1
|
Sahu M, Xiao Y, Porras JL, Amanian A, Jain A, Thamboo A, Taylor RH, Creighton FX, Ishii M. A Label-Efficient Framework for Automated Sinonasal CT Segmentation in Image-Guided Surgery. Otolaryngol Head Neck Surg 2024. [PMID: 38922721 DOI: 10.1002/ohn.868] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/12/2023] [Revised: 04/20/2024] [Accepted: 06/03/2024] [Indexed: 06/28/2024]
Abstract
OBJECTIVE Segmentation, the partitioning of patient imaging into multiple, labeled segments, has several potential clinical benefits but when performed manually is tedious and resource intensive. Automated deep learning (DL)-based segmentation methods can streamline the process. The objective of this study was to evaluate a label-efficient DL pipeline that requires only a small number of annotated scans for semantic segmentation of sinonasal structures in CT scans. STUDY DESIGN Retrospective cohort study. SETTING Academic institution. METHODS Forty CT scans were used in this study including 16 scans in which the nasal septum (NS), inferior turbinate (IT), maxillary sinus (MS), and optic nerve (ON) were manually annotated using an open-source software. A label-efficient DL framework was used to train jointly on a few manually labeled scans and the remaining unlabeled scans. Quantitative analysis was then performed to obtain the number of annotated scans needed to achieve submillimeter average surface distances (ASDs). RESULTS Our findings reveal that merely four labeled scans are necessary to achieve median submillimeter ASDs for large sinonasal structures-NS (0.96 mm), IT (0.74 mm), and MS (0.43 mm), whereas eight scans are required for smaller structures-ON (0.80 mm). CONCLUSION We have evaluated a label-efficient pipeline for segmentation of sinonasal structures. Empirical results demonstrate that automated DL methods can achieve submillimeter accuracy using a small number of labeled CT scans. Our pipeline has the potential to improve pre-operative planning workflows, robotic- and image-guidance navigation systems, computer-assisted diagnosis, and the construction of statistical shape models to quantify population variations. LEVEL OF EVIDENCE N/A.
Collapse
Affiliation(s)
- Manish Sahu
- Laboratory for Computational Sensing and Robotics, Johns Hopkins University, Baltimore, Maryland, USA
| | - Yuliang Xiao
- Laboratory for Computational Sensing and Robotics, Johns Hopkins University, Baltimore, Maryland, USA
| | - Jose L Porras
- Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
| | - Ameen Amanian
- Division of Otolaryngology, Department of Surgery, University of British Columbia, Vancouver, British Columbia, Canada
| | - Aseem Jain
- Laboratory for Computational Sensing and Robotics, Johns Hopkins University, Baltimore, Maryland, USA
| | - Andrew Thamboo
- Division of Otolaryngology, Department of Surgery, University of British Columbia, Vancouver, British Columbia, Canada
| | - Russell H Taylor
- Laboratory for Computational Sensing and Robotics, Johns Hopkins University, Baltimore, Maryland, USA
| | - Francis X Creighton
- Department of Otolaryngology, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
| | - Masaru Ishii
- Department of Otolaryngology, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
| |
Collapse
|
2
|
Court LE, Aggarwal A, Jhingran A, Naidoo K, Netherton T, Olanrewaju A, Peterson C, Parkes J, Simonds H, Trauernicht C, Zhang L, Beadle BM. Artificial Intelligence-Based Radiotherapy Contouring and Planning to Improve Global Access to Cancer Care. JCO Glob Oncol 2024; 10:e2300376. [PMID: 38484191 PMCID: PMC10954080 DOI: 10.1200/go.23.00376] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/03/2023] [Revised: 11/22/2023] [Accepted: 01/22/2024] [Indexed: 03/19/2024] Open
Abstract
PURPOSE Increased automation has been identified as one approach to improving global cancer care. The Radiation Planning Assistant (RPA) is a web-based tool offering automated radiotherapy (RT) contouring and planning to low-resource clinics. In this study, the RPA workflow and clinical acceptability were assessed by physicians around the world. METHODS The RPA output for 75 cases was reviewed by at least three physicians; 31 radiation oncologists at 16 institutions in six countries on five continents reviewed RPA contours and plans for clinical acceptability using a 5-point Likert scale. RESULTS For cervical cancer, RPA plans using bony landmarks were scored as usable as-is in 81% (with minor edits 93%); using soft tissue contours, plans were scored as usable as-is in 79% (with minor edits 96%). For postmastectomy breast cancer, RPA plans were scored as usable as-is in 44% (with minor edits 91%). For whole-brain treatment, RPA plans were scored as usable as-is in 67% (with minor edits 99%). For head/neck cancer, the normal tissue autocontours were acceptable as-is in 89% (with minor edits 97%). The clinical target volumes (CTVs) were acceptable as-is in 40% (with minor edits 93%). The volumetric-modulated arc therapy (VMAT) plans were acceptable as-is in 87% (with minor edits 96%). For cervical cancer, the normal tissue autocontours were acceptable as-is in 92% (with minor edits 99%). The CTVs for cervical cancer were scored as acceptable as-is in 83% (with minor edits 92%). The VMAT plans for cervical cancer were acceptable as-is in 99% (with minor edits 100%). CONCLUSION The RPA, a web-based tool designed to improve access to high-quality RT in low-resource settings, has high rates of clinical acceptability by practicing clinicians around the world. It has significant potential for successful implementation in low-resource clinics.
Collapse
Affiliation(s)
| | - Ajay Aggarwal
- Guy's and St Thomas Hospitals, London, United Kingdom
| | - Anuja Jhingran
- University of Texas MD Anderson Cancer Center, Houston, TX
| | | | | | | | | | | | | | | | - Lifei Zhang
- University of Texas MD Anderson Cancer Center, Houston, TX
| | | |
Collapse
|
3
|
Boussioux L, Ma Y, Thomas NK, Bertsimas D, Shusharina N, Pursley J, Chen YL, DeLaney TF, Qian J, Bortfeld T. Automated Segmentation of Sacral Chordoma and Surrounding Muscles Using Deep Learning Ensemble. Int J Radiat Oncol Biol Phys 2023; 117:738-749. [PMID: 37451472 PMCID: PMC10665084 DOI: 10.1016/j.ijrobp.2023.03.078] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/16/2022] [Revised: 03/18/2023] [Accepted: 03/30/2023] [Indexed: 07/18/2023]
Abstract
PURPOSE The manual segmentation of organ structures in radiation oncology treatment planning is a time-consuming and highly skilled task, particularly when treating rare tumors like sacral chordomas. This study evaluates the performance of automated deep learning (DL) models in accurately segmenting the gross tumor volume (GTV) and surrounding muscle structures of sacral chordomas. METHODS AND MATERIALS An expert radiation oncologist contoured 5 muscle structures (gluteus maximus, gluteus medius, gluteus minimus, paraspinal, piriformis) and sacral chordoma GTV on computed tomography images from 48 patients. We trained 6 DL auto-segmentation models based on 3-dimensional U-Net and residual 3-dimensional U-Net architectures. We then implemented an average and an optimally weighted average ensemble to improve prediction performance. We evaluated algorithms with the average and standard deviation of the volumetric Dice similarity coefficient, surface Dice similarity coefficient with 2- and 3-mm thresholds, and average symmetric surface distance. One independent expert radiation oncologist assessed the clinical viability of the DL contours and determined the necessary amount of editing before they could be used in clinical practice. RESULTS Quantitatively, the ensembles performed the best across all structures. The optimal ensemble (volumetric Dice similarity coefficient, average symmetric surface distance) was (85.5 ± 6.4, 2.6 ± 0.8; GTV), (94.4 ± 1.5, 1.0 ± 0.4; gluteus maximus), (92.6 ± 0.9, 0.9 ± 0.1; gluteus medius), (85.0 ± 2.7, 1.1 ± 0.3; gluteus minimus), (92.1 ± 1.5, 0.8 ± 0.2; paraspinal), and (78.3 ± 5.7, 1.5 ± 0.6; piriformis). The qualitative evaluation suggested that the best model could reduce the total muscle and tumor delineation time to a 19-minute average. CONCLUSIONS Our methodology produces expert-level muscle and sacral chordoma tumor segmentation using DL and ensemble modeling. It can substantially augment the streamlining and accuracy of treatment planning and represents a critical step toward automated delineation of the clinical target volume in sarcoma and other disease sites.
Collapse
Affiliation(s)
- Leonard Boussioux
- Operations Research Center, Massachusetts Institute of Technology, Cambridge, Massachusetts; Sloan School of Management, Massachusetts Institute of Technology, Cambridge, Massachusetts; University of Washington, Michael G. Foster School of Business, Department of Information Systems and Operations Management, Seattle, Washington.
| | - Yu Ma
- Operations Research Center, Massachusetts Institute of Technology, Cambridge, Massachusetts; Sloan School of Management, Massachusetts Institute of Technology, Cambridge, Massachusetts
| | - Nancy Knight Thomas
- Operations Research Center, Massachusetts Institute of Technology, Cambridge, Massachusetts; Sloan School of Management, Massachusetts Institute of Technology, Cambridge, Massachusetts
| | - Dimitris Bertsimas
- Operations Research Center, Massachusetts Institute of Technology, Cambridge, Massachusetts; Sloan School of Management, Massachusetts Institute of Technology, Cambridge, Massachusetts
| | - Nadya Shusharina
- Department of Radiation Oncology, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts
| | - Jennifer Pursley
- Department of Radiation Oncology, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts
| | - Yen-Lin Chen
- Department of Radiation Oncology, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts
| | - Thomas F DeLaney
- Department of Radiation Oncology, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts
| | - Jack Qian
- Department of Radiation Oncology, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts
| | - Thomas Bortfeld
- Department of Radiation Oncology, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts
| |
Collapse
|
4
|
Doolan PJ, Charalambous S, Roussakis Y, Leczynski A, Peratikou M, Benjamin M, Ferentinos K, Strouthos I, Zamboglou C, Karagiannis E. A clinical evaluation of the performance of five commercial artificial intelligence contouring systems for radiotherapy. Front Oncol 2023; 13:1213068. [PMID: 37601695 PMCID: PMC10436522 DOI: 10.3389/fonc.2023.1213068] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/27/2023] [Accepted: 07/17/2023] [Indexed: 08/22/2023] Open
Abstract
Purpose/objectives Auto-segmentation with artificial intelligence (AI) offers an opportunity to reduce inter- and intra-observer variability in contouring, to improve the quality of contours, as well as to reduce the time taken to conduct this manual task. In this work we benchmark the AI auto-segmentation contours produced by five commercial vendors against a common dataset. Methods and materials The organ at risk (OAR) contours generated by five commercial AI auto-segmentation solutions (Mirada (Mir), MVision (MV), Radformation (Rad), RayStation (Ray) and TheraPanacea (Ther)) were compared to manually-drawn expert contours from 20 breast, 20 head and neck, 20 lung and 20 prostate patients. Comparisons were made using geometric similarity metrics including volumetric and surface Dice similarity coefficient (vDSC and sDSC), Hausdorff distance (HD) and Added Path Length (APL). To assess the time saved, the time taken to manually draw the expert contours, as well as the time to correct the AI contours, were recorded. Results There are differences in the number of CT contours offered by each AI auto-segmentation solution at the time of the study (Mir 99; MV 143; Rad 83; Ray 67; Ther 86), with all offering contours of some lymph node levels as well as OARs. Averaged across all structures, the median vDSCs were good for all systems and compared favorably with existing literature: Mir 0.82; MV 0.88; Rad 0.86; Ray 0.87; Ther 0.88. All systems offer substantial time savings, ranging between: breast 14-20 mins; head and neck 74-93 mins; lung 20-26 mins; prostate 35-42 mins. The time saved, averaged across all structures, was similar for all systems: Mir 39.8 mins; MV 43.6 mins; Rad 36.6 min; Ray 43.2 mins; Ther 45.2 mins. Conclusions All five commercial AI auto-segmentation solutions evaluated in this work offer high quality contours in significantly reduced time compared to manual contouring, and could be used to render the radiotherapy workflow more efficient and standardized.
Collapse
Affiliation(s)
- Paul J. Doolan
- Department of Medical Physics, German Oncology Center, Limassol, Cyprus
| | | | - Yiannis Roussakis
- Department of Medical Physics, German Oncology Center, Limassol, Cyprus
| | - Agnes Leczynski
- Department of Radiation Oncology, German Oncology Center, Limassol, Cyprus
| | - Mary Peratikou
- Department of Radiation Oncology, German Oncology Center, Limassol, Cyprus
| | - Melka Benjamin
- Department of Radiation Oncology, German Oncology Center, Limassol, Cyprus
| | - Konstantinos Ferentinos
- Department of Radiation Oncology, German Oncology Center, Limassol, Cyprus
- School of Medicine, European University Cyprus, Nicosia, Cyprus
| | - Iosif Strouthos
- Department of Radiation Oncology, German Oncology Center, Limassol, Cyprus
- School of Medicine, European University Cyprus, Nicosia, Cyprus
| | - Constantinos Zamboglou
- Department of Radiation Oncology, German Oncology Center, Limassol, Cyprus
- School of Medicine, European University Cyprus, Nicosia, Cyprus
- Department of Radiation Oncology, Medical Center – University of Freiberg, Freiberg, Germany
| | - Efstratios Karagiannis
- Department of Radiation Oncology, German Oncology Center, Limassol, Cyprus
- School of Medicine, European University Cyprus, Nicosia, Cyprus
| |
Collapse
|
5
|
Baroudi H, Brock KK, Cao W, Chen X, Chung C, Court LE, El Basha MD, Farhat M, Gay S, Gronberg MP, Gupta AC, Hernandez S, Huang K, Jaffray DA, Lim R, Marquez B, Nealon K, Netherton TJ, Nguyen CM, Reber B, Rhee DJ, Salazar RM, Shanker MD, Sjogreen C, Woodland M, Yang J, Yu C, Zhao Y. Automated Contouring and Planning in Radiation Therapy: What Is 'Clinically Acceptable'? Diagnostics (Basel) 2023; 13:diagnostics13040667. [PMID: 36832155 PMCID: PMC9955359 DOI: 10.3390/diagnostics13040667] [Citation(s) in RCA: 19] [Impact Index Per Article: 19.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/18/2022] [Revised: 01/21/2023] [Accepted: 01/30/2023] [Indexed: 02/12/2023] Open
Abstract
Developers and users of artificial-intelligence-based tools for automatic contouring and treatment planning in radiotherapy are expected to assess clinical acceptability of these tools. However, what is 'clinical acceptability'? Quantitative and qualitative approaches have been used to assess this ill-defined concept, all of which have advantages and disadvantages or limitations. The approach chosen may depend on the goal of the study as well as on available resources. In this paper, we discuss various aspects of 'clinical acceptability' and how they can move us toward a standard for defining clinical acceptability of new autocontouring and planning tools.
Collapse
Affiliation(s)
- Hana Baroudi
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Kristy K. Brock
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- Department of Imaging Physics, Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Wenhua Cao
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Xinru Chen
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Caroline Chung
- Department of Radiation Oncology, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Laurence E. Court
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- Correspondence:
| | - Mohammad D. El Basha
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Maguy Farhat
- Department of Radiation Oncology, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Skylar Gay
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Mary P. Gronberg
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Aashish Chandra Gupta
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
- Department of Imaging Physics, Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Soleil Hernandez
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Kai Huang
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - David A. Jaffray
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- Department of Imaging Physics, Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Rebecca Lim
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Barbara Marquez
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Kelly Nealon
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Tucker J. Netherton
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Callistus M. Nguyen
- Department of Imaging Physics, Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Brandon Reber
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
- Department of Imaging Physics, Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Dong Joo Rhee
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Ramon M. Salazar
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Mihir D. Shanker
- The University of Queensland, Saint Lucia 4072, Australia
- The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Carlos Sjogreen
- Department of Physics, University of Houston, Houston, TX 77004, USA
| | - McKell Woodland
- Department of Imaging Physics, Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- Department of Computer Science, Rice University, Houston, TX 77005, USA
| | - Jinzhong Yang
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
| | - Cenji Yu
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| | - Yao Zhao
- Department of Radiation Physics, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
- The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences, Houston, TX 77030, USA
| |
Collapse
|
6
|
Jin D, Guo D, Ge J, Ye X, Lu L. Towards automated organs at risk and target volumes contouring: Defining precision radiation therapy in the modern era. JOURNAL OF THE NATIONAL CANCER CENTER 2022; 2:306-313. [PMID: 39036546 PMCID: PMC11256697 DOI: 10.1016/j.jncc.2022.09.003] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2022] [Revised: 09/06/2022] [Accepted: 09/27/2022] [Indexed: 12/05/2022] Open
Abstract
Precision radiotherapy is a critical and indispensable cancer treatment means in the modern clinical workflow with the goal of achieving "quality-up and cost-down" in patient care. The challenge of this therapy lies in developing computerized clinical-assistant solutions with precision, automation, and reproducibility built-in to deliver it at scale. In this work, we provide a comprehensive yet ongoing, incomplete survey of and discussions on the recent progress of utilizing advanced deep learning, semantic organ parsing, multimodal imaging fusion, neural architecture search and medical image analytical techniques to address four corner-stone problems or sub-problems required by all precision radiotherapy workflows, namely, organs at risk (OARs) segmentation, gross tumor volume (GTV) segmentation, metastasized lymph node (LN) detection, and clinical tumor volume (CTV) segmentation. Without loss of generality, we mainly focus on using esophageal and head-and-neck cancers as examples, but the methods can be extrapolated to other types of cancers. High-precision, automated and highly reproducible OAR/GTV/LN/CTV auto-delineation techniques have demonstrated their effectiveness in reducing the inter-practitioner variabilities and the time cost to permit rapid treatment planning and adaptive replanning for the benefit of patients. Through the presentation of the achievements and limitations of these techniques in this review, we hope to encourage more collective multidisciplinary precision radiotherapy workflows to transpire.
Collapse
Affiliation(s)
- Dakai Jin
- DAMO Academy, Alibaba Group, New York, United States
| | - Dazhou Guo
- DAMO Academy, Alibaba Group, New York, United States
| | - Jia Ge
- Department of Radiation Oncology, The First Affiliated Hospital of Zhejiang University, Hangzhou, China
| | - Xianghua Ye
- Department of Radiation Oncology, The First Affiliated Hospital of Zhejiang University, Hangzhou, China
| | - Le Lu
- DAMO Academy, Alibaba Group, New York, United States
| |
Collapse
|
7
|
Kihara S, Koike Y, Takegawa H, Anetai Y, Nakamura S, Tanigawa N, Koizumi M. Clinical target volume segmentation based on gross tumor volume using deep learning for head and neck cancer treatment. Med Dosim 2022; 48:20-24. [PMID: 36273950 DOI: 10.1016/j.meddos.2022.09.004] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/14/2021] [Revised: 02/07/2022] [Accepted: 09/17/2022] [Indexed: 02/04/2023]
Abstract
Accurate clinical target volume (CTV) delineation is important for head and neck intensity-modulated radiation therapy. However, delineation is time-consuming and susceptible to interobserver variability (IOV). Based on a manual contouring process commonly used in clinical practice, we developed a deep learning (DL)-based method to delineate a low-risk CTV with computed tomography (CT) and gross tumor volume (GTV) input and compared it with a CT-only input. A total of 310 patients with oropharynx cancer were randomly divided into the training set (250) and test set (60). The low-risk CTV and primary GTV contours were used to generate label data for the input and ground truth. A 3D U-Net with a two-channel input of CT and GTV (U-NetGTV) was proposed and its performance was compared with a U-Net with only CT input (U-NetCT). The Dice similarity coefficient (DSC) and average Hausdorff distance (AHD) were evaluated. The time required to predict the CTV was 0.86 s per patient. U-NetGTV showed a significantly higher mean DSC value than U-NetCT (0.80 ± 0.03 and 0.76 ± 0.05) and a significantly lower mean AHD value (3.0 ± 0.5 mm vs 3.5 ± 0.7 mm). Compared to the existing DL method with only CT input, the proposed GTV-based segmentation using DL showed a more precise low-risk CTV segmentation for head and neck cancer. Our findings suggest that the proposed method could reduce the contouring time of a low-risk CTV, allowing the standardization of target delineations for head and neck cancer.
Collapse
Affiliation(s)
- Sayaka Kihara
- Department of Medical Physics and Engineering, Osaka University Graduate School of Medicine, 1-7 Yamadaoka, Suita, Osaka, 565-0871, Japan
| | - Yuhei Koike
- Department of Radiology, Kansai Medical University, 2-5-1 Shinmachi, Hirakata, Osaka, 573-1010, Japan.
| | - Hideki Takegawa
- Department of Radiology, Kansai Medical University, 2-5-1 Shinmachi, Hirakata, Osaka, 573-1010, Japan
| | - Yusuke Anetai
- Department of Radiology, Kansai Medical University, 2-5-1 Shinmachi, Hirakata, Osaka, 573-1010, Japan
| | - Satoaki Nakamura
- Department of Radiology, Kansai Medical University, 2-5-1 Shinmachi, Hirakata, Osaka, 573-1010, Japan
| | - Noboru Tanigawa
- Department of Radiology, Kansai Medical University, 2-5-1 Shinmachi, Hirakata, Osaka, 573-1010, Japan
| | - Masahiko Koizumi
- Department of Medical Physics and Engineering, Osaka University Graduate School of Medicine, 1-7 Yamadaoka, Suita, Osaka, 565-0871, Japan
| |
Collapse
|
8
|
Wang J, Chen Z, Yang C, Qu B, Ma L, Fan W, Zhou Q, Zheng Q, Xu S. Evaluation Exploration of Atlas-Based and Deep Learning-Based Automatic Contouring for Nasopharyngeal Carcinoma. Front Oncol 2022; 12:833816. [PMID: 35433460 PMCID: PMC9008357 DOI: 10.3389/fonc.2022.833816] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/12/2021] [Accepted: 02/25/2022] [Indexed: 11/13/2022] Open
Abstract
Purpose The purpose of this study was to evaluate and explore the difference between an atlas-based and deep learning (DL)-based auto-segmentation scheme for organs at risk (OARs) of nasopharyngeal carcinoma cases to provide valuable help for clinical practice. Methods 120 nasopharyngeal carcinoma cases were established in the MIM Maestro (atlas) database and trained by a DL-based model (AccuContour®), and another 20 nasopharyngeal carcinoma cases were randomly selected outside the atlas database. The experienced physicians contoured 14 OARs from 20 patients based on the published consensus guidelines, and these were defined as the reference volumes (Vref). Meanwhile, these OARs were auto-contoured using an atlas-based model, a pre-built DL-based model, and an on-site trained DL-based model. These volumes were named Vatlas, VDL-pre-built, and VDL-trained, respectively. The similarities between Vatlas, VDL-pre-built, VDL-trained, and Vref were assessed using the Dice similarity coefficient (DSC), Jaccard coefficient (JAC), maximum Hausdorff distance (HDmax), and deviation of centroid (DC) methods. A one-way ANOVA test was carried out to show the differences (between each two of them). Results The results of the three methods were almost similar for the brainstem and eyes. For inner ears and temporomandibular joints, the results of the pre-built DL-based model are the worst, as well as the results of atlas-based auto-segmentation for the lens. For the segmentation of optic nerves, the trained DL-based model shows the best performance (p < 0.05). For the contouring of the oral cavity, the DSC value of VDL-pre-built is the smallest, and VDL-trained is the most significant (p < 0.05). For the parotid glands, the DSC of Vatlas is the minimum (about 0.80 or so), and VDL-pre-built and VDL-trained are slightly larger (about 0.82 or so). In addition to the oral cavity, parotid glands, and the brainstem, the maximum Hausdorff distances of the other organs are below 0.5 cm using the trained DL-based segmentation model. The trained DL-based segmentation method behaves well in the contouring of all the organs that the maximum average deviation of the centroid is no more than 0.3 cm. Conclusion The trained DL-based segmentation performs significantly better than atlas-based segmentation for nasopharyngeal carcinoma, especially for the OARs with small volumes. Although some delineation results still need further modification, auto-segmentation methods improve the work efficiency and provide a level of help for clinical work.
Collapse
Affiliation(s)
- Jinyuan Wang
- Department of Radiation Oncology, The First Medical Center of the Chinese PLA General Hospital, Beijing, China
| | | | | | - Baolin Qu
- Department of Radiation Oncology, The First Medical Center of the Chinese PLA General Hospital, Beijing, China
| | - Lin Ma
- Department of Radiation Oncology, The First Medical Center of the Chinese PLA General Hospital, Beijing, China
| | - Wenjun Fan
- Department of Radiation Oncology, The First Medical Center of the Chinese PLA General Hospital, Beijing, China
| | - Qichao Zhou
- Manteia Technologies Co., Ltd., Xiamen, China
| | - Qingzeng Zheng
- Department of Radiation Oncology, Beijing Geriatric Hospital, Beijing, China
| | - Shouping Xu
- Department of Radiation Oncology, The First Medical Center of the Chinese PLA General Hospital, Beijing, China
| |
Collapse
|
9
|
Harrison K, Pullen H, Welsh C, Oktay O, Alvarez-Valle J, Jena R. Machine Learning for Auto-Segmentation in Radiotherapy Planning. Clin Oncol (R Coll Radiol) 2022; 34:74-88. [PMID: 34996682 DOI: 10.1016/j.clon.2021.12.003] [Citation(s) in RCA: 37] [Impact Index Per Article: 18.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/29/2021] [Revised: 11/27/2021] [Accepted: 12/03/2021] [Indexed: 12/12/2022]
Abstract
Manual segmentation of target structures and organs at risk is a crucial step in the radiotherapy workflow. It has the disadvantages that it can require several hours of clinician time per patient and is prone to inter- and intra-observer variability. Automatic segmentation (auto-segmentation), using computer algorithms, seeks to address these issues. Advances in machine learning and computer vision have led to the development of methods for accurate and efficient auto-segmentation. This review surveys auto-segmentation techniques and applications in radiotherapy planning. It provides an overview of traditional approaches to auto-segmentation, including intensity analysis, shape modelling and atlas-based methods. The focus, though, is on uses of machine learning and deep learning, including convolutional neural networks. Finally, the future of machine-learning-driven auto-segmentation in clinical settings is considered, and the barriers that must be overcome for it to be widely accepted into routine practice are highlighted.
Collapse
Affiliation(s)
- K Harrison
- Cavendish Laboratory, University of Cambridge, Cambridge, UK.
| | - H Pullen
- Cavendish Laboratory, University of Cambridge, Cambridge, UK
| | - C Welsh
- Department of Oncology, University of Cambridge, Cambridge, UK
| | - O Oktay
- Health Intelligence, Microsoft Research, Cambridge, UK
| | | | - R Jena
- Department of Oncology, University of Cambridge, Cambridge, UK; Department of Oncology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK
| |
Collapse
|
10
|
Bandeira Diniz JO, Ferreira JL, Bandeira Diniz PH, Silva AC, Paiva AC. A deep learning method with residual blocks for automatic spinal cord segmentation in planning CT. Biomed Signal Process Control 2022. [DOI: 10.1016/j.bspc.2021.103074] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
|
11
|
Talwar V, Chufal KS, Joga S. Artificial Intelligence: A New Tool in Oncologist's Armamentarium. Indian J Med Paediatr Oncol 2021. [DOI: 10.1055/s-0041-1735577] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/19/2022] Open
Abstract
AbstractArtificial intelligence (AI) has become an essential tool in human life because of its pivotal role in communications, transportation, media, and social networking. Inspired by the complex neuronal network and its functions in human beings, AI, using computer-based algorithms and training, had been explored since the 1950s. To tackle the enormous amount of patients' clinical data, imaging, histopathological data, and the increasing pace of research on new treatments and clinical trials, and ever-changing guidelines for treatment with the advent of novel drugs and evidence, AI is the need of the hour. There are numerous publications and active work on AI's role in the field of oncology. In this review, we discuss the fundamental terminology of AI, its applications in oncology on the whole, and its limitations. There is an inter-relationship between AI, machine learning and, deep learning. The virtual branch of AI deals with machine learning. While the physical branch of AI deals with the delivery of different forms of treatment—surgery, targeted drug delivery, and elderly care. The applications of AI in oncology include cancer screening, diagnosis (clinical, imaging, and histopathological), radiation therapy (image acquisition, tumor and organs at risk segmentation, image registration, planning, and delivery), prediction of treatment outcomes and toxicities, prediction of cancer cell sensitivity to therapeutics and clinical decision-making. A specific area of interest is in the development of effective drug combinations tailored to every patient and tumor with the help of AI. Radiomics, the new kid on the block, deals with the planning and administration of radiotherapy. As with any new invention, AI has its fallacies. The limitations include lack of external validation and proof of generalizability, difficulty in data access for rare diseases, ethical and legal issues, no precise logic behind the prediction, and last but not the least, lack of education and expertise among medical professionals. A collaboration between departments of clinical oncology, bioinformatics, and data sciences can help overcome these problems in the near future.
Collapse
Affiliation(s)
- Vineet Talwar
- Department of Medical Oncology, Rajiv Gandhi Cancer Institute & Research Centre, New Delhi, India
| | - Kundan Singh Chufal
- Department of Radiation Oncology, Rajiv Gandhi Cancer Institute & Research Centre, New Delhi, India
| | - Srujana Joga
- Department of Medical Oncology, Rajiv Gandhi Cancer Institute & Research Centre, New Delhi, India
| |
Collapse
|
12
|
Korte JC, Hardcastle N, Ng SP, Clark B, Kron T, Jackson P. Cascaded deep learning-based auto-segmentation for head and neck cancer patients: Organs at risk on T2-weighted magnetic resonance imaging. Med Phys 2021; 48:7757-7772. [PMID: 34676555 DOI: 10.1002/mp.15290] [Citation(s) in RCA: 10] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/16/2021] [Revised: 08/30/2021] [Accepted: 09/24/2021] [Indexed: 12/09/2022] Open
Abstract
PURPOSE To investigate multiple deep learning methods for automated segmentation (auto-segmentation) of the parotid glands, submandibular glands, and level II and level III lymph nodes on magnetic resonance imaging (MRI). Outlining radiosensitive organs on images used to assist radiation therapy (radiotherapy) of patients with head and neck cancer (HNC) is a time-consuming task, in which variability between observers may directly impact on patient treatment outcomes. Auto-segmentation on computed tomography imaging has been shown to result in significant time reductions and more consistent outlines of the organs at risk. METHODS Three convolutional neural network (CNN)-based auto-segmentation architectures were developed using manual segmentations and T2-weighted MRI images provided from the American Association of Physicists in Medicine (AAPM) radiotherapy MRI auto-contouring (RT-MAC) challenge dataset (n = 31). Auto-segmentation performance was evaluated with segmentation similarity and surface distance metrics on the RT-MAC dataset with institutional manual segmentations (n = 10). The generalizability of the auto-segmentation methods was assessed on an institutional MRI dataset (n = 10). RESULTS Auto-segmentation performance on the RT-MAC images with institutional segmentations was higher than previously reported MRI methods for the parotid glands (Dice: 0.860 ± 0.067, mean surface distance [MSD]: 1.33 ± 0.40 mm) and the first report of MRI performance for submandibular glands (Dice: 0.830 ± 0.032, MSD: 1.16 ± 0.47 mm). We demonstrate that high-resolution auto-segmentations with improved geometric accuracy can be generated for the parotid and submandibular glands by cascading a localizer CNN and a cropped high-resolution CNN. Improved MSDs were observed between automatic and manual segmentations of the submandibular glands when a low-resolution auto-segmentation was used as prior knowledge in the second-stage CNN. Reduced auto-segmentation performance was observed on our institutional MRI dataset when trained on external RT-MAC images; only the parotid gland auto-segmentations were considered clinically feasible for manual correction (Dice: 0.775 ± 0.105, MSD: 1.20 ± 0.60 mm). CONCLUSIONS This work demonstrates that CNNs are a suitable method to auto-segment the parotid and submandibular glands on MRI images of patients with HNC, and that cascaded CNNs can generate high-resolution segmentations with improved geometric accuracy. Deep learning methods may be suitable for auto-segmentation of the parotid glands on T2-weighted MRI images from different scanners, but further work is required to improve the performance and generalizability of these methods for auto-segmentation of the submandibular glands and lymph nodes.
Collapse
Affiliation(s)
- James C Korte
- Department of Physical Science, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia.,Department of Biomedical Engineering, University of Melbourne, Melbourne, Victoria, Australia
| | - Nicholas Hardcastle
- Department of Physical Science, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia.,Centre for Medical Radiation Physics, University of Wollongong, Wollongong, New South Wales, Australia.,Sir Peter MacCallum Department of Oncology, University of Melbourne, Melbourne, Victoria, Australia
| | - Sweet Ping Ng
- Department of Radiation Oncology, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia.,Department of Radiation Oncology, Olivia Newton-John Cancer and Wellness Centre, Austin Health, Melbourne, Victoria, Australia
| | - Brett Clark
- Department of Physical Science, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia.,Department of Biomedical Engineering, University of Melbourne, Melbourne, Victoria, Australia
| | - Tomas Kron
- Department of Physical Science, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia.,Sir Peter MacCallum Department of Oncology, University of Melbourne, Melbourne, Victoria, Australia
| | - Price Jackson
- Department of Physical Science, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia.,Sir Peter MacCallum Department of Oncology, University of Melbourne, Melbourne, Victoria, Australia
| |
Collapse
|
13
|
He Y, Zhang S, Luo Y, Yu H, Fu Y, Wu Z, Jiang X, Li P. Quantitative Comparisons of Deep-learning-based and Atlas-based Auto-segmentation of the Intermediate Risk Clinical Target Volume for Nasopharyngeal Carcinoma. Curr Med Imaging 2021; 18:335-345. [PMID: 34455965 DOI: 10.2174/1573405617666210827165031] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/05/2021] [Revised: 05/22/2021] [Accepted: 06/02/2021] [Indexed: 02/08/2023]
Abstract
BACKGROUND Manual segment target volumes were time-consuming and inter-observer variability couldn't be avoided. With the development of computer science, auto-segmentation had the potential to solve this problem. OBJECTIVE To evaluate the accuracy and stability of Atlas-based and deep-learning-based auto-segmentation of the intermediate risk clinical target volume, composed of CTV2 and CTVnd, for nasopharyngeal carcinoma quantitatively. METHODS AND MATERIALS A cascade-deep-residual neural network was constructed to automatically segment CTV2 and CTVnd by deep learning method. Meanwhile, a commercially available software was used to automatically segment the same regions by Atlas-based method. The datasets included contrast computed tomography scans from 102 patients. For each patient, the two regions were manually delineated by one experienced physician. The similarity between the two auto-segmentation methods was quantitatively evaluated by Dice similarity coefficient, the 95th Hausdorff distance, volume overlap error and relative volume difference, respectively. Statistical analyses were performed using the ranked Wilcoxon test. RESULTS The average Dice similarity coefficient (±standard deviation) given by the deep-learning-based and Atlas-based auto-segmentation were 0.84(±0.03) and 0.74(±0.04) for CTV2, 0.79(±0.02) and 0.68(±0.03) for CTVnd, respectively. For the 95th Hausdorff distance, the corresponding values were 6.30±3.55mm and 9.34±3.39mm for CTV2, 7.09±2.27mm and 14.33±3.98mm for CTVnd. Besides, volume overlap error and relative volume difference could also predict the same situations. Statistical analyses showed significant difference between the two auto-segmentation methods (p<0.01). CONCLUSIONS Compared with the Atlas-based segmentation approach, the deep-learning-based segmentation method performed better both in accuracy and stability for meaningful anatomical areas other than organs at risk.
Collapse
Affiliation(s)
- Yisong He
- Department of Radiation Oncology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province. China
| | - Shengyuan Zhang
- Key Laboratory of Radiation Physics and Technology of Ministry of Education, Institute of Nuclear Science and Technology, Sichuan University, Chengdu 610064, Sichuan Province. China
| | - Yong Luo
- Department of Radiation Oncology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province. China
| | - Hang Yu
- Department of Radiation Oncology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province. China
| | - Yuchuan Fu
- Department of Radiation Oncology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province. China
| | - Zhangwen Wu
- Key Laboratory of Radiation Physics and Technology of Ministry of Education, Institute of Nuclear Science and Technology, Sichuan University, Chengdu 610064, Sichuan Province. China
| | - Xiaoxuan Jiang
- Department of Radiation Oncology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province. China
| | - Ping Li
- Department of Radiation Oncology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province. China
| |
Collapse
|
14
|
Nikolov S, Blackwell S, Zverovitch A, Mendes R, Livne M, De Fauw J, Patel Y, Meyer C, Askham H, Romera-Paredes B, Kelly C, Karthikesalingam A, Chu C, Carnell D, Boon C, D'Souza D, Moinuddin SA, Garie B, McQuinlan Y, Ireland S, Hampton K, Fuller K, Montgomery H, Rees G, Suleyman M, Back T, Hughes CO, Ledsam JR, Ronneberger O. Clinically Applicable Segmentation of Head and Neck Anatomy for Radiotherapy: Deep Learning Algorithm Development and Validation Study. J Med Internet Res 2021; 23:e26151. [PMID: 34255661 PMCID: PMC8314151 DOI: 10.2196/26151] [Citation(s) in RCA: 118] [Impact Index Per Article: 39.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/30/2020] [Revised: 02/10/2021] [Accepted: 04/30/2021] [Indexed: 12/16/2022] Open
Abstract
BACKGROUND Over half a million individuals are diagnosed with head and neck cancer each year globally. Radiotherapy is an important curative treatment for this disease, but it requires manual time to delineate radiosensitive organs at risk. This planning process can delay treatment while also introducing interoperator variability, resulting in downstream radiation dose differences. Although auto-segmentation algorithms offer a potentially time-saving solution, the challenges in defining, quantifying, and achieving expert performance remain. OBJECTIVE Adopting a deep learning approach, we aim to demonstrate a 3D U-Net architecture that achieves expert-level performance in delineating 21 distinct head and neck organs at risk commonly segmented in clinical practice. METHODS The model was trained on a data set of 663 deidentified computed tomography scans acquired in routine clinical practice and with both segmentations taken from clinical practice and segmentations created by experienced radiographers as part of this research, all in accordance with consensus organ at risk definitions. RESULTS We demonstrated the model's clinical applicability by assessing its performance on a test set of 21 computed tomography scans from clinical practice, each with 21 organs at risk segmented by 2 independent experts. We also introduced surface Dice similarity coefficient, a new metric for the comparison of organ delineation, to quantify the deviation between organ at risk surface contours rather than volumes, better reflecting the clinical task of correcting errors in automated organ segmentations. The model's generalizability was then demonstrated on 2 distinct open-source data sets, reflecting different centers and countries to model training. CONCLUSIONS Deep learning is an effective and clinically applicable technique for the segmentation of the head and neck anatomy for radiotherapy. With appropriate validation studies and regulatory approvals, this system could improve the efficiency, consistency, and safety of radiotherapy pathways.
Collapse
Affiliation(s)
| | | | | | - Ruheena Mendes
- University College London Hospitals NHS Foundation Trust, London, United Kingdom
| | | | | | | | | | | | | | | | | | | | - Dawn Carnell
- University College London Hospitals NHS Foundation Trust, London, United Kingdom
| | - Cheng Boon
- Clatterbridge Cancer Centre NHS Foundation Trust, Liverpool, United Kingdom
| | - Derek D'Souza
- University College London Hospitals NHS Foundation Trust, London, United Kingdom
| | - Syed Ali Moinuddin
- University College London Hospitals NHS Foundation Trust, London, United Kingdom
| | | | | | | | | | | | | | - Geraint Rees
- University College London, London, United Kingdom
| | | | | | | | | | | |
Collapse
|
15
|
Qiu B, Guo J, Kraeima J, Glas HH, Zhang W, Borra RJH, Witjes MJH, van Ooijen PMA. Recurrent Convolutional Neural Networks for 3D Mandible Segmentation in Computed Tomography. J Pers Med 2021; 11:jpm11060492. [PMID: 34072714 PMCID: PMC8229770 DOI: 10.3390/jpm11060492] [Citation(s) in RCA: 10] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/03/2021] [Revised: 05/26/2021] [Accepted: 05/28/2021] [Indexed: 12/24/2022] Open
Abstract
Purpose: Classic encoder–decoder-based convolutional neural network (EDCNN) approaches cannot accurately segment detailed anatomical structures of the mandible in computed tomography (CT), for instance, condyles and coronoids of the mandible, which are often affected by noise and metal artifacts. The main reason is that EDCNN approaches ignore the anatomical connectivity of the organs. In this paper, we propose a novel CNN-based 3D mandible segmentation approach that has the ability to accurately segment detailed anatomical structures. Methods: Different from the classic EDCNNs that need to slice or crop the whole CT scan into 2D slices or 3D patches during the segmentation process, our proposed approach can perform mandible segmentation on complete 3D CT scans. The proposed method, namely, RCNNSeg, adopts the structure of the recurrent neural networks to form a directed acyclic graph in order to enable recurrent connections between adjacent nodes to retain their connectivity. Each node then functions as a classic EDCNN to segment a single slice in the CT scan. Our proposed approach can perform 3D mandible segmentation on sequential data of any varied lengths and does not require a large computation cost. The proposed RCNNSeg was evaluated on 109 head and neck CT scans from a local dataset and 40 scans from the PDDCA public dataset. The final accuracy of the proposed RCNNSeg was evaluated by calculating the Dice similarity coefficient (DSC), average symmetric surface distance (ASD), and 95% Hausdorff distance (95HD) between the reference standard and the automated segmentation. Results: The proposed RCNNSeg outperforms the EDCNN-based approaches on both datasets and yields superior quantitative and qualitative performances when compared to the state-of-the-art approaches on the PDDCA dataset. The proposed RCNNSeg generated the most accurate segmentations with an average DSC of 97.48%, ASD of 0.2170 mm, and 95HD of 2.6562 mm on 109 CT scans, and an average DSC of 95.10%, ASD of 0.1367 mm, and 95HD of 1.3560 mm on the PDDCA dataset. Conclusions: The proposed RCNNSeg method generated more accurate automated segmentations than those of the other classic EDCNN segmentation techniques in terms of quantitative and qualitative evaluation. The proposed RCNNSeg has potential for automatic mandible segmentation by learning spatially structured information.
Collapse
Affiliation(s)
- Bingjiang Qiu
- 3D Lab, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands; (B.Q.); (J.K.); (H.H.G.); (M.J.H.W.)
- Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands;
- Data Science Center in Health (DASH), University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands
| | - Jiapan Guo
- Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands;
- Data Science Center in Health (DASH), University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands
- Correspondence:
| | - Joep Kraeima
- 3D Lab, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands; (B.Q.); (J.K.); (H.H.G.); (M.J.H.W.)
- Department of Oral and Maxillofacial Surgery, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands
| | - Haye Hendrik Glas
- 3D Lab, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands; (B.Q.); (J.K.); (H.H.G.); (M.J.H.W.)
- Department of Oral and Maxillofacial Surgery, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands
| | - Weichuan Zhang
- Institute for Integrated and Intelligent System, Griffith University, Nathan, QLD 4111, Australia;
- CSIRO Data61, Epping, NSW 1710, Australia
| | - Ronald J. H. Borra
- Medical Imaging Center (MIC), University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands;
| | - Max Johannes Hendrikus Witjes
- 3D Lab, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands; (B.Q.); (J.K.); (H.H.G.); (M.J.H.W.)
- Department of Oral and Maxillofacial Surgery, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands
| | - Peter M. A. van Ooijen
- Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands;
- Data Science Center in Health (DASH), University Medical Center Groningen, University of Groningen, Hanzeplein 1, 9713GZ Groningen, The Netherlands
| |
Collapse
|
16
|
Multiview Self-Supervised Segmentation for OARs Delineation in Radiotherapy. EVIDENCE-BASED COMPLEMENTARY AND ALTERNATIVE MEDICINE 2021; 2021:8894222. [PMID: 33747116 PMCID: PMC7954615 DOI: 10.1155/2021/8894222] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 09/15/2020] [Revised: 10/31/2020] [Accepted: 02/24/2021] [Indexed: 11/17/2022]
Abstract
Radiotherapy has become a common treatment option for head and neck (H&N) cancer, and organs at risk (OARs) need to be delineated to implement a high conformal dose distribution. Manual drawing of OARs is time consuming and inaccurate, so automatic drawing based on deep learning models has been proposed to accurately delineate the OARs. However, state-of-the-art performance usually requires a decent amount of delineation, but collecting pixel-level manual delineations is labor intensive and may not be necessary for representation learning. Encouraged by the recent progress in self-supervised learning, this study proposes and evaluates a novel multiview contrastive representation learning to boost the models from unlabelled data. The proposed learning architecture leverages three views of CTs (coronal, sagittal, and transverse plane) to collect positive and negative training samples. Specifically, a CT in 3D is first projected into three 2D views (coronal, sagittal, and transverse planes), then a convolutional neural network takes 3 views as inputs and outputs three individual representations in latent space, and finally, a contrastive loss is used to pull representation of different views of the same image closer (“positive pairs”) and push representations of views from different images (“negative pairs”) apart. To evaluate performance, we collected 220 CT images in H&N cancer patients. The experiment demonstrates that our method significantly improves quantitative performance over the state-of-the-art (from 83% to 86% in absolute Dice scores). Thus, our method provides a powerful and principled means to deal with the label-scarce problem.
Collapse
|
17
|
Aliotta E, Nourzadeh H, Choi W, Leandro Alves VG, Siebers JV. An Automated Workflow to Improve Efficiency in Radiation Therapy Treatment Planning by Prioritizing Organs at Risk. Adv Radiat Oncol 2020; 5:1324-1333. [PMID: 33305095 PMCID: PMC7718498 DOI: 10.1016/j.adro.2020.06.012] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/22/2020] [Revised: 04/15/2020] [Accepted: 06/16/2020] [Indexed: 11/28/2022] Open
Abstract
PURPOSE Manual delineation (MD) of organs at risk (OAR) is time and labor intensive. Auto-delineation (AD) can reduce the need for MD, but because current algorithms are imperfect, manual review and modification is still typically used. Recognizing that many OARs are sufficiently far from important dose levels that they do not pose a realistic risk, we hypothesize that some OARs can be excluded from MD and manual review with no clinical effect. The purpose of this study was to develop a method that automatically identifies these OARs and enables more efficient workflows that incorporate AD without degrading clinical quality. METHODS AND MATERIALS Preliminary dose map estimates were generated for n = 10 patients with head and neck cancers using only prescription and target-volume information. Conservative estimates of clinical OAR objectives were computed using AD structures with spatial expansion buffers to account for potential delineation uncertainties. OARs with estimated dose metrics below clinical tolerances were deemed low priority and excluded from MD and/or manual review. Final plans were then optimized using high-priority MD OARs and low-priority AD OARs and compared with reference plans generated using all MD OARs. Multiple different spatial buffers were used to accommodate different potential delineation uncertainties. RESULTS Sixty-seven out of 201 total OARs were identified as low-priority using the proposed methodology, which permitted a 33% reduction in structures requiring manual delineation/review. Plans optimized using low-priority AD OARs without review or modification met all planning objectives that were met when all MD OARs were used, indicating clinical equivalence. CONCLUSIONS Prioritizing OARs using estimated dose distributions allowed a substantial reduction in required MD and review without affecting clinically relevant dosimetry.
Collapse
Affiliation(s)
- Eric Aliotta
- Department of Radiation Oncology, University of Virginia, Charlottesville, Virginia
| | - Hamidreza Nourzadeh
- Department of Radiation Oncology, University of Virginia, Charlottesville, Virginia
| | - Wookjin Choi
- Department of Radiation Oncology, University of Virginia, Charlottesville, Virginia
| | | | - Jeffrey V. Siebers
- Department of Radiation Oncology, University of Virginia, Charlottesville, Virginia
| |
Collapse
|
18
|
Unkelbach J, Bortfeld T, Cardenas CE, Gregoire V, Hager W, Heijmen B, Jeraj R, Korreman SS, Ludwig R, Pouymayou B, Shusharina N, Söderberg J, Toma-Dasu I, Troost EGC, Vasquez Osorio E. The role of computational methods for automating and improving clinical target volume definition. Radiother Oncol 2020; 153:15-25. [PMID: 33039428 DOI: 10.1016/j.radonc.2020.10.002] [Citation(s) in RCA: 26] [Impact Index Per Article: 6.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/10/2020] [Revised: 10/01/2020] [Accepted: 10/01/2020] [Indexed: 12/25/2022]
Abstract
Treatment planning in radiotherapy distinguishes three target volume concepts: the gross tumor volume (GTV), the clinical target volume (CTV), and the planning target volume (PTV). Over time, GTV definition and PTV margins have improved through the development of novel imaging techniques and better image guidance, respectively. CTV definition is sometimes considered the weakest element in the planning process. CTV definition is particularly complex since the extension of microscopic disease cannot be seen using currently available in-vivo imaging techniques. Instead, CTV definition has to incorporate knowledge of the patterns of tumor progression. While CTV delineation has largely been considered the domain of radiation oncologists, this paper, arising from a 2019 ESTRO Physics research workshop, discusses the contributions that medical physics and computer science can make by developing computational methods to support CTV definition. First, we overview the role of image segmentation algorithms, which may in part automate CTV delineation through segmentation of lymph node stations or normal tissues representing anatomical boundaries of microscopic tumor progression. The recent success of deep convolutional neural networks has also enabled learning entire CTV delineations from examples. Second, we discuss the use of mathematical models of tumor progression for CTV definition, using as example the application of glioma growth models to facilitate GTV-to-CTV expansion for glioblastoma that is consistent with neuroanatomy. We further consider statistical machine learning models to quantify lymphatic metastatic progression of tumors, which may eventually improve elective CTV definition. Lastly, we discuss approaches to incorporate uncertainty in CTV definition into treatment plan optimization as well as general limitations of the CTV concept in the case of infiltrating tumors without natural boundaries.
Collapse
Affiliation(s)
- Jan Unkelbach
- Department of Radiation Oncology, University Hospital Zurich, Switzerland.
| | - Thomas Bortfeld
- Division of Radiation Biophysics, Massachusetts General Hospital and Harvard Medical School, Boston, USA
| | - Carlos E Cardenas
- Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, Houston, USA
| | | | - Wille Hager
- Department of Physics, Medical Radiation Physics, Stockholm University and Department of Oncology and Pathology, Medical Radiation Physics, Karolinska Institutet, Stockholm, Sweden
| | - Ben Heijmen
- Department of Radiation Oncology, Erasmus University Medical Center (Erasmus MC), Rotterdam, The Netherlands
| | - Robert Jeraj
- Department of Medical Physics, University of Wisconsin, Madison, USA
| | - Stine S Korreman
- Department of Oncology and Danish Center for Particle Therapy, Aarhus University Hospital, Department of Clinical Medicine, Aarhus University, Aarhus, Denmark
| | - Roman Ludwig
- Department of Radiation Oncology, University Hospital Zurich, Switzerland
| | - Bertrand Pouymayou
- Department of Radiation Oncology, University Hospital Zurich, Switzerland
| | - Nadya Shusharina
- Division of Radiation Biophysics, Massachusetts General Hospital and Harvard Medical School, Boston, USA
| | | | - Iuliana Toma-Dasu
- Department of Physics, Medical Radiation Physics, Stockholm University and Department of Oncology and Pathology, Medical Radiation Physics, Karolinska Institutet, Stockholm, Sweden
| | - Esther G C Troost
- Dept. of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany; OncoRay - National Center for Radiation Research in Oncology, Dresden, Germany; Helmholtz-Zentrum Dresden - Rossendorf, Institute of Radiooncology - OncoRay, Dresden, Germany
| | - Eliana Vasquez Osorio
- Division of Cancer Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, UK
| |
Collapse
|
19
|
Zhang D, Yang Z, Jiang S, Zhou Z, Meng M, Wang W. Automatic segmentation and applicator reconstruction for CT-based brachytherapy of cervical cancer using 3D convolutional neural networks. J Appl Clin Med Phys 2020; 21:158-169. [PMID: 32991783 PMCID: PMC7592978 DOI: 10.1002/acm2.13024] [Citation(s) in RCA: 26] [Impact Index Per Article: 6.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/06/2020] [Revised: 08/05/2020] [Accepted: 08/07/2020] [Indexed: 12/29/2022] Open
Abstract
In this study, we present deep learning-based approaches to automatic segmentation and applicator reconstruction with high accuracy and efficiency in the planning computed tomography (CT) for cervical cancer brachytherapy (BT). A novel three-dimensional (3D) convolutional neural network (CNN) architecture was proposed and referred to as DSD-UNET. The dataset of 91 patients received CT-based BT of cervical cancer was used to train and test DSD-UNET model for auto-segmentation of high-risk clinical target volume (HR-CTV) and organs at risk (OARs). Automatic applicator reconstruction was achieved with DSD-UNET-based segmentation of applicator components followed by 3D skeletonization and polynomial curve fitting. Digitization of the channel paths for tandem and ovoid applicator in the planning CT was evaluated utilizing the data from 32 patients. Dice similarity coefficient (DSC), Jaccard Index (JI), and Hausdorff distance (HD) were used to quantitatively evaluate the accuracy. The segmentation performance of DSD-UNET was compared with that of 3D U-Net. Results showed that DSD-UNET method outperformed 3D U-Net on segmentations of all the structures. The mean DSC values of DSD-UNET method were 86.9%, 82.9%, and 82.1% for bladder, HR-CTV, and rectum, respectively. For the performance of automatic applicator reconstruction, outstanding segmentation accuracy was first achieved for the intrauterine and ovoid tubes (average DSC value of 92.1%, average HD value of 2.3 mm). Finally, HDs between the channel paths determined automatically and manually were 0.88 ± 0.12 mm, 0.95 ± 0.16 mm, and 0.96 ± 0.15 mm for the intrauterine, left ovoid, and right ovoid tubes, respectively. The proposed DSD-UNET method outperformed the 3D U-Net and could segment HR-CTV, bladder, and rectum with relatively good accuracy. Accurate digitization of the channel paths could be achieved with the DSD-UNET-based method. The proposed approaches could be useful to improve the efficiency and consistency of treatment planning for cervical cancer BT.
Collapse
Affiliation(s)
- Daguang Zhang
- School of Mechanical EngineeringTianjin UniversityTianjinChina
| | - Zhiyong Yang
- School of Mechanical EngineeringTianjin UniversityTianjinChina
| | - Shan Jiang
- School of Mechanical EngineeringTianjin UniversityTianjinChina
| | - Zeyang Zhou
- School of Mechanical EngineeringTianjin UniversityTianjinChina
| | - Maobin Meng
- Department of Radiation OncologyTianjin Medical University Cancer Institute and HospitalTianjinChina
| | - Wei Wang
- Department of Radiation OncologyTianjin Medical University Cancer Institute and HospitalTianjinChina
| |
Collapse
|
20
|
Vrtovec T, Močnik D, Strojan P, Pernuš F, Ibragimov B. Auto-segmentation of organs at risk for head and neck radiotherapy planning: From atlas-based to deep learning methods. Med Phys 2020; 47:e929-e950. [PMID: 32510603 DOI: 10.1002/mp.14320] [Citation(s) in RCA: 80] [Impact Index Per Article: 20.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/26/2019] [Revised: 05/27/2020] [Accepted: 05/29/2020] [Indexed: 02/06/2023] Open
Abstract
Radiotherapy (RT) is one of the basic treatment modalities for cancer of the head and neck (H&N), which requires a precise spatial description of the target volumes and organs at risk (OARs) to deliver a highly conformal radiation dose to the tumor cells while sparing the healthy tissues. For this purpose, target volumes and OARs have to be delineated and segmented from medical images. As manual delineation is a tedious and time-consuming task subjected to intra/interobserver variability, computerized auto-segmentation has been developed as an alternative. The field of medical imaging and RT planning has experienced an increased interest in the past decade, with new emerging trends that shifted the field of H&N OAR auto-segmentation from atlas-based to deep learning-based approaches. In this review, we systematically analyzed 78 relevant publications on auto-segmentation of OARs in the H&N region from 2008 to date, and provided critical discussions and recommendations from various perspectives: image modality - both computed tomography and magnetic resonance image modalities are being exploited, but the potential of the latter should be explored more in the future; OAR - the spinal cord, brainstem, and major salivary glands are the most studied OARs, but additional experiments should be conducted for several less studied soft tissue structures; image database - several image databases with the corresponding ground truth are currently available for methodology evaluation, but should be augmented with data from multiple observers and multiple institutions; methodology - current methods have shifted from atlas-based to deep learning auto-segmentation, which is expected to become even more sophisticated; ground truth - delineation guidelines should be followed and participation of multiple experts from multiple institutions is recommended; performance metrics - the Dice coefficient as the standard volumetric overlap metrics should be accompanied with at least one distance metrics, and combined with clinical acceptability scores and risk assessments; segmentation performance - the best performing methods achieve clinically acceptable auto-segmentation for several OARs, however, the dosimetric impact should be also studied to provide clinically relevant endpoints for RT planning.
Collapse
Affiliation(s)
- Tomaž Vrtovec
- Faculty Electrical Engineering, University of Ljubljana, Tržaška cesta 25, Ljubljana, SI-1000, Slovenia
| | - Domen Močnik
- Faculty Electrical Engineering, University of Ljubljana, Tržaška cesta 25, Ljubljana, SI-1000, Slovenia
| | - Primož Strojan
- Institute of Oncology Ljubljana, Zaloška cesta 2, Ljubljana, SI-1000, Slovenia
| | - Franjo Pernuš
- Faculty Electrical Engineering, University of Ljubljana, Tržaška cesta 25, Ljubljana, SI-1000, Slovenia
| | - Bulat Ibragimov
- Faculty Electrical Engineering, University of Ljubljana, Tržaška cesta 25, Ljubljana, SI-1000, Slovenia.,Department of Computer Science, University of Copenhagen, Universitetsparken 1, Copenhagen, D-2100, Denmark
| |
Collapse
|
21
|
Chen W, Li Y, Dyer BA, Feng X, Rao S, Benedict SH, Chen Q, Rong Y. Deep learning vs. atlas-based models for fast auto-segmentation of the masticatory muscles on head and neck CT images. Radiat Oncol 2020; 15:176. [PMID: 32690103 PMCID: PMC7372849 DOI: 10.1186/s13014-020-01617-0] [Citation(s) in RCA: 35] [Impact Index Per Article: 8.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/20/2020] [Accepted: 07/13/2020] [Indexed: 12/16/2022] Open
Abstract
BACKGROUND Impaired function of masticatory muscles will lead to trismus. Routine delineation of these muscles during planning may improve dose tracking and facilitate dose reduction resulting in decreased radiation-related trismus. This study aimed to compare a deep learning model with a commercial atlas-based model for fast auto-segmentation of the masticatory muscles on head and neck computed tomography (CT) images. MATERIAL AND METHODS Paired masseter (M), temporalis (T), medial and lateral pterygoid (MP, LP) muscles were manually segmented on 56 CT images. CT images were randomly divided into training (n = 27) and validation (n = 29) cohorts. Two methods were used for automatic delineation of masticatory muscles (MMs): Deep learning auto-segmentation (DLAS) and atlas-based auto-segmentation (ABAS). The automatic algorithms were evaluated using Dice similarity coefficient (DSC), recall, precision, Hausdorff distance (HD), HD95, and mean surface distance (MSD). A consolidated score was calculated by normalizing the metrics against interobserver variability and averaging over all patients. Differences in dose (∆Dose) to MMs for DLAS and ABAS segmentations were assessed. A paired t-test was used to compare the geometric and dosimetric difference between DLAS and ABAS methods. RESULTS DLAS outperformed ABAS in delineating all MMs (p < 0.05). The DLAS mean DSC for M, T, MP, and LP ranged from 0.83 ± 0.03 to 0.89 ± 0.02, the ABAS mean DSC ranged from 0.79 ± 0.05 to 0.85 ± 0.04. The mean value for recall, HD, HD95, MSD also improved with DLAS for auto-segmentation. Interobserver variation revealed the highest variability in DSC and MSD for both T and MP, and the highest scores were achieved for T by both automatic algorithms. With few exceptions, the mean ∆D98%, ∆D95%, ∆D50%, and ∆D2% for all structures were below 10% for DLAS and ABAS and had no detectable statistical difference (P > 0.05). DLAS based contours had dose endpoints more closely matched with that of the manually segmented when compared with ABAS. CONCLUSIONS DLAS auto-segmentation of masticatory muscles for the head and neck radiotherapy had improved segmentation accuracy compared with ABAS with no qualitative difference in dosimetric endpoints compared to manually segmented contours.
Collapse
Affiliation(s)
- Wen Chen
- Department of Radiation Oncology, Xiangya Hospital, Central South University, Changsha, China.,Department of Radiation Oncology, University of California Davis Medical Center, 4501 X Street, Suite 0152, Sacramento, California, 95817, USA
| | - Yimin Li
- Department of Radiation Oncology, Xiamen Cancer Center, The First Affiliated Hospital of Xiamen University, Xiamen, Fujian, China
| | - Brandon A Dyer
- Department of Radiation Oncology, University of California Davis Medical Center, 4501 X Street, Suite 0152, Sacramento, California, 95817, USA.,Department of Radiation Oncology, University of Washington, Seattle, WA, USA
| | - Xue Feng
- Carina Medical LLC, 145 Graham Ave, A168, Lexington, KY, 40536, USA
| | - Shyam Rao
- Department of Radiation Oncology, University of California Davis Medical Center, 4501 X Street, Suite 0152, Sacramento, California, 95817, USA
| | - Stanley H Benedict
- Department of Radiation Oncology, University of California Davis Medical Center, 4501 X Street, Suite 0152, Sacramento, California, 95817, USA
| | - Quan Chen
- Carina Medical LLC, 145 Graham Ave, A168, Lexington, KY, 40536, USA. .,Department of Radiation Oncology, Markey Cancer Center, University of Kentucky, RM CC063, 800 Rose St, Lexington, KY, 40536, USA.
| | - Yi Rong
- Department of Radiation Oncology, University of California Davis Medical Center, 4501 X Street, Suite 0152, Sacramento, California, 95817, USA.
| |
Collapse
|
22
|
Kim N, Chang JS, Kim YB, Kim JS. Atlas-based auto-segmentation for postoperative radiotherapy planning in endometrial and cervical cancers. Radiat Oncol 2020; 15:106. [PMID: 32404123 PMCID: PMC7218589 DOI: 10.1186/s13014-020-01562-y] [Citation(s) in RCA: 16] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/22/2020] [Accepted: 05/05/2020] [Indexed: 12/22/2022] Open
Abstract
Background Since intensity-modulated radiation therapy (IMRT) has become popular for the treatment of gynecologic cancers, the contouring process has become more critical. This study evaluated the feasibility of atlas-based auto-segmentation (ABAS) for contouring in patients with endometrial and cervical cancers. Methods A total of 75 sets of planning CT images from 75 patients were collected. Contours for the pelvic nodal clinical target volume (CTV), femur, and bladder were carefully generated by two skilled radiation oncologists. Of 75 patients, 60 were randomly registered in three different atlas libraries for ABAS in groups of 20, 40, or 60. ABAS was conducted in 15 patients, followed by manual correction (ABASc). The time required to generate all contours was recorded, and the accuracy of segmentation was assessed using Dice’s coefficient (DC) and the Hausdorff distance (HD) and compared to those of manually delineated contours. Results For ABAS-CTV, the best results were achieved with groups of 60 patients (DC, 0.79; HD, 19.7 mm) and the worst results with groups of 20 patients (DC, 0.75; p = 0.012; HD, 21.3 mm; p = 0.002). ABASc-CTV performed better than ABAS-CTV in terms of both HD and DC (ABASc [n = 60]; DC, 0.84; HD, 15.6 mm; all p < 0.017). ABAS required an average of 45.1 s, whereas ABASc required 191.1 s; both methods required less time than the manual methods (p < 0.001). Both ABAS-Femur and simultaneous ABAS-Bilateral-femurs showed satisfactory performance, regardless of the atlas library used (DC > 0.9 and HD ≤10.0 mm), with significant time reduction compared to that needed for manual delineation (p < 0.001). However, ABAS-Bladder did not prove to be feasible, with inferior results regardless of library size (DC < 0.6 and HD > 40 mm). Furthermore, ABASc-Bladder required a longer processing time than manual contouring to achieve the same accuracy. Conclusions ABAS could help physicians to delineate the CTV and organs-at-risk (e.g., femurs) in IMRT planning considering its consistency, efficacy, and accuracy.
Collapse
Affiliation(s)
- Nalee Kim
- Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul, 03722, South Korea.,Department of Radiation Oncology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul, 06351, Korea
| | - Jee Suk Chang
- Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul, 03722, South Korea
| | - Yong Bae Kim
- Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul, 03722, South Korea
| | - Jin Sung Kim
- Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul, 03722, South Korea.
| |
Collapse
|
23
|
Shusharina N, Söderberg J, Edmunds D, Löfman F, Shih H, Bortfeld T. Automated delineation of the clinical target volume using anatomically constrained 3D expansion of the gross tumor volume. Radiother Oncol 2020; 146:37-43. [PMID: 32114264 PMCID: PMC10660950 DOI: 10.1016/j.radonc.2020.01.028] [Citation(s) in RCA: 23] [Impact Index Per Article: 5.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/27/2019] [Revised: 01/29/2020] [Accepted: 01/30/2020] [Indexed: 01/05/2023]
Abstract
PURPOSE Delineation of the clinical target volume (CTV) is arguably the weakest link in the treatment planning chain. This work aims to support clinicians in this crucial task. METHODS AND MATERIALS While the CTV itself is ambiguous, it is much easier to identify structures that do not belong to the CTV and serve as barriers to the spread of the disease. We segment the known barrier structures using a convolutional neural network (CNN). The CTV is then obtained by starting from the manually delineated gross tumor volume (GTV) and expanding it while taking into account the barrier structures. Mathematically, we define the CTV as an iso-surface in the 3D map of shortest paths of all voxels from the GTV. The shortest paths are found with the Dijkstra algorithm. While the method is generally applicable, we test it on 206 glioma and glioblastoma cases. RESULTS The auto-segmented barrier structures for the brain cases include the ventricles, falx cerebri, tentorium cerebelli, brain sinuses, and the outer surface of the brain. Manual and auto-segmented barrier structures agree with surface Dice Similarity Coefficients (DSC) ranging from 0.91 to 0.97 at 2 mm tolerance. Comparison of manual and automatically delineated CTVs shows a median surface DSC of 0.79. CONCLUSIONS Barrier structures for CTV definition can be auto-delineated with outstanding precision using a CNN. An algorithm for automated calculation of the CTV by 3D expansion of the GTV while respecting anatomical barriers has been developed. It shows good agreement with manual CTV definition for brain tumors.
Collapse
Affiliation(s)
- Nadya Shusharina
- Division of Radiation Biophysics, Massachusetts General Hospital and Harvard Medical School, Boston, USA
| | | | - David Edmunds
- Division of Radiation Biophysics, Massachusetts General Hospital and Harvard Medical School, Boston, USA
| | | | - Helen Shih
- Department of Radiation Oncology, Massachusetts General Hospital and Harvard Medical School, Boston, USA
| | - Thomas Bortfeld
- Division of Radiation Biophysics, Massachusetts General Hospital and Harvard Medical School, Boston, USA.
| |
Collapse
|
24
|
Haq R, Berry SL, Deasy JO, Hunt M, Veeraraghavan H. Dynamic multiatlas selection-based consensus segmentation of head and neck structures from CT images. Med Phys 2019; 46:5612-5622. [PMID: 31587300 DOI: 10.1002/mp.13854] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/06/2018] [Revised: 09/11/2019] [Accepted: 09/16/2019] [Indexed: 12/11/2022] Open
Abstract
PURPOSE Manual delineation of head and neck (H&N) organ-at-risk (OAR) structures for radiation therapy planning is time consuming and highly variable. Therefore, we developed a dynamic multiatlas selection-based approach for fast and reproducible segmentation. METHODS Our approach dynamically selects and weights the appropriate number of atlases for weighted label fusion and generates segmentations and consensus maps indicating voxel-wise agreement between different atlases. Atlases were selected for a target as those exceeding an alignment weight called dynamic atlas attention index. Alignment weights were computed at the image level and called global weighted voting (GWV) or at the structure level and called structure weighted voting (SWV) by using a normalized metric computed as the sum of squared distances of computed tomography (CT)-radiodensity and modality-independent neighborhood descriptors (extracting edge information). Performance comparisons were performed using 77 H&N CT images from an internal Memorial Sloan-Kettering Cancer Center dataset (N = 45) and an external dataset (N = 32) using Dice similarity coefficient (DSC), Hausdorff distance (HD), 95th percentile of HD, median of maximum surface distance, and volume ratio error against expert delineation. Pairwise DSC accuracy comparisons of proposed (GWV, SWV) vs single best atlas (BA) or majority voting (MV) methods were performed using Wilcoxon rank-sum tests. RESULTS Both SWV and GWV methods produced significantly better segmentation accuracy than BA (P < 0.001) and MV (P < 0.001) for all OARs within both datasets. SWV generated the most accurate segmentations with DSC of: 0.88 for oral cavity, 0.85 for mandible, 0.84 for cord, 0.76 for brainstem and parotids, 0.71 for larynx, and 0.60 for submandibular glands. SWV's accuracy exceeded GWV's for submandibular glands (DSC = 0.60 vs 0.52, P = 0.019). CONCLUSIONS The contributed SWV and GWV methods generated more accurate automated segmentations than the other two multiatlas-based segmentation techniques. The consensus maps could be combined with segmentations to visualize voxel-wise consensus between atlases within OARs during manual review.
Collapse
Affiliation(s)
- Rabia Haq
- Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, 10065, USA
| | - Sean L Berry
- Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, 10065, USA
| | - Joseph O Deasy
- Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, 10065, USA
| | - Margie Hunt
- Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, 10065, USA
| | - Harini Veeraraghavan
- Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY, 10065, USA
| |
Collapse
|
25
|
Tang H, Chen X, Liu Y, Lu Z, You J, Yang M, Yao S, Zhao G, Xu Y, Chen T, Liu Y, Xie X. Clinically applicable deep learning framework for organs at risk delineation in CT images. NAT MACH INTELL 2019. [DOI: 10.1038/s42256-019-0099-z] [Citation(s) in RCA: 63] [Impact Index Per Article: 12.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/09/2023]
|
26
|
Sabater S, Pastor-Juan M, Berenguer R, Lozano-Setien E, Andres I, Tercero-Azorin M, Sevillano M, Jimenez-Jimenez E, Rovirosa A, Korte E, Arenas M. An MRI comparative image evaluation under diagnostic and radiotherapy planning set-ups using a carbon fibre tabletop for pelvic radiotherapy. Cancer Radiother 2019; 23:296-303. [DOI: 10.1016/j.canrad.2019.01.001] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/14/2018] [Revised: 01/04/2019] [Accepted: 01/10/2019] [Indexed: 01/31/2023]
|
27
|
Kosmin M, Ledsam J, Romera-Paredes B, Mendes R, Moinuddin S, de Souza D, Gunn L, Kelly C, Hughes C, Karthikesalingam A, Nutting C, Sharma R. Rapid advances in auto-segmentation of organs at risk and target volumes in head and neck cancer. Radiother Oncol 2019; 135:130-140. [DOI: 10.1016/j.radonc.2019.03.004] [Citation(s) in RCA: 52] [Impact Index Per Article: 10.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/17/2018] [Revised: 02/10/2019] [Accepted: 03/04/2019] [Indexed: 11/25/2022]
|
28
|
Rattan R, Kataria T, Banerjee S, Goyal S, Gupta D, Pandita A, Bisht S, Narang K, Mishra SR. Artificial intelligence in oncology, its scope and future prospects with specific reference to radiation oncology. BJR Open 2019; 1:20180031. [PMID: 33178922 PMCID: PMC7592433 DOI: 10.1259/bjro.20180031] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.2] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/12/2018] [Revised: 04/01/2019] [Accepted: 04/12/2019] [Indexed: 01/02/2023] Open
Abstract
OBJECTIVE Artificial intelligence (AI) seems to be bridging the gap between the acquisition of data and its meaningful interpretation. These approaches, have shown outstanding capabilities, outperforming most classification and regression methods to date and the ability to automatically learn the most suitable data representation for the task at hand and present it for better correlation. This article tries to sensitize the practising radiation oncologists to understand where the potential role of AI lies and what further can be achieved with it. METHODS AND MATERIALS Contemporary literature was searched and the available literature was sorted and an attempt at writing a comprehensive non-systematic review was made. RESULTS The article addresses various areas in oncology, especially in the field of radiation oncology, where the work based on AI has been done. Whether it's the screening modalities, or diagnosis or the prognostic assays, AI has come with more accurately defining results and survival of patients. Various steps and protocols in radiation oncology are now using AI-based methods, like in the steps of planning, segmentation and delivery of radiation. Benefit of AI across all the platforms of health sector may lead to a more refined and personalized medicine in near future. CONCLUSION AI with the use of machine learning and artificial neural networks has come up with faster and more accurate solutions for the problems faced by oncologist. The uses of AI,are likely to get increased exponentially . However, concerns regarding demographic discrepancies in relation to patients, disease and their natural history and reports of manipulation of AI, the ultimate responsibility will rest on the treating physicians.
Collapse
Affiliation(s)
- Rajit Rattan
- Division of Radiation Oncology, Medanta- The Medicity, Gurgaon, Haryana, India
| | - Tejinder Kataria
- Division of Radiation Oncology, Medanta- The Medicity, Gurgaon, Haryana, India
| | - Susovan Banerjee
- Division of Radiation Oncology, Medanta- The Medicity, Gurgaon, Haryana, India
| | - Shikha Goyal
- Division of Radiation Oncology, Medanta- The Medicity, Gurgaon, Haryana, India
| | - Deepak Gupta
- Division of Radiation Oncology, Medanta- The Medicity, Gurgaon, Haryana, India
| | - Akshi Pandita
- Department of Dermatology, P. N. Behl Skin Institute, New Delhi, India
| | - Shyam Bisht
- Division of Radiation Oncology, Medanta- The Medicity, Gurgaon, Haryana, India
- Department of Dermatology, P. N. Behl Skin Institute, New Delhi, India
| | - Kushal Narang
- Division of Radiation Oncology, Medanta- The Medicity, Gurgaon, Haryana, India
| | | |
Collapse
|
29
|
Wu X, Udupa JK, Tong Y, Odhner D, Pednekar GV, Simone CB, McLaughlin D, Apinorasethkul C, Apinorasethkul O, Lukens J, Mihailidis D, Shammo G, James P, Tiwari A, Wojtowicz L, Camaratta J, Torigian DA. AAR-RT - A system for auto-contouring organs at risk on CT images for radiation therapy planning: Principles, design, and large-scale evaluation on head-and-neck and thoracic cancer cases. Med Image Anal 2019; 54:45-62. [PMID: 30831357 PMCID: PMC6499546 DOI: 10.1016/j.media.2019.01.008] [Citation(s) in RCA: 23] [Impact Index Per Article: 4.6] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/04/2018] [Revised: 12/04/2018] [Accepted: 01/26/2019] [Indexed: 12/25/2022]
Abstract
Contouring (segmentation) of Organs at Risk (OARs) in medical images is required for accurate radiation therapy (RT) planning. In current clinical practice, OAR contouring is performed with low levels of automation. Although several approaches have been proposed in the literature for improving automation, it is difficult to gain an understanding of how well these methods would perform in a realistic clinical setting. This is chiefly due to three key factors - small number of patient studies used for evaluation, lack of performance evaluation as a function of input image quality, and lack of precise anatomic definitions of OARs. In this paper, extending our previous body-wide Automatic Anatomy Recognition (AAR) framework to RT planning of OARs in the head and neck (H&N) and thoracic body regions, we present a methodology called AAR-RT to overcome some of these hurdles. AAR-RT follows AAR's 3-stage paradigm of model-building, object-recognition, and object-delineation. Model-building: Three key advances were made over AAR. (i) AAR-RT (like AAR) starts off with a computationally precise definition of the two body regions and all of their OARs. Ground truth delineations of OARs are then generated following these definitions strictly. We retrospectively gathered patient data sets and the associated contour data sets that have been created previously in routine clinical RT planning from our Radiation Oncology department and mended the contours to conform to these definitions. We then derived an Object Quality Score (OQS) for each OAR sample and an Image Quality Score (IQS) for each study, both on a 1-to-10 scale, based on quality grades assigned to each OAR sample following 9 key quality criteria. Only studies with high IQS and high OQS for all of their OARs were selected for model building. IQS and OQS were employed for evaluating AAR-RT's performance as a function of image/object quality. (ii) In place of the previous hand-crafted hierarchy for organizing OARs in AAR, we devised a method to find an optimal hierarchy for each body region. Optimality was based on minimizing object recognition error. (iii) In addition to the parent-to-child relationship encoded in the hierarchy in previous AAR, we developed a directed probability graph technique to further improve recognition accuracy by learning and encoding in the model "steady" relationships that may exist among OAR boundaries in the three orthogonal planes. Object-recognition: The two key improvements over the previous approach are (i) use of the optimal hierarchy for actual recognition of OARs in a given image, and (ii) refined recognition by making use of the trained probability graph. Object-delineation: We use a kNN classifier confined to the fuzzy object mask localized by the recognition step and then fit optimally the fuzzy mask to the kNN-derived voxel cluster to bring back shape constraint on the object. We evaluated AAR-RT on 205 thoracic and 298 H&N (total 503) studies, involving both planning and re-planning scans and a total of 21 organs (9 - thorax, 12 - H&N). The studies were gathered from two patient age groups for each gender - 40-59 years and 60-79 years. The number of 3D OAR samples analyzed from the two body regions was 4301. IQS and OQS tended to cluster at the two ends of the score scale. Accordingly, we considered two quality groups for each gender - good and poor. Good quality data sets typically had OQS ≥ 6 and had distortions, artifacts, pathology etc. in not more than 3 slices through the object. The number of model-worthy data sets used for training were 38 for thorax and 36 for H&N, and the remaining 479 studies were used for testing AAR-RT. Accordingly, we created 4 anatomy models, one each for: Thorax male (20 model-worthy data sets), Thorax female (18 model-worthy data sets), H&N male (20 model-worthy data sets), and H&N female (16 model-worthy data sets). On "good" cases, AAR-RT's recognition accuracy was within 2 voxels and delineation boundary distance was within ∼1 voxel. This was similar to the variability observed between two dosimetrists in manually contouring 5-6 OARs in each of 169 studies. On "poor" cases, AAR-RT's errors hovered around 5 voxels for recognition and 2 voxels for boundary distance. The performance was similar on planning and replanning cases, and there was no gender difference in performance. AAR-RT's recognition operation is much more robust than delineation. Understanding object and image quality and how they influence performance is crucial for devising effective object recognition and delineation algorithms. OQS seems to be more important than IQS in determining accuracy. Streak artifacts arising from dental implants and fillings and beam hardening from bone pose the greatest challenge to auto-contouring methods.
Collapse
Affiliation(s)
- Xingyu Wu
- Medical Image Processing Group, Department of Radiology, University of Pennsylvania, 602 Goddard building, 3710 Hamilton Walk, 6th Floor, Rm 602W, Philadelphia, PA 19104, United States
| | - Jayaram K Udupa
- Medical Image Processing Group, Department of Radiology, University of Pennsylvania, 602 Goddard building, 3710 Hamilton Walk, 6th Floor, Rm 602W, Philadelphia, PA 19104, United States.
| | - Yubing Tong
- Medical Image Processing Group, Department of Radiology, University of Pennsylvania, 602 Goddard building, 3710 Hamilton Walk, 6th Floor, Rm 602W, Philadelphia, PA 19104, United States
| | - Dewey Odhner
- Medical Image Processing Group, Department of Radiology, University of Pennsylvania, 602 Goddard building, 3710 Hamilton Walk, 6th Floor, Rm 602W, Philadelphia, PA 19104, United States
| | - Gargi V Pednekar
- Quantitative Radiology Solutions, 3624 Market Street, Suite 5E, Philadelphia, PA 19104, United States
| | - Charles B Simone
- Department of Radiation Oncology, Maryland Proton Treatment Center, School of Medicine, University of Maryland 850W, Baltimore, MD 21201, United States
| | - David McLaughlin
- Quantitative Radiology Solutions, 3624 Market Street, Suite 5E, Philadelphia, PA 19104, United States
| | - Chavanon Apinorasethkul
- Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Ontida Apinorasethkul
- Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - John Lukens
- Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Dimitris Mihailidis
- Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Geraldine Shammo
- Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Paul James
- Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Akhil Tiwari
- Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Lisa Wojtowicz
- Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Joseph Camaratta
- Quantitative Radiology Solutions, 3624 Market Street, Suite 5E, Philadelphia, PA 19104, United States
| | - Drew A Torigian
- Medical Image Processing Group, Department of Radiology, University of Pennsylvania, 602 Goddard building, 3710 Hamilton Walk, 6th Floor, Rm 602W, Philadelphia, PA 19104, United States
| |
Collapse
|
30
|
Giraud P, Giraud P, Gasnier A, El Ayachy R, Kreps S, Foy JP, Durdux C, Huguet F, Burgun A, Bibault JE. Radiomics and Machine Learning for Radiotherapy in Head and Neck Cancers. Front Oncol 2019; 9:174. [PMID: 30972291 PMCID: PMC6445892 DOI: 10.3389/fonc.2019.00174] [Citation(s) in RCA: 60] [Impact Index Per Article: 12.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/12/2018] [Accepted: 02/28/2019] [Indexed: 12/13/2022] Open
Abstract
Introduction: An increasing number of parameters can be considered when making decisions in oncology. Tumor characteristics can also be extracted from imaging through the use of radiomics and add to this wealth of clinical data. Machine learning can encompass these parameters and thus enhance clinical decision as well as radiotherapy workflow. Methods: We performed a description of machine learning applications at each step of treatment by radiotherapy in head and neck cancers. We then performed a systematic review on radiomics and machine learning outcome prediction models in head and neck cancers. Results: Machine Learning has several promising applications in treatment planning with automatic organ at risk delineation improvements and adaptative radiotherapy workflow automation. It may also provide new approaches for Normal Tissue Complication Probability models. Radiomics may provide additional data on tumors for improved machine learning powered predictive models, not only on survival, but also on risk of distant metastasis, in field recurrence, HPV status and extra nodal spread. However, most studies provide preliminary data requiring further validation. Conclusion: Promising perspectives arise from machine learning applications and radiomics based models, yet further data are necessary for their implementation in daily care.
Collapse
Affiliation(s)
- Paul Giraud
- Radiation Oncology Department, Georges Pompidou European Hospital, Assistance Publique-Hôpitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,Cancer Research and Personalized Medicine-Integrated Cancer Research Center (SIRIC), Georges Pompidou European Hospital, Assistance Publique-Hôitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France
| | - Philippe Giraud
- Radiation Oncology Department, Georges Pompidou European Hospital, Assistance Publique-Hôpitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,Cancer Research and Personalized Medicine-Integrated Cancer Research Center (SIRIC), Georges Pompidou European Hospital, Assistance Publique-Hôitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France
| | - Anne Gasnier
- Radiation Oncology Department, Georges Pompidou European Hospital, Assistance Publique-Hôpitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,Cancer Research and Personalized Medicine-Integrated Cancer Research Center (SIRIC), Georges Pompidou European Hospital, Assistance Publique-Hôitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France
| | - Radouane El Ayachy
- Radiation Oncology Department, Georges Pompidou European Hospital, Assistance Publique-Hôpitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,Cancer Research and Personalized Medicine-Integrated Cancer Research Center (SIRIC), Georges Pompidou European Hospital, Assistance Publique-Hôitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France
| | - Sarah Kreps
- Radiation Oncology Department, Georges Pompidou European Hospital, Assistance Publique-Hôpitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,Cancer Research and Personalized Medicine-Integrated Cancer Research Center (SIRIC), Georges Pompidou European Hospital, Assistance Publique-Hôitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France
| | - Jean-Philippe Foy
- Department of Oral and Maxillo-Facial Surgery, Sorbonne University, Pitié-Salpêtriére Hospital, Paris, France.,Univ Lyon, Université Claude Bernard Lyon 1, INSERM 1052, CNRS 5286, Centre Léon Bérard, Centre de Recherche en Cancérologie de Lyon, Lyon, France
| | - Catherine Durdux
- Radiation Oncology Department, Georges Pompidou European Hospital, Assistance Publique-Hôpitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,Cancer Research and Personalized Medicine-Integrated Cancer Research Center (SIRIC), Georges Pompidou European Hospital, Assistance Publique-Hôitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France
| | - Florence Huguet
- Department of Radiation Oncology, Tenon University Hospital, Hôpitaux Universitaires Est Parisien, Sorbonne University Medical Faculty, Paris, France
| | - Anita Burgun
- Cancer Research and Personalized Medicine-Integrated Cancer Research Center (SIRIC), Georges Pompidou European Hospital, Assistance Publique-Hôitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,INSERM UMR 1138 Team 22: Information Sciences to support Personalized Medicine, Paris Descartes University, Sorbonne Paris Cité, Paris, France
| | - Jean-Emmanuel Bibault
- Radiation Oncology Department, Georges Pompidou European Hospital, Assistance Publique-Hôpitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,Cancer Research and Personalized Medicine-Integrated Cancer Research Center (SIRIC), Georges Pompidou European Hospital, Assistance Publique-Hôitaux de Paris, Paris Descartes University, Paris Sorbonne Cité, Paris, France.,INSERM UMR 1138 Team 22: Information Sciences to support Personalized Medicine, Paris Descartes University, Sorbonne Paris Cité, Paris, France
| |
Collapse
|
31
|
Zhu W, Huang Y, Zeng L, Chen X, Liu Y, Qian Z, Du N, Fan W, Xie X. AnatomyNet: Deep learning for fast and fully automated whole-volume segmentation of head and neck anatomy. Med Phys 2018; 46:576-589. [PMID: 30480818 DOI: 10.1002/mp.13300] [Citation(s) in RCA: 222] [Impact Index Per Article: 37.0] [Reference Citation Analysis] [Abstract] [Key Words] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/13/2018] [Revised: 11/06/2018] [Accepted: 11/07/2018] [Indexed: 12/20/2022] Open
Abstract
PURPOSE Radiation therapy (RT) is a common treatment option for head and neck (HaN) cancer. An important step involved in RT planning is the delineation of organs-at-risks (OARs) based on HaN computed tomography (CT). However, manually delineating OARs is time-consuming as each slice of CT images needs to be individually examined and a typical CT consists of hundreds of slices. Automating OARs segmentation has the benefit of both reducing the time and improving the quality of RT planning. Existing anatomy autosegmentation algorithms use primarily atlas-based methods, which require sophisticated atlas creation and cannot adequately account for anatomy variations among patients. In this work, we propose an end-to-end, atlas-free three-dimensional (3D) convolutional deep learning framework for fast and fully automated whole-volume HaN anatomy segmentation. METHODS Our deep learning model, called AnatomyNet, segments OARs from head and neck CT images in an end-to-end fashion, receiving whole-volume HaN CT images as input and generating masks of all OARs of interest in one shot. AnatomyNet is built upon the popular 3D U-net architecture, but extends it in three important ways: (a) a new encoding scheme to allow autosegmentation on whole-volume CT images instead of local patches or subsets of slices, (b) incorporating 3D squeeze-and-excitation residual blocks in encoding layers for better feature representation, and (c) a new loss function combining Dice scores and focal loss to facilitate the training of the neural model. These features are designed to address two main challenges in deep learning-based HaN segmentation: (a) segmenting small anatomies (i.e., optic chiasm and optic nerves) occupying only a few slices, and (b) training with inconsistent data annotations with missing ground truth for some anatomical structures. RESULTS We collected 261 HaN CT images to train AnatomyNet and used MICCAI Head and Neck Auto Segmentation Challenge 2015 as a benchmark dataset to evaluate the performance of AnatomyNet. The objective is to segment nine anatomies: brain stem, chiasm, mandible, optic nerve left, optic nerve right, parotid gland left, parotid gland right, submandibular gland left, and submandibular gland right. Compared to previous state-of-the-art results from the MICCAI 2015 competition, AnatomyNet increases Dice similarity coefficient by 3.3% on average. AnatomyNet takes about 0.12 s to fully segment a head and neck CT image of dimension 178 × 302 × 225, significantly faster than previous methods. In addition, the model is able to process whole-volume CT images and delineate all OARs in one pass, requiring little pre- or postprocessing. CONCLUSION Deep learning models offer a feasible solution to the problem of delineating OARs from CT images. We demonstrate that our proposed model can improve segmentation accuracy and simplify the autosegmentation pipeline. With this method, it is possible to delineate OARs of a head and neck CT within a fraction of a second.
Collapse
Affiliation(s)
- Wentao Zhu
- Department of Computer Science, University of California, Irvine, CA, USA
| | | | | | - Xuming Chen
- Department of Radiation Oncology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
| | - Yong Liu
- Department of Radiation Oncology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
| | - Zhen Qian
- Tencent Medical AI Lab, Palo Alto, CA, USA
| | - Nan Du
- Tencent Medical AI Lab, Palo Alto, CA, USA
| | - Wei Fan
- Tencent Medical AI Lab, Palo Alto, CA, USA
| | - Xiaohui Xie
- Department of Computer Science, University of California, Irvine, CA, USA
| |
Collapse
|
32
|
Men K, Geng H, Cheng C, Zhong H, Huang M, Fan Y, Plastaras JP, Lin A, Xiao Y. Technical Note: More accurate and efficient segmentation of organs-at-risk in radiotherapy with convolutional neural networks cascades. Med Phys 2018; 46:286-292. [PMID: 30450825 DOI: 10.1002/mp.13296] [Citation(s) in RCA: 24] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/03/2018] [Revised: 11/12/2018] [Accepted: 11/13/2018] [Indexed: 01/17/2023] Open
Abstract
PURPOSE Manual delineation of organs-at-risk (OARs) in radiotherapy is both time-consuming and subjective. Automated and more accurate segmentation is of the utmost importance in clinical application. The purpose of this study is to further improve the segmentation accuracy and efficiency with a novel network named convolutional neural networks (CNN) Cascades. METHODS CNN Cascades was a two-step, coarse-to-fine approach that consisted of a simple region detector (SRD) and a fine segmentation unit (FSU). The SRD first used a relative shallow network to define the region of interest (ROI) where the organ was located, and then, the FSU took the smaller ROI as input and adopted a deep network for fine segmentation. The imaging data (14,651 slices) of 100 head-and-neck patients with segmentations were used for this study. The performance was compared with the state-of-the-art single CNN in terms of accuracy with metrics of Dice similarity coefficient (DSC) and Hausdorff distance (HD) values. RESULTS The proposed CNN Cascades outperformed the single CNN on accuracy for each OAR. Similarly, for the average of all OARs, it was also the best with mean DSC of 0.90 (SRD: 0.86, FSU: 0.87, and U-Net: 0.85) and the mean HD of 3.0 mm (SRD: 4.0, FSU: 3.6, and U-Net: 4.4). Meanwhile, the CNN Cascades reduced the mean segmentation time per patient by 48% (FSU) and 5% (U-Net), respectively. CONCLUSIONS The proposed two-step network demonstrated superior performance by reducing the input region. This potentially can be an effective segmentation method that provides accurate and consistent delineation with reduced clinician interventions for clinical applications as well as for quality assurance of a multicenter clinical trial.
Collapse
Affiliation(s)
- Kuo Men
- University of Pennsylvania, Philadelphia, PA, 19104, USA
- National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100021, China
| | - Huaizhi Geng
- University of Pennsylvania, Philadelphia, PA, 19104, USA
| | - Chingyun Cheng
- University of Pennsylvania, Philadelphia, PA, 19104, USA
| | - Haoyu Zhong
- University of Pennsylvania, Philadelphia, PA, 19104, USA
| | - Mi Huang
- University of Pennsylvania, Philadelphia, PA, 19104, USA
| | - Yong Fan
- University of Pennsylvania, Philadelphia, PA, 19104, USA
| | | | - Alexander Lin
- University of Pennsylvania, Philadelphia, PA, 19104, USA
| | - Ying Xiao
- University of Pennsylvania, Philadelphia, PA, 19104, USA
| |
Collapse
|
33
|
Sandström H, Jokura H, Chung C, Toma-Dasu I. Multi-institutional study of the variability in target delineation for six targets commonly treated with radiosurgery. Acta Oncol 2018; 57:1515-1520. [PMID: 29786462 DOI: 10.1080/0284186x.2018.1473636] [Citation(s) in RCA: 13] [Impact Index Per Article: 2.2] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/16/2022]
Abstract
BACKGROUND Although accurate delineation of the target is a key factor of success in radiosurgery there are no consensus guidelines for target contouring. AIM The aim of the present study was therefore to quantify the variability in target delineation and discuss the potential clinical implications, for six targets regarded as common in stereotactic radiosurgery. MATERIAL AND METHODS Twelve Gamma Knife centers participated in the study by contouring the targets and organs at risks and performing the treatment plans. Analysis of target delineation variability was based on metrics defined based on agreement volumes derived from overlapping structures following a previously developed method. The 50% agreement volume (AV50), the common and the encompassing volumes as well as the Agreement Volume Index (AVI) were determined. RESULTS Results showed that the lowest AVI (0.16) was found for one of the analyzed metastases (range of delineated volumes 1.27-3.33 cm3). AVI for the other two metastases was 0.62 and 0.37, respectively. Corresponding AVIs for the cavernous sinus meningioma, pituitary adenoma and vestibular schwannoma were 0.22, 0.37 and 0.50. CONCLUSIONS This study showed that the variability in the contouring was much higher than expected and therefore further work in standardizing the contouring practice in radiosurgery is warranted.
Collapse
Affiliation(s)
- Helena Sandström
- Medical Radiation Physics, Department of Physics, Stockholm University, Stockholm, Sweden
- Medical Radiation Physics, Department of Oncology and Pathology, Karolinska Institutet, Solna, Sweden
| | - Hidefumi Jokura
- Furukawa Seiryo Hospital, Jiro Suzuki Memorial Gamma House, Osaki, Japan
| | - Caroline Chung
- University of Texas MD Anderson Cancer Center, Houston, TX, USA
| | - Iuliana Toma-Dasu
- Medical Radiation Physics, Department of Physics, Stockholm University, Stockholm, Sweden
- Medical Radiation Physics, Department of Oncology and Pathology, Karolinska Institutet, Solna, Sweden
| |
Collapse
|
34
|
Bibault JE, Denis F, Roué A, Gibon D, Fumagalli I, Hennequin C, Barillot I, Quéro L, Paumier A, Mahé MA, Servagi Vernat S, Créhange G, Lapeyre M, Blanchard P, Pointreau Y, Lafond C, Huguet F, Mornex F, Latorzeff I, de Crevoisier R, Martin V, Kreps S, Durdux C, Antoni D, Noël G, Giraud P. [Siriade 2.0: An e-learning platform for radiation oncology contouring]. Cancer Radiother 2018; 22:773-777. [PMID: 30360973 DOI: 10.1016/j.canrad.2018.02.003] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/17/2017] [Revised: 01/23/2018] [Accepted: 02/08/2018] [Indexed: 12/26/2022]
Abstract
PURPOSE In 2008, the French national society of radiation oncology (SFRO) and the association for radiation oncology continued education (AFCOR) created Siriade, an e-learning website dedicated to contouring. MATERIAL AND METHODS Between 2015 and 2017, this platform was updated using the latest digital online tools available. Two main sections were needed: a theoretical part and another section of online workshops. RESULTS Teaching courses are available as online commented videos, available on demand. The practical section of the website is an online contouring workshop that automatically generates a report quantifying the quality of the user's delineation compared with the experts'. CONCLUSION Siriade 2.0 is an innovating digital tool for radiation oncology initial and continuous education.
Collapse
Affiliation(s)
- J-E Bibault
- Service d'oncologie radiothérapie, hôpital européen Georges-Pompidou, 20, rue Leblanc, 75015 Paris, France; Université Paris Descartes, Paris Sorbonne Cité, 20, rue Leblanc, 75015 Paris, France
| | - F Denis
- Service de radiothérapie, centre Jean-Bernard, 9, rue Beauverger, 72000 Le Mans, France
| | - A Roué
- Institut national des sciences et techniques nucléaires, centre CEA de Saclay, D36, 91191 Gif-sur-Yvette, France
| | - D Gibon
- Aquilab, parc Eurasanté, biocentre Fleming, 250, rue Salvador-Allende, 59120 Loos, France
| | - I Fumagalli
- Service d'oncologie radiothérapie, hôpital Saint-Louis, 1, avenue Claude-Vellefau, 75010 Paris, France
| | - C Hennequin
- Service d'oncologie radiothérapie, hôpital Saint-Louis, 1, avenue Claude-Vellefau, 75010 Paris, France
| | - I Barillot
- Service d'oncologie radiothérapie, centre universitaire de cancérologie Henry-S.-Kaplan, 2, boulevard Tonnellé, 37044 Tours, France; Université François-Rabelais, 2, boulevard Tonnellé, 37044 Tours, France
| | - L Quéro
- Service d'oncologie radiothérapie, hôpital Saint-Louis, 1, avenue Claude-Vellefau, 75010 Paris, France
| | - A Paumier
- Service d'oncologie radiothérapie, institut de cancérologie de l'Ouest René-Gauducheau, boulevard Professeur-Jacques-Monod, 44805 Saint-Herblain, France
| | - M-A Mahé
- Service d'oncologie radiothérapie, institut de cancérologie de l'Ouest René-Gauducheau, boulevard Professeur-Jacques-Monod, 44805 Saint-Herblain, France
| | - S Servagi Vernat
- Service d'oncologie radiothérapie, institut Jean-Godinot, 1, rue Koenig, 51100 Reims, France
| | - G Créhange
- Service d'oncologie radiothérapie, centre Georges-François-Leclerc, 1, rue du Professeur-Marion, 21000 Dijon, France
| | - M Lapeyre
- Service d'oncologie radiothérapie, centre Jean-Perrin, 58, rue Montalembert, 63011 Clermont-Ferrand, France
| | - P Blanchard
- Service d'oncologie radiothérapie Gustave-Roussy, 114, rue Édouard-Vaillant, 94805 Villejuif, France
| | - Y Pointreau
- Service de radiothérapie, centre Jean-Bernard, 9, rue Beauverger, 72000 Le Mans, France
| | - C Lafond
- Service de radiothérapie, centre Jean-Bernard, 9, rue Beauverger, 72000 Le Mans, France
| | - F Huguet
- Service d'oncologie radiothérapie, hôpital Tenon, Hôpitaux universitaires de l'Est parisien, 4, rue de la Chine, 75020 Paris, France; Université Pierre-et-Marie-Curie, 4, rue de la Chine, 75020 Paris, France
| | - F Mornex
- Service d'oncologie radiothérapie, CHU Lyon Sud, 65, chemin du Grand-Revoyet, 69495 Pierre-Bénite, France
| | - I Latorzeff
- Service d'oncologie radiothérapie, clinique Pasteur, 1, rue de la Petite-Vitesse, 31300 Toulouse, France
| | - R de Crevoisier
- Service d'oncologie radiothérapie, centre Eugène-Marquis, avenue de la Bataille-Flandre-Dunkerque, 35700 Rennes, France
| | - V Martin
- Service d'oncologie radiothérapie, hôpital Saint-Louis, 1, avenue Claude-Vellefau, 75010 Paris, France
| | - S Kreps
- Service d'oncologie radiothérapie, hôpital européen Georges-Pompidou, 20, rue Leblanc, 75015 Paris, France; Université Paris Descartes, Paris Sorbonne Cité, 20, rue Leblanc, 75015 Paris, France
| | - C Durdux
- Service d'oncologie radiothérapie, hôpital européen Georges-Pompidou, 20, rue Leblanc, 75015 Paris, France; Université Paris Descartes, Paris Sorbonne Cité, 20, rue Leblanc, 75015 Paris, France
| | - D Antoni
- Département universitaire de radiothérapie, centre Paul-Strauss, 3, rue de la Porte-de-l'Hôpital, 67065 Strasbourg, France
| | - G Noël
- Département universitaire de radiothérapie, centre Paul-Strauss, 3, rue de la Porte-de-l'Hôpital, 67065 Strasbourg, France
| | - P Giraud
- Service d'oncologie radiothérapie, hôpital européen Georges-Pompidou, 20, rue Leblanc, 75015 Paris, France; Université Paris Descartes, Paris Sorbonne Cité, 20, rue Leblanc, 75015 Paris, France.
| |
Collapse
|
35
|
Ting FF, Sim KS, Lim CP. Case-control comparison brain lesion segmentation for early infarct detection. Comput Med Imaging Graph 2018; 69:82-95. [PMID: 30219737 DOI: 10.1016/j.compmedimag.2018.08.011] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/18/2017] [Revised: 08/24/2018] [Accepted: 08/24/2018] [Indexed: 11/16/2022]
Abstract
Computed Tomography (CT) images are widely used for the identification of abnormal brain tissues following infarct and hemorrhage of a stroke. The treatment of this medical condition mainly depends on doctors' experience. While manual lesion delineation by medical doctors is currently considered as the standard approach, it is time-consuming and dependent on each doctor's expertise and experience. In this study, a case-control comparison brain lesion segmentation (CCBLS) method is proposed to segment the region pertaining to brain injury by comparing the voxel intensity of CT images between control subjects and stroke patients. The method is able to segment the brain lesion from the stacked CT images automatically without prior knowledge of the location or the presence of the lesion. The aim is to reduce medical doctors' burden and assist them in making an accurate diagnosis. A case study with 300 sets of CT images from control subjects and stroke patients is conducted. Comparing with other existing methods, the outcome ascertains the effectiveness of the proposed method in detecting brain infarct of stroke patients.
Collapse
Affiliation(s)
- Fung Fung Ting
- Faculty of Engineering and Technology, Multimedia University, Jalan Ayer Keroh Lama, 75450, Melaka, Malaysia.
| | - Kok Swee Sim
- Faculty of Engineering and Technology, Multimedia University, Jalan Ayer Keroh Lama, 75450, Melaka, Malaysia.
| | - Chee Peng Lim
- Institute for Intelligent Systems Research and Innovation, Deakin University, Geelong, Victoria, Australia.
| |
Collapse
|
36
|
Kieselmann JP, Kamerling CP, Burgos N, Menten MJ, Fuller CD, Nill S, Cardoso MJ, Oelfke U. Geometric and dosimetric evaluations of atlas-based segmentation methods of MR images in the head and neck region. Phys Med Biol 2018; 63:145007. [PMID: 29882749 PMCID: PMC6296440 DOI: 10.1088/1361-6560/aacb65] [Citation(s) in RCA: 22] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/14/2018] [Revised: 06/01/2018] [Accepted: 06/08/2018] [Indexed: 11/19/2022]
Abstract
Owing to its excellent soft-tissue contrast, magnetic resonance (MR) imaging has found an increased application in radiation therapy (RT). By harnessing these properties for treatment planning, automated segmentation methods can alleviate the manual workload burden to the clinical workflow. We investigated atlas-based segmentation methods of organs at risk (OARs) in the head and neck (H&N) region using one approach that selected the most similar atlas from a library of segmented images and two multi-atlas approaches. The latter were based on weighted majority voting and an iterative atlas-fusion approach called STEPS. We built the atlas library from pre-treatment T1-weighted MR images of 12 patients with manual contours of the parotids, spinal cord and mandible, delineated by a clinician. Following a leave-one-out cross-validation strategy, we measured the geometric accuracy by calculating Dice similarity coefficients (DSC), standard and 95% Hausdorff distances (HD and HD95), and the mean surface distance (MSD), whereby the manual contours served as the gold standard. To benchmark the algorithm, we determined the inter-observer variability (IOV) between three observers. To investigate the dosimetric effect of segmentation inaccuracies, we implemented an auto-planning strategy within the treatment planning system Monaco (Elekta AB, Stockholm, Sweden). For each set of auto-segmented OARs, we generated a plan for a 9-beam step and shoot intensity modulated RT treatment, designed according to our institution's clinical H&N protocol. Superimposing the dose distributions on the gold standard OARs, we calculated dose differences to OARs caused by delineation differences between auto-segmented and gold standard OARs. We investigated the correlations between geometric and dosimetric differences. The mean DSC was larger than 0.8 and the mean MSD smaller than 2 mm for the multi-atlas approaches, resulting in a geometric accuracy comparable to previously published results and within the range of the IOV. While dosimetric differences could be as large as 23% of the clinical goal, treatment plans fulfilled all imposed clinical goals for the gold standard OARs. Correlations between geometric and dosimetric measures were low with R2 < 0.5. The geometric accuracy and the ability to achieve clinically acceptable treatment plans indicate the suitability of using atlas-based contours for RT treatment planning purposes. The low correlations between geometric and dosimetric measures suggest that geometric measures alone are not sufficient to predict the dosimetric impact of segmentation inaccuracies on treatment planning for the data utilised in this study.
Collapse
Affiliation(s)
- J P Kieselmann
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| | - C P Kamerling
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| | - N Burgos
- University
College London, Centre for Medical Image Computing, London,
United Kingdom
- Inria, Aramis project-team, Institut du Cerveau et de la Moelle
épinière, Sorbonne Université, Paris,
France
| | - M J Menten
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| | - C D Fuller
- Department of Radiation Oncology,
MD Anderson Cancer Center,
Houston, TX, United States of America
| | - S Nill
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| | - M J Cardoso
- University
College London, Centre for Medical Image Computing, London,
United Kingdom
- School of
Biomedical Engineering and Imaging Sciences, King’s College,
London, United Kingdom
| | - U Oelfke
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| |
Collapse
|
37
|
Kieselmann JP, Kamerling CP, Burgos N, Menten MJ, Fuller CD, Nill S, Cardoso MJ, Oelfke U. Geometric and dosimetric evaluations of atlas-based segmentation methods of MR images in the head and neck region. Phys Med Biol 2018; 63:145007. [PMID: 29882749 PMCID: PMC6296440 DOI: 10.1088/1361-6560/aacb65;145007] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/21/2023]
Abstract
Owing to its excellent soft-tissue contrast, magnetic resonance (MR) imaging has found an increased application in radiation therapy (RT). By harnessing these properties for treatment planning, automated segmentation methods can alleviate the manual workload burden to the clinical workflow. We investigated atlas-based segmentation methods of organs at risk (OARs) in the head and neck (H&N) region using one approach that selected the most similar atlas from a library of segmented images and two multi-atlas approaches. The latter were based on weighted majority voting and an iterative atlas-fusion approach called STEPS. We built the atlas library from pre-treatment T1-weighted MR images of 12 patients with manual contours of the parotids, spinal cord and mandible, delineated by a clinician. Following a leave-one-out cross-validation strategy, we measured the geometric accuracy by calculating Dice similarity coefficients (DSC), standard and 95% Hausdorff distances (HD and HD95), and the mean surface distance (MSD), whereby the manual contours served as the gold standard. To benchmark the algorithm, we determined the inter-observer variability (IOV) between three observers. To investigate the dosimetric effect of segmentation inaccuracies, we implemented an auto-planning strategy within the treatment planning system Monaco (Elekta AB, Stockholm, Sweden). For each set of auto-segmented OARs, we generated a plan for a 9-beam step and shoot intensity modulated RT treatment, designed according to our institution's clinical H&N protocol. Superimposing the dose distributions on the gold standard OARs, we calculated dose differences to OARs caused by delineation differences between auto-segmented and gold standard OARs. We investigated the correlations between geometric and dosimetric differences. The mean DSC was larger than 0.8 and the mean MSD smaller than 2 mm for the multi-atlas approaches, resulting in a geometric accuracy comparable to previously published results and within the range of the IOV. While dosimetric differences could be as large as 23% of the clinical goal, treatment plans fulfilled all imposed clinical goals for the gold standard OARs. Correlations between geometric and dosimetric measures were low with R2 < 0.5. The geometric accuracy and the ability to achieve clinically acceptable treatment plans indicate the suitability of using atlas-based contours for RT treatment planning purposes. The low correlations between geometric and dosimetric measures suggest that geometric measures alone are not sufficient to predict the dosimetric impact of segmentation inaccuracies on treatment planning for the data utilised in this study.
Collapse
Affiliation(s)
- J P Kieselmann
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom,
| | - C P Kamerling
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| | - N Burgos
- University
College London, Centre for Medical Image Computing, London,
United Kingdom,Inria, Aramis project-team, Institut du Cerveau et de la Moelle
épinière, Sorbonne Université, Paris,
France
| | - M J Menten
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| | - C D Fuller
- Department of Radiation Oncology,
MD Anderson Cancer Center,
Houston, TX, United States of America
| | - S Nill
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| | - M J Cardoso
- University
College London, Centre for Medical Image Computing, London,
United Kingdom,School of
Biomedical Engineering and Imaging Sciences, King’s College,
London, United Kingdom
| | - U Oelfke
- Joint Department of Physics, The Institute of Cancer Research and The Royal Marsden
NHS Foundation Trust, London, United
Kingdom
| |
Collapse
|
38
|
Segmentation of parotid glands from registered CT and MR images. Phys Med 2018; 52:33-41. [PMID: 30139607 DOI: 10.1016/j.ejmp.2018.06.012] [Citation(s) in RCA: 24] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/19/2018] [Revised: 06/11/2018] [Accepted: 06/12/2018] [Indexed: 01/16/2023] Open
Abstract
PURPOSE To develop an automatic multimodal method for segmentation of parotid glands (PGs) from pre-registered computed tomography (CT) and magnetic resonance (MR) images and compare its results to the results of an existing state-of-the-art algorithm that segments PGs from CT images only. METHODS Magnetic resonance images of head and neck were registered to the accompanying CT images using two different state-of-the-art registration procedures. The reference domains of registered image pairs were divided on the complementary PG regions and backgrounds according to the manual delineation of PGs on CT images, provided by a physician. Patches of intensity values from both image modalities, centered around randomly sampled voxels from the reference domain, served as positive or negative samples in the training of the convolutional neural network (CNN) classifier. The trained CNN accepted a previously unseen (registered) image pair and classified its voxels according to the resemblance of its patches to the patches used for training. The final segmentation was refined using a graph-cut algorithm, followed by the dilate-erode operations. RESULTS Using the same image dataset, segmentation of PGs was performed using the proposed multimodal algorithm and an existing monomodal algorithm, which segments PGs from CT images only. The mean value of the achieved Dice overlapping coefficient for the proposed algorithm was 78.8%, while the corresponding mean value for the monomodal algorithm was 76.5%. CONCLUSIONS Automatic PG segmentation on the planning CT image can be augmented with the MR image modality, leading to an improved RT planning of head and neck cancer.
Collapse
|
39
|
Wu X, Udupa JK, Tong Y, Odhner D, Pednekar GV, Simone CB, McLaughlin D, Apinorasethkul C, Lukens J, Mihailidis D, Shammo G, James P, Camaratta J, Torigian DA. Auto-contouring via Automatic Anatomy Recognition of Organs at Risk in Head and Neck Cancer on CT images. PROCEEDINGS OF SPIE--THE INTERNATIONAL SOCIETY FOR OPTICAL ENGINEERING 2018; 10576:1057617. [PMID: 30190629 PMCID: PMC6122857 DOI: 10.1117/12.2293946] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/14/2022]
Abstract
Contouring of the organs at risk is a vital part of routine radiation therapy planning. For the head and neck (H&N) region, this is more challenging due to the complexity of anatomy, the presence of streak artifacts, and the variations of object appearance. In this paper, we describe the latest advances in our Automatic Anatomy Recognition (AAR) approach, which aims to automatically contour multiple objects in the head and neck region on planning CT images. Our method has three major steps: model building, object recognition, and object delineation. First, the better-quality images from our cohort of H&N CT studies are used to build fuzzy models and find the optimal hierarchy for arranging objects based on the relationship between objects. Then, the object recognition step exploits the rich prior anatomic information encoded in the hierarchy to derive the location and pose for each object, which leads to generalizable and robust methods and mitigation of object localization challenges. Finally, the delineation algorithms employ local features to contour the boundary based on object recognition results. We make several improvements within the AAR framework, including finding recognition-error-driven optimal hierarchy, modeling boundary relationships, combining texture and intensity, and evaluating object quality. Experiments were conducted on the largest ensemble of clinical data sets reported to date, including 216 planning CT studies and over 2,600 object samples. The preliminary results show that on data sets with minimal (<4 slices) streak artifacts and other deviations, overall recognition accuracy reaches 2 voxels, with overall delineation Dice coefficient close to 0.8 and Hausdorff Distance within 1 voxel.
Collapse
Affiliation(s)
- Xingyu Wu
- Medical Image Processing Group, 3710 Hamilton Walk, Department of Radiology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Jayaram K. Udupa
- Medical Image Processing Group, 3710 Hamilton Walk, Department of Radiology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Yubing Tong
- Medical Image Processing Group, 3710 Hamilton Walk, Department of Radiology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Dewey Odhner
- Medical Image Processing Group, 3710 Hamilton Walk, Department of Radiology, University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Gargi V. Pednekar
- Quantitative Radiology Solutions, 3624 Market Street, Suite 5E, Philadelphia, PA 19104, United States
| | - Charles B. Simone
- University of Maryland School of Medicine, Department of Radiation Oncology, Maryland Proton Treatment Center, 850 W. Baltimore, MD 21201, United States
| | - David McLaughlin
- Quantitative Radiology Solutions, 3624 Market Street, Suite 5E, Philadelphia, PA 19104, United States
| | - Chavanon Apinorasethkul
- Radiation Oncology Department at University of Pennsylvania, Philadelphia, PA 19104, United States
| | - John Lukens
- Radiation Oncology Department at University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Dimitris Mihailidis
- Radiation Oncology Department at University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Geraldine Shammo
- Radiation Oncology Department at University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Paul James
- Radiation Oncology Department at University of Pennsylvania, Philadelphia, PA 19104, United States
| | - Joseph Camaratta
- Quantitative Radiology Solutions, 3624 Market Street, Suite 5E, Philadelphia, PA 19104, United States
| | - Drew A. Torigian
- Medical Image Processing Group, 3710 Hamilton Walk, Department of Radiology, University of Pennsylvania, Philadelphia, PA 19104, United States
| |
Collapse
|
40
|
Meillan N, Bibault JE, Vautier J, Daveau-Bergerault C, Kreps S, Tournat H, Durdux C, Giraud P. Automatic Intracranial Segmentation: Is the Clinician Still Needed? Technol Cancer Res Treat 2018; 17:1533034617748839. [PMID: 29343204 PMCID: PMC5784565 DOI: 10.1177/1533034617748839] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/28/2017] [Revised: 10/23/2017] [Accepted: 11/17/2017] [Indexed: 12/31/2022] Open
Abstract
INTRODUCTION Stereotactic hypofractionated radiotherapy is an effective treatment for brain metastases in oligometastatic patients. Its planning is however time-consuming because of the number of organs at risk to be manually segmented. This study evaluates 2 automated segmentation commercial software. METHODS Patients were scanned in the treatment position. The computed tomography scan was registered on a magnetic resonance imaging and volumes were manually segmented by a clinician. Then 2 automated segmentations were performed (with iPlan and Smart Segmentation). RT STRUCT files were compared with Aquilab's Artistruct segment comparison module. We selected common segmented volume ratio as the main judging criterion. Secondary criteria were Dice-Sørensen coefficients, overlap ratio, and additional segmented volume. RESULTS Twenty consecutive patients were included. Agreement between manual and automated contouring was poor. Common segmented volumes ranged from 7.71% to 82.54%, Dice-Sørensen coefficient ranged from 0.0745 to 0.8398, overlap ratio ranged from 0.0414 to 0.7275, and additional segmented volume ranged from 9.80% to 92.25%. Each software outperformed the other on some organs while performing worse on others. CONCLUSION No software seemed clearly better than the other. Common segmented volumes were much too low for routine use in stereotactic hypofractionated brain radiotherapy. Manual editing is still needed.
Collapse
Affiliation(s)
- Nicolas Meillan
- Service de Cancérologie Radiothérapie, Hopital Saint-Louis, Paris, France
| | | | - Julien Vautier
- Service d’Onco-Radiothérapie, Hopital Europeen Georges Pompidou, Paris, France
| | | | - Sarah Kreps
- Service d’Onco-Radiothérapie, Hopital Europeen Georges Pompidou, Paris, France
| | - Hélène Tournat
- Service d’Onco-Radiothérapie, Hopital Europeen Georges Pompidou, Paris, France
| | - Catherine Durdux
- Service d’Onco-Radiothérapie, Hopital Europeen Georges Pompidou, Paris, France
| | - Philippe Giraud
- Service d’Onco-Radiothérapie, Hopital Europeen Georges Pompidou, Paris, France
| |
Collapse
|
41
|
Clinical evaluation of atlas and deep learning based automatic contouring for lung cancer. Radiother Oncol 2017; 126:312-317. [PMID: 29208513 DOI: 10.1016/j.radonc.2017.11.012] [Citation(s) in RCA: 219] [Impact Index Per Article: 31.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/26/2017] [Revised: 11/16/2017] [Accepted: 11/21/2017] [Indexed: 02/07/2023]
Abstract
BACKGROUND AND PURPOSE Contouring of organs at risk (OARs) is an important but time consuming part of radiotherapy treatment planning. The aim of this study was to investigate whether using institutional created software-generated contouring will save time if used as a starting point for manual OAR contouring for lung cancer patients. MATERIAL AND METHODS Twenty CT scans of stage I-III NSCLC patients were used to compare user adjusted contours after an atlas-based and deep learning contour, against manual delineation. The lungs, esophagus, spinal cord, heart and mediastinum were contoured for this study. The time to perform the manual tasks was recorded. RESULTS With a median time of 20 min for manual contouring, the total median time saved was 7.8 min when using atlas-based contouring and 10 min for deep learning contouring. Both atlas based and deep learning adjustment times were significantly lower than manual contouring time for all OARs except for the left lung and esophagus of the atlas based contouring. CONCLUSIONS User adjustment of software generated contours is a viable strategy to reduce contouring time of OARs for lung radiotherapy while conforming to local clinical standards. In addition, deep learning contouring shows promising results compared to existing solutions.
Collapse
|
42
|
Rivin Del Campo E, Rivera S, Martínez-Paredes M, Hupé P, Slocker Escarpa A, Borget I, Mazeron R, Scholl S, Palacios Eito A, Haie-Meder C, Chargari C, Deutsch E. Assessment of the novel online delineation workshop dummy run approach using FALCON within a European multicentre trial in cervical cancer (RAIDs). Radiother Oncol 2017; 124:130-138. [PMID: 28532608 DOI: 10.1016/j.radonc.2017.05.008] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/29/2017] [Revised: 04/10/2017] [Accepted: 05/04/2017] [Indexed: 11/28/2022]
Abstract
BACKGROUND AND PURPOSE Online delineation workshops (ODW) permit training of geographically dispersed participants. The purpose is to evaluate the methodology of an ODW using FALCON to harmonize delineation within a European multicentre trial on locally advanced cervical cancer (LACC). MATERIAL AND METHODS Two ODW included 46 clinicians (14 centres). Clinicians completed baseline (C1), guideline (C2) and final contours (C3) for external beam radiotherapy (EBRT) and brachytherapy (BT) for LACC. Interobserver and intraobserver variability was evaluated quantitatively (using the DICE index) and qualitatively compared to expert contours. RESULTS Nine clinicians submitted for EBRT and BT for C1-C3. Thirty-two sent any contour. Interobserver quantitative comparisons for EBRT showed significant improvement for C2 vs. C1 for bowel, CTV node, CTV-p and GTV node with significant detriment for GTV node (C3 vs. C1; C2), CTV-p (C3 vs. C2) and bowel (C3 vs. C2), showing in general an improvement in C2 vs. C1, with a detriment in C3 vs. C2 for two target volumes and an organ at risk. For BT there was significant improvement for C2 vs. C1 for bladder, GTV, HR-CTV and IR-CTV, with significant detriment for bladder (C3 vs. C2), thus overall improvement in C2 vs. C1, with only a detriment in C3 vs. C2 for bladder. Centres using MRI imaging for BT contouring did significantly better in the BT case for HR-CTV than those which used other techniques (C2 vs. C1: p<0.005; C3 vs. C1: p=0.02). Intraobserver quantitative comparisons showed significant improvement contouring a region of interest between C2 vs. C1, C3 vs. C1 and C3 vs. C2 for EBRT and between C2 and C1 for BT. CONCLUSIONS ODW offer training, initial contouring harmonization and allow assessment of centres.
Collapse
Affiliation(s)
| | - Sofia Rivera
- Department of Radiation Oncology, Gustave Roussy Cancer Campus, Villejuif, France
| | - María Martínez-Paredes
- Department of Radiology and Medical Physics, Medical School, University of Cordoba, Spain
| | - Philippe Hupé
- Institut Curie, Paris, France; PSL Research University, Paris, France; Inserm, U900, Paris, France; Mines Paris Tech, Fontainebleau, France; CNRS, UMR144, Paris, France
| | | | - Isabelle Borget
- Service de Biostatistique et d'Epidemiologie, Gustave Roussy, Villejuif, France; INSERM U1018, CESP, Université Paris-Sud, Université Paris-Saclay, Villejuif, France
| | - Renaud Mazeron
- Department of Radiation Oncology, Gustave Roussy Cancer Campus, Villejuif, France
| | - Suzy Scholl
- Department of Medical Oncology, Institut Curie, Paris, France
| | - Amalia Palacios Eito
- Department of Radiation Oncology, Reina Sofia University Hospital, Cordoba, Spain
| | - Christine Haie-Meder
- Department of Radiation Oncology, Gustave Roussy Cancer Campus, Villejuif, France
| | - Cyrus Chargari
- Department of Radiation Oncology, Gustave Roussy Cancer Campus, Villejuif, France
| | - Eric Deutsch
- Department of Radiation Oncology, Gustave Roussy Cancer Campus, Villejuif, France
| |
Collapse
|
43
|
Fogliata A, Reggiori G, Stravato A, Lobefalo F, Franzese C, Franceschini D, Tomatis S, Mancosu P, Scorsetti M, Cozzi L. RapidPlan head and neck model: the objectives and possible clinical benefit. Radiat Oncol 2017; 12:73. [PMID: 28449704 PMCID: PMC5408433 DOI: 10.1186/s13014-017-0808-x] [Citation(s) in RCA: 62] [Impact Index Per Article: 8.9] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/06/2017] [Accepted: 04/14/2017] [Indexed: 11/27/2022] Open
Abstract
Background To evaluate a knowledge based planning model for RapidPlan (RP) generated for advanced head and neck cancer (HNC) patient treatments, as well its ability to possibly improve the clinical plan quality. The stability of the model was assessed also for a different beam geometry, different dose fractionation and different management of bilateral structures (parotids). Methods Dosimetric and geometric data from plans of 83 patients presenting HNC were selected for the model training. All the plans used volumetric modulated arc therapy (VMAT, RapidArc) to treat two targets at dose levels of 69.96 and 54.45 Gy in 33 fractions with simultaneous integrated boost. Two models were generated, the first separating the ipsi- and contra-lateral parotids, while the second associating the two parotids to a single structure for training. The optimization objectives were adjusted to the final model to better translate the institutional planning and dosimetric strategies and trade-offs. The models were validated on 20 HNC patients, comparing the RP generated plans and the clinical plans. RP generated plans were also compared between the clinical beam arrangement and a simpler geometry, as well as for a different fractionation scheme. Results RP improved significantly the clinical plan quality, with a reduction of 2 Gy, 5 Gy, and 10 Gy of the mean parotid, oral cavity and laryngeal doses, respectively. A simpler beam geometry was deteriorating the plan quality, but in a small amount, keeping a significant improvement relative to the clinical plan. The two models, with one or two parotid structures, showed very similar results. NTCP evaluations indicated the possibility of improving (NTCP decreasing of about 7%) the toxicity profile when using the RP solution. Conclusions The HNC RP model showed improved plan quality and planning stability for beam geometry and fractionation. An adequate choice of the objectives in the model is necessary for the trade-offs strategies.
Collapse
Affiliation(s)
- A Fogliata
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy.
| | - G Reggiori
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy
| | - A Stravato
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy
| | - F Lobefalo
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy
| | - C Franzese
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy
| | - D Franceschini
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy
| | - S Tomatis
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy
| | - P Mancosu
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy
| | - M Scorsetti
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy.,Department of Biomedical Sciences, Humanitas University, Milan, Rozzano, Italy
| | - L Cozzi
- Humanitas Research Hospital and Cancer Center, Radiotherapy and Radiosurgery Department, Milan, Rozzano, Italy.,Department of Biomedical Sciences, Humanitas University, Milan, Rozzano, Italy
| |
Collapse
|
44
|
Ibragimov B, Xing L. Segmentation of organs-at-risks in head and neck CT images using convolutional neural networks. Med Phys 2017; 44:547-557. [PMID: 28205307 DOI: 10.1002/mp.12045] [Citation(s) in RCA: 320] [Impact Index Per Article: 45.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/02/2016] [Revised: 10/31/2016] [Accepted: 11/23/2016] [Indexed: 12/14/2022] Open
Abstract
PURPOSE Accurate segmentation of organs-at-risks (OARs) is the key step for efficient planning of radiation therapy for head and neck (HaN) cancer treatment. In the work, we proposed the first deep learning-based algorithm, for segmentation of OARs in HaN CT images, and compared its performance against state-of-the-art automated segmentation algorithms, commercial software, and interobserver variability. METHODS Convolutional neural networks (CNNs)-a concept from the field of deep learning-were used to study consistent intensity patterns of OARs from training CT images and to segment the OAR in a previously unseen test CT image. For CNN training, we extracted a representative number of positive intensity patches around voxels that belong to the OAR of interest in training CT images, and negative intensity patches around voxels that belong to the surrounding structures. These patches then passed through a sequence of CNN layers that captured local image features such as corners, end-points, and edges, and combined them into more complex high-order features that can efficiently describe the OAR. The trained network was applied to classify voxels in a region of interest in the test image where the corresponding OAR is expected to be located. We then smoothed the obtained classification results by using Markov random fields algorithm. We finally extracted the largest connected component of the smoothed voxels classified as the OAR by CNN, performed dilate-erode operations to remove cavities of the component, which resulted in segmentation of the OAR in the test image. RESULTS The performance of CNNs was validated on segmentation of spinal cord, mandible, parotid glands, submandibular glands, larynx, pharynx, eye globes, optic nerves, and optic chiasm using 50 CT images. The obtained segmentation results varied from 37.4% Dice coefficient (DSC) for chiasm to 89.5% DSC for mandible. We also analyzed the performance of state-of-the-art algorithms and commercial software reported in the literature, and observed that CNNs demonstrate similar or superior performance on segmentation of spinal cord, mandible, parotid glands, larynx, pharynx, eye globes, and optic nerves, but inferior performance on segmentation of submandibular glands and optic chiasm. CONCLUSION We concluded that convolution neural networks can accurately segment most of OARs using a representative database of 50 HaN CT images. At the same time, inclusion of additional information, for example, MR images, may be beneficial to some OARs with poorly visible boundaries.
Collapse
Affiliation(s)
- Bulat Ibragimov
- Department of Radiation Oncology, Stanford University School of Medicine, Stanford, California, 94305, USA
| | - Lei Xing
- Department of Radiation Oncology, Stanford University School of Medicine, Stanford, California, 94305, USA
| |
Collapse
|