1
|
Jermain PR, Muir B, McEwen M, Niu Y, Pang D. Accurate machine-specific reference and small-field dosimetry for a self-shielded neuro-radiosurgical system. Med Phys 2024; 51:4423-4433. [PMID: 38695760 DOI: 10.1002/mp.17111] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/05/2024] [Revised: 03/14/2024] [Accepted: 04/18/2024] [Indexed: 06/05/2024] Open
Abstract
BACKGROUND The newly available ZAP-X stereotactic radiosurgical system is designed for the treatment of intracranial lesions, with several unique features that include a self-shielding, gyroscopic gantry, wheel collimation, non-orthogonal kV imaging, short source-axis distance, and low-energy megavoltage beam. Systematic characterization of its radiation as well as other properties is imperative to ensure its safe and effective clinical application. PURPOSE To accurately determine the radiation output of the ZAP-X with a special focus on the smaller diameter cones and an aim to provide useful recommendations on quantification of small field dosimetry. METHODS Six different types of detectors were used to measure relative output factors at field sizes ranging from 4 to 25 mm, including the PTW microSilicon and microdiamond diodes, Exradin W2 plastic scintillator, Exradin A16 and A1SL ionization chambers, and the alanine dosimeter. The 25 mm cone served as the reference field size. Absolute dose was determined with both TG-51-based dosimetry using a calibrated PTW Semiflex ion chamber and measurements using alanine dosimeters. RESULTS The average radiation output factors (maximum deviation from the average) measured with the microDiamond, microSilicon, and W2 detectors were: for the 4 mm cone, 0.741 (1.0%); for the 5 mm cone: 0.817 (1.0%); for the 7.5 mm cone: 0.908 (1.0%); for the 10 mm cone: 0.946 (0.4%); for the 12.5 mm cone: 0.964 (0.2%); for the 15 mm cone: 0.976 (0.1%); for the 20 mm cone: 0.990 (0.1%). For field sizes larger than 10 mm, the A1SL and A16 micro-chambers also yielded consistent output factors within 1.5% of those obtained using the microSilicon, microdiamond, and W2 detectors. The absolute dose measurement obtained with alanine was within 1.2%, consistent with combined uncertainties, compared to the PTW Semiflex chamber for the 25 mm reference cone. CONCLUSION For field sizes less than 10 mm, the microSilicon diode, microDiamond detector, and W2 scintillator are suitable devices for accurate small field dosimetry of the ZAP-X system. For larger fields, the A1SL and A16 micro-chambers can also be used. Furthermore, alanine dosimetry can be an accurate verification of reference and absolute dose typically measured with ion chambers. Use of multiple suitable detectors and uncertainty analyses were recommended for reliable determination of small field radiation outputs.
Collapse
Affiliation(s)
- Peter R Jermain
- Department of Radiation Medicine, Medstar Georgetown University Hospital, Washington, District of Columbia, USA
| | - Bryan Muir
- Metrology Research Centre, National Research Council, Ottawa, Ontario, Canada
| | - Malcolm McEwen
- Metrology Research Centre, National Research Council, Ottawa, Ontario, Canada
| | - Ying Niu
- Department of Radiation Medicine, Medstar Georgetown University Hospital, Washington, District of Columbia, USA
| | - Dalong Pang
- Department of Radiation Medicine, Medstar Georgetown University Hospital, Washington, District of Columbia, USA
| |
Collapse
|
2
|
Zhao J, Vaios E, Wang Y, Yang Z, Cui Y, Reitman ZJ, Lafata KJ, Fecci P, Kirkpatrick J, Fang Yin F, Floyd S, Wang C. Dose-Incorporated Deep Ensemble Learning for Improving Brain Metastasis Stereotactic Radiosurgery Outcome Prediction. Int J Radiat Oncol Biol Phys 2024:S0360-3016(24)00505-4. [PMID: 38615888 DOI: 10.1016/j.ijrobp.2024.04.006] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/29/2023] [Revised: 03/19/2024] [Accepted: 04/02/2024] [Indexed: 04/16/2024]
Abstract
PURPOSE To develop a novel deep ensemble learning model for accurate prediction of brain metastasis (BM) local control outcomes after stereotactic radiosurgery (SRS). METHODS AND MATERIALS A total of 114 brain metastases (BMs) from 82 patients were evaluated, including 26 BMs that developed biopsy-confirmed local failure post-SRS. The SRS spatial dose distribution (Dmap) of each BM was registered to the planning contrast-enhanced T1 (T1-CE) magnetic resonance imaging (MRI). Axial slices of the Dmap, T1-CE, and planning target volume (PTV) segmentation (PTVseg) intersecting the BM center were extracted within a fixed field of view determined by the 60% isodose volume in Dmap. A spherical projection was implemented to transform planar image content onto a spherical surface using multiple projection centers, and the resultant T1-CE/Dmap/PTVseg projections were stacked as a 3-channel variable. Four Visual Geometry Group (VGG-19) deep encoders were used in an ensemble design, with each submodel using a different spherical projection formula as input for BM outcome prediction. In each submodel, clinical features after positional encoding were fused with VGG-19 deep features to generate logit results. The ensemble's outcome was synthesized from the 4 submodel results via logistic regression. In total, 10 model versions with random validation sample assignments were trained to study model robustness. Performance was compared with (1) a single VGG-19 encoder, (2) an ensemble with a T1-CE MRI as the sole image input after projections, and (3) an ensemble with the same image input design without clinical feature inclusion. RESULTS The ensemble model achieved an excellent area under the receiver operating characteristic curve (AUCROC: 0.89 ± 0.02) with high sensitivity (0.82 ± 0.05), specificity (0.84 ± 0.11), and accuracy (0.84 ± 0.08) results. This outperformed the MRI-only VGG-19 encoder (sensitivity: 0.35 ± 0.01, AUCROC: 0.64 ± 0.08), the MRI-only deep ensemble (sensitivity: 0.60 ± 0.09, AUCROC: 0.68 ± 0.06), and the 3-channel ensemble without clinical feature fusion (sensitivity: 0.78 ± 0.08, AUCROC: 0.84 ± 0.03). CONCLUSIONS Facilitated by the spherical image projection method, a deep ensemble model incorporating Dmap and clinical variables demonstrated excellent performance in predicting BM post-SRS local failure. Our novel approach could improve other radiation therapy outcome models and warrants further evaluation.
Collapse
Affiliation(s)
- Jingtong Zhao
- Duke University Medical Center, Durham, North Carolina
| | - Eugene Vaios
- Duke University Medical Center, Durham, North Carolina
| | - Yuqi Wang
- Duke University Medical Center, Durham, North Carolina
| | - Zhenyu Yang
- Duke University Medical Center, Durham, North Carolina
| | - Yunfeng Cui
- Duke University Medical Center, Durham, North Carolina
| | | | - Kyle J Lafata
- Duke University Medical Center, Durham, North Carolina
| | - Peter Fecci
- Duke University Medical Center, Durham, North Carolina
| | | | | | - Scott Floyd
- Duke University Medical Center, Durham, North Carolina
| | - Chunhao Wang
- Duke University Medical Center, Durham, North Carolina.
| |
Collapse
|
3
|
Jeong H, Park JE, Kim N, Yoon SK, Kim HS. Deep learning-based detection and quantification of brain metastases on black-blood imaging can provide treatment suggestions: a clinical cohort study. Eur Radiol 2024; 34:2062-2071. [PMID: 37658885 PMCID: PMC10873231 DOI: 10.1007/s00330-023-10120-5] [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: 01/17/2023] [Revised: 06/25/2023] [Accepted: 07/01/2023] [Indexed: 09/05/2023]
Abstract
OBJECTIVES We aimed to evaluate whether deep learning-based detection and quantification of brain metastasis (BM) may suggest treatment options for patients with BMs. METHODS The deep learning system (DLS) for detection and quantification of BM was developed in 193 patients and applied to 112 patients that were newly detected on black-blood contrast-enhanced T1-weighted imaging. Patients were assigned to one of 3 treatment suggestion groups according to the European Association of Neuro-Oncology (EANO)-European Society for Medical Oncology (ESMO) recommendations using number and volume of the BMs detected by the DLS: short-term imaging follow-up without treatment (group A), surgery or stereotactic radiosurgery (limited BM, group B), or whole-brain radiotherapy or systemic chemotherapy (extensive BM, group C). The concordance between the DLS-based groups and clinical decisions was analyzed with or without consideration of targeted agents. The performance of distinguishing high-risk (B + C) was calculated. RESULTS Among 112 patients (mean age 64.3 years, 63 men), group C had the largest number and volume of BM, followed by group B (4.4 and 851.6 mm3) and A (1.5 and 15.5 mm3). The DLS-based groups were concordant with the actual clinical decisions, with an accuracy of 76.8% (86 of 112). Modified accuracy considering targeted agents was 81.3% (91 of 112). The DLS showed 95% (82/86) sensitivity and 81% (21/26) specificity for distinguishing the high risk. CONCLUSION DLS-based detection and quantification of BM have the potential to be helpful in the determination of treatment options for both low- and high-risk groups of limited and extensive BMs. CLINICAL RELEVANCE STATEMENT For patients with newly diagnosed brain metastasis, deep learning-based detection and quantification may be used in clinical settings where prompt and accurate treatment decisions are required, which can lead to better patient outcomes. KEY POINTS • Deep learning-based brain metastasis detection and quantification showed excellent agreement with ground-truth classifications. • By setting an algorithm to suggest treatment based on the number and volume of brain metastases detected by the deep learning system, the concordance was 81.3%. • When dividing patients into low- and high-risk groups, the sensitivity for detecting the latter was 95%.
Collapse
Affiliation(s)
- Hana Jeong
- Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 43 Olympic-ro 88, Songpa-Gu, 05505, Seoul, Korea
| | - Ji Eun Park
- Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 43 Olympic-ro 88, Songpa-Gu, 05505, Seoul, Korea.
| | | | - Shin-Kyo Yoon
- Department of Oncology, Asan Medical Center, Seoul, South Korea
| | - Ho Sung Kim
- Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 43 Olympic-ro 88, Songpa-Gu, 05505, Seoul, Korea
| |
Collapse
|
4
|
Kübler J, Wester-Ebbinghaus M, Wenz F, Stieler F, Bathen B, Mai SK, Wolff R, Hänggi D, Blanck O, Giordano FA. Postoperative stereotactic radiosurgery and hypofractionated radiotherapy for brain metastases using Gamma Knife and CyberKnife: a dual-center analysis. J Neurosurg Sci 2024; 68:22-30. [PMID: 32031357 DOI: 10.23736/s0390-5616.20.04830-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/02/2024]
Abstract
BACKGROUND Postoperative stereotactic radiosurgery (SRS) and hypofractionated stereotactic radiotherapy (hFSRT) to tumor cavities is emerging as a new standard of care after resection of brain metastases. Both Gamma Knife (GK) and CyberKnife (CK) are modalities commonly used for stereotactic radiotherapy, but fractional schemes are not consistent. The objective of this study was to evaluate outcomes in patients receiving postoperative stereotactic radiotherapy of resected brain metastases (BM) using different fractionation schedules and modalities in two large centers. METHODS Patients with newly diagnosed BM who underwent postoperative SRS or hFSRT with either GK or CK at two large cancer centers were retrospectively evaluated. We analyzed local control (LC), regional control (RC) and overall survival (OS). RESULTS From April 14th to May 18th, 2020, 79 patients with 81 resection cavities were treated. Forty-seven patients (59.5%) received GK and 32 patients (40.5%) received CK treatment. Fifty-four cavities (66.7%) were treated with hFSRT and 27 (33.3%) with SRS. The most common hFSRT and SRS scheme was 3x10 Gy and 1x16 Gy, respectively. Median OS was 11.7 months with survival rates of 44.7% at 1 year and 18.5% at 2 years. LC was 83.3% after 1 year. Median time to regional progression was 12.0 months with RC rates of 61.1% at 6 months and 41.0% at 12 months. There was no difference in OS, LC or RC between GK and CK treatments or SRS and hFSRT. CONCLUSIONS Both SRS and hFSRT provide high local control rates in resected BM regardless of the applied modality.
Collapse
Affiliation(s)
- Jens Kübler
- Department of Radiation Oncology, University Hospital Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
| | - Michael Wester-Ebbinghaus
- Department of Radiation Oncology, University Hospital Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
| | | | - Florian Stieler
- Department of Radiation Oncology, University Hospital Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
| | - Bastian Bathen
- Saphir Radiosurgery Center Frankfurt, Frankfurt am Main, Germany
- Department of Radiation Oncology, University Hospital Frankfurt, Frankfurt am Main, Germany
| | - Sabine K Mai
- Department of Radiation Oncology, University Hospital Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
| | - Robert Wolff
- Saphir Radiosurgery Center Frankfurt, Frankfurt am Main, Germany
- Department of Neurosurgery, University Hospital Frankfurt, Frankfurt am Main, Germany
| | - Daniel Hänggi
- Department of Neurosurgery, University Hospital Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
| | - Oliver Blanck
- Saphir Radiosurgery Center Frankfurt, Frankfurt am Main, Germany
- Department of Radiation Oncology, University Medical Center Schleswig-Holstein, Kiel, Germany
| | - Frank A Giordano
- Department of Radiation Oncology, University Hospital Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany -
| |
Collapse
|
5
|
Dasgupta A, Saifuddin M, McNabb E, Ho L, Lu L, Vesprini D, Karam I, Soliman H, Chow E, Gandhi S, Trudeau M, Tran W, Curpen B, Stanisz G, Sahgal A, Kolios M, Czarnota GJ. Novel MRI-guided focussed ultrasound stimulated microbubble radiation enhancement treatment for breast cancer. Sci Rep 2023; 13:13566. [PMID: 37604988 PMCID: PMC10442356 DOI: 10.1038/s41598-023-40551-5] [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: 02/20/2023] [Accepted: 08/12/2023] [Indexed: 08/23/2023] Open
Abstract
Preclinical studies have demonstrated focused ultrasound (FUS) stimulated microbubble (MB) rupture leads to the activation of acid sphingomyelinase-ceramide pathway in the endothelial cells. When radiotherapy (RT) is delivered concurrently with FUS-MB, apoptotic pathway leads to increased cell death resulting in potent radiosensitization. Here we report the first human trial of using magnetic resonance imaging (MRI) guided FUS-MB treatment in the treatment of breast malignancies. In the phase 1 prospective interventional study, patients with breast cancer were treated with fractionated RT (5 or 10 fractions) to the disease involving breast or chest wall. FUS-MB treatment was delivered before 1st and 5th fractions of RT (within 1 h). Eight patients with 9 tumours were treated. All 7 evaluable patients with at least 3 months follow-up treated for 8 tumours had a complete response in the treated site. The maximum acute toxicity observed was grade 2 dermatitis in 1 site, and grade 1 in 8 treated sites, at one month post RT, which recovered at 3 months. No RT-related late effect or FUS-MB related toxicity was noted. This study demonstrated safety of combined FUS-MB and RT treatment. Promising response rates suggest potential strong radiosensitization effects of the investigational modality.Trial registration: clinicaltrials.gov, identifier NCT04431674.
Collapse
Affiliation(s)
- Archya Dasgupta
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
- Department of Radiation Oncology, University of Toronto, Toronto, Canada
- Physical Sciences, Sunnybrook Research Institute, Toronto, Canada
| | | | - Evan McNabb
- Physical Sciences, Sunnybrook Research Institute, Toronto, Canada
| | - Ling Ho
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
| | - Lin Lu
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
| | - Danny Vesprini
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
- Department of Radiation Oncology, University of Toronto, Toronto, Canada
| | - Irene Karam
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
- Department of Radiation Oncology, University of Toronto, Toronto, Canada
| | - Hany Soliman
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
- Department of Radiation Oncology, University of Toronto, Toronto, Canada
| | - Edward Chow
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
- Department of Radiation Oncology, University of Toronto, Toronto, Canada
| | - Sonal Gandhi
- Department of Medical Oncology, Sunnybrook Health Sciences Centre, Toronto, Canada
- Department of Medicine, University of Toronto, Toronto, Canada
| | - Maureen Trudeau
- Department of Medical Oncology, Sunnybrook Health Sciences Centre, Toronto, Canada
- Department of Medicine, University of Toronto, Toronto, Canada
| | - William Tran
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
- Department of Radiation Oncology, University of Toronto, Toronto, Canada
| | - Belinda Curpen
- Department of Medical Imaging, Sunnybrook Health Sciences, Toronto, Canada
- Department of Medical Imaging, University of Toronto, Toronto, Canada
| | - Greg Stanisz
- Physical Sciences, Sunnybrook Research Institute, Toronto, Canada
- Department of Biophysics, University of Toronto, Toronto, Canada
- Canada Research Chair in Cancer Imaging, Canadian Institutes of Health Research, Toronto, Canada
| | - Arjun Sahgal
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada
- Department of Radiation Oncology, University of Toronto, Toronto, Canada
| | | | - Gregory J Czarnota
- Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, ON, M4N3M5, Canada.
- Department of Radiation Oncology, University of Toronto, Toronto, Canada.
- Physical Sciences, Sunnybrook Research Institute, Toronto, Canada.
- Department of Biophysics, University of Toronto, Toronto, Canada.
| |
Collapse
|
6
|
Park H, Chung HT, Kim JW, Dho YS, Lee EJ. A 3-month survival model after Gamma Knife surgery in patients with brain metastasis from lung cancer with Karnofsky performance status ≤ 70. Sci Rep 2023; 13:13159. [PMID: 37573417 PMCID: PMC10423256 DOI: 10.1038/s41598-023-40356-6] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/25/2023] [Accepted: 08/09/2023] [Indexed: 08/14/2023] Open
Abstract
Gamma Knife surgery (GKS) for brain metastasis (BM) has been generally advocated for patients with a Karnofsky performance status (KPS) scale of ≥ 70. However, some patients with a poor KPS scale of < 70 are recoverable after GKS and show durable survival. A purpose of this study is to devise a 3-month survival prediction model to screen patients with BM with a KPS of ≤ 70 in whom GKS is needed. A retrospective analysis of 67 patients with a KPS scale of 60-70 undergoing GKS for BM of non-small cell lung cancer (NSCLC) from 2016 to 2020 in our institute was performed. Univariate and multivariate logistic regression analyses were performed to investigate factors related to survival for more than 3 months after GKS. The probability (P) prediction model was designed by giving a weight corresponding to the odds ratio of the variables. The overall survival was 9.9 ± 12.7 months (range 0.2-53.2), with a 3-month survival rate of 59.7% (n = 40). In multivariate logistic regression analysis, extracranial disease (ECD) control (p = .033), focal neurological deficit (FND) (p = .014), and cumulative tumor volume (∑ TV) (p = .005) were associated with 3-month survival. The prediction model of 3-month survival (Harrell's C index = 0.767) was devised based on associated factors. In conclusion, GKS for BMs is recommended in selected patients, even if the KPS scale is ≤ 70.
Collapse
Affiliation(s)
- Hangeul Park
- Department of Neurosurgery, Seoul National University Hospital, Seoul, Republic of Korea
| | - Hyun-Tai Chung
- Department of Neurosurgery, Seoul National University Hospital, Seoul, Republic of Korea
- Gamma Knife Radiosurgery Center, Seoul National University Hospital, Seoul, Republic of Korea
- Seoul National University College of Medicine, Seoul, Republic of Korea
| | - Jin-Wook Kim
- Department of Neurosurgery, Seoul National University Hospital, Seoul, Republic of Korea
- Gamma Knife Radiosurgery Center, Seoul National University Hospital, Seoul, Republic of Korea
- Seoul National University College of Medicine, Seoul, Republic of Korea
| | - Yun-Sik Dho
- Neuro-Oncology Clinic, National Cancer Center, Goyang, Republic of Korea
| | - Eun Jung Lee
- Department of Neurosurgery, Seoul National University Hospital, Seoul, Republic of Korea.
- Gamma Knife Radiosurgery Center, Seoul National University Hospital, Seoul, Republic of Korea.
- Seoul National University College of Medicine, Seoul, Republic of Korea.
| |
Collapse
|
7
|
Byun J, Kim JH. Revisiting the Role of Surgical Resection for Brain Metastasis. Brain Tumor Res Treat 2023; 11:1-7. [PMID: 36762802 PMCID: PMC9911712 DOI: 10.14791/btrt.2022.0028] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/11/2022] [Revised: 01/07/2023] [Accepted: 01/09/2023] [Indexed: 02/05/2023] Open
Abstract
Brain metastasis (BM) is the most common type of brain tumor in adults. The contemporary management of BM remains challenging. Advancements in systemic cancer treatment have increased the survival of patients with cancer. Although the treatment of BM is still complicated, advances in radiotherapy, including stereotactic radiosurgery and chemotherapy, have improved treatment outcomes. Surgical resection is the traditional treatment for BM and its role in the surgical resection of BM has been well established. However, refinement of the surgical resection technique and strategy for BM is needed. Herein, we discuss the evolving role of surgery in patients with BM and the future of BM treatment.
Collapse
Affiliation(s)
- Joonho Byun
- Department of Neurosurgery, Korea University Guro Hospital, Korea University College of Medicine, Seoul, Korea
| | - Jong Hyun Kim
- Department of Neurosurgery, Korea University Guro Hospital, Korea University College of Medicine, Seoul, Korea.
| |
Collapse
|
8
|
Guerini AE, Nici S, Magrini SM, Riga S, Toraci C, Pegurri L, Facheris G, Cozzaglio C, Farina D, Liserre R, Gasparotti R, Ravanelli M, Rondi P, Spiazzi L, Buglione M. Adoption of Hybrid MRI-Linac Systems for the Treatment of Brain Tumors: A Systematic Review of the Current Literature Regarding Clinical and Technical Features. Technol Cancer Res Treat 2023; 22:15330338231199286. [PMID: 37774771 PMCID: PMC10542234 DOI: 10.1177/15330338231199286] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/28/2023] [Revised: 07/24/2023] [Accepted: 08/08/2023] [Indexed: 10/01/2023] Open
Abstract
BACKGROUND Possible advantages of magnetic resonance (MR)-guided radiation therapy (MRgRT) for the treatment of brain tumors include improved definition of treatment volumes and organs at risk (OARs) that could allow margin reductions, resulting in limited dose to the OARs and/or dose escalation to target volumes. Recently, hybrid systems integrating a linear accelerator and an magnetic resonance imaging (MRI) scan (MRI-linacs, MRL) have been introduced, that could potentially lead to a fully MRI-based treatment workflow. METHODS We performed a systematic review of the published literature regarding the adoption of MRL for the treatment of primary or secondary brain tumors (last update November 3, 2022), retrieving a total of 2487 records; after a selection based on title and abstracts, the full text of 74 articles was analyzed, finally resulting in the 52 papers included in this review. RESULTS AND DISCUSSION Several solutions have been implemented to achieve a paradigm shift from CT-based radiotherapy to MRgRT, such as the management of geometric integrity and the definition of synthetic CT models that estimate electron density. Multiple sequences have been optimized to acquire images with adequate quality with on-board MR scanner in limited times. Various sophisticated algorithms have been developed to compensate the impact of magnetic field on dose distribution and calculate daily adaptive plans in a few minutes with satisfactory dosimetric parameters for the treatment of primary brain tumors and cerebral metastases. Dosimetric studies and preliminary clinical experiences demonstrated the feasibility of treating brain lesions with MRL. CONCLUSIONS The adoption of an MRI-only workflow is feasible and could offer several advantages for the treatment of brain tumors, including superior image quality for lesions and OARs and the possibility to adapt the treatment plan on the basis of daily MRI. The growing body of clinical data will clarify the potential benefit in terms of toxicity and response to treatment.
Collapse
Affiliation(s)
- Andrea Emanuele Guerini
- Department of Radiation Oncology, University and Spedali Civili Hospital, Brescia, Italy
- Co-first authors
| | - Stefania Nici
- Medical Physics Department, ASST Spedali Civili Hospital, Brescia, Italy
- Co-first authors
| | - Stefano Maria Magrini
- Department of Radiation Oncology, University and Spedali Civili Hospital, Brescia, Italy
| | - Stefano Riga
- Medical Physics Department, ASST Spedali Civili Hospital, Brescia, Italy
| | - Cristian Toraci
- Medical Physics Department, ASST Spedali Civili Hospital, Brescia, Italy
| | - Ludovica Pegurri
- Department of Radiation Oncology, University and Spedali Civili Hospital, Brescia, Italy
| | - Giorgio Facheris
- Department of Radiation Oncology, University and Spedali Civili Hospital, Brescia, Italy
| | - Claudia Cozzaglio
- Department of Radiation Oncology, University and Spedali Civili Hospital, Brescia, Italy
- Medical Physics Department, ASST Spedali Civili Hospital, Brescia, Italy
| | - Davide Farina
- Radiology Unit, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy
| | - Roberto Liserre
- Department of Radiology, Neuroradiology Unit, ASST Spedali Civili University Hospital, Brescia, Italy
| | - Roberto Gasparotti
- Neuroradiology Unit, Department of Medical-Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy
| | - Marco Ravanelli
- Radiology Unit, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy
| | - Paolo Rondi
- Radiology Unit, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy
| | - Luigi Spiazzi
- Medical Physics Department, ASST Spedali Civili Hospital, Brescia, Italy
- Co-last author
| | - Michela Buglione
- Department of Radiation Oncology, University and Spedali Civili Hospital, Brescia, Italy
- Co-last author
| |
Collapse
|
9
|
Huang K, Hernandez S, Wang C, Nguyen C, Briere TM, Cardenas C, Court L, Xiao Y. Automated field-in-field whole brain radiotherapy planning. J Appl Clin Med Phys 2022; 24:e13819. [PMID: 36354957 PMCID: PMC9924111 DOI: 10.1002/acm2.13819] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/01/2022] [Revised: 10/03/2022] [Accepted: 10/11/2022] [Indexed: 11/12/2022] Open
Abstract
PURPOSE We developed and tested an automatic field-in-field (FIF) solution for whole-brain radiotherapy (WBRT) planning that creates a homogeneous dose distribution by minimizing hotspots, resulting in clinically acceptable plans. METHODS A configurable auto-planning algorithm was developed to automatically generate FIF WBRT plans independent of the treatment planning system. Configurable parameters include the definition of hotspots, target volume, maximum number of subfields, and minimum number of monitor units per field. This algorithm iteratively identifies a hotspot, creates two opposing subfields, calculates the dose, and optimizes the beam weight based on user-configured constraints of dose-volume histogram coverage and least-squared cost functions. The algorithm was retrospectively tested on 17 whole-brain patients. First, an in-house landmark-based automated beam aperture technique was used to generate the treatment fields and initial plans. Second, the FIF algorithm was employed to optimize the plans using physician-defined goals of 99.9% of the brain volume receiving 100% of the prescription dose (30 Gy in 10 fractions) and a target hotspot definition of 107% of the prescription dose. The final auto-optimized plans were assessed for clinical acceptability by an experienced radiation oncologist using a five-point scale. RESULTS The FIF algorithm reduced the mean (± SD) plan hotspot percentage dose from 35.0 Gy (116.6%) ± 0.6 Gy (2.0%) to 32.6 Gy (108.8%) ± 0.4 Gy (1.2%). Also, it decreased the mean (± SD) hotspot V107% [cm3 ] from 959 ± 498 cm3 to 145 ± 224 cm3 . On average, plans were produced in 16 min without any user intervention. Furthermore, 76.5% of the auto-plans were clinically acceptable (needing no or minor stylistic edits), and all of them were clinically acceptable after minor clinically necessary edits. CONCLUSIONS This algorithm successfully produced high-quality WBRT plans and can improve treatment planning efficiency when incorporated into an automatic planning workflow.
Collapse
Affiliation(s)
- Kai Huang
- The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical SciencesHoustonTexasUSA,Department of Radiation PhysicsThe University of Texas MD Anderson Cancer CenterHoustonTexasUSA
| | - Soleil Hernandez
- The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical SciencesHoustonTexasUSA,Department of Radiation PhysicsThe University of Texas MD Anderson Cancer CenterHoustonTexasUSA
| | - Chenyang Wang
- Department of Radiation OncologyThe University of Texas MD Anderson Cancer CenterHoustonTexasUSA
| | - Callistus Nguyen
- Department of Radiation PhysicsThe University of Texas MD Anderson Cancer CenterHoustonTexasUSA
| | - Tina Marie Briere
- Department of Radiation PhysicsThe University of Texas MD Anderson Cancer CenterHoustonTexasUSA
| | - Carlos Cardenas
- Department of Radiation OncologyThe University of Alabama at BirminghamBirminghamAlabamaUSA
| | - Laurence Court
- Department of Radiation PhysicsThe University of Texas MD Anderson Cancer CenterHoustonTexasUSA
| | - Yao Xiao
- Department of Radiation PhysicsThe University of Texas MD Anderson Cancer CenterHoustonTexasUSA
| |
Collapse
|
10
|
Liang Y, Lee K, Bovi JA, Palmer JD, Brown PD, Gondi V, Tomé WA, Benzinger TLS, Mehta MP, Li XA. Deep Learning-Based Automatic Detection of Brain Metastases in Heterogenous Multi-Institutional Magnetic Resonance Imaging Sets: An Exploratory Analysis of NRG-CC001. Int J Radiat Oncol Biol Phys 2022; 114:529-536. [PMID: 35787927 PMCID: PMC9641965 DOI: 10.1016/j.ijrobp.2022.06.081] [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: 01/14/2022] [Revised: 06/09/2022] [Accepted: 06/21/2022] [Indexed: 10/31/2022]
Abstract
PURPOSE Deep learning-based algorithms have been shown to be able to automatically detect and segment brain metastases (BMs) in magnetic resonance imaging, mostly based on single-institutional data sets. This work aimed to investigate the use of deep convolutional neural networks (DCNN) for BM detection and segmentation on a highly heterogeneous multi-institutional data set. METHODS AND MATERIALS A total of 407 patients from 98 institutions were randomly split into 326 patients from 78 institutions for training/validation and 81 patients from 20 institutions for unbiased testing. The data set contained T1-weighted gadolinium and T2-weighted fluid-attenuated inversion recovery magnetic resonance imaging acquired on diverse scanners using different pulse sequences and various acquisition parameters. Several variants of 3-dimensional U-Net based DCNN models were trained and tuned using 5-fold cross validation on the training set. Performances of different models were compared based on Dice similarity coefficient for segmentation and sensitivity and false positive rate (FPR) for detection. The best performing model was evaluated on the test set. RESULTS A DCNN with an input size of 64 × 64 × 64 and an equal number of 128 kernels for all convolutional layers using instance normalization was identified as the best performing model (Dice similarity coefficient 0.73, sensitivity 0.86, and FPR 1.9) in the 5-fold cross validation experiments. The best performing model demonstrated consistent behavior on the test set (Dice similarity coefficient 0.73, sensitivity 0.91, and FPR 1.7) and successfully detected 7 BMs (out of 327) that were missed during manual delineation. For large BMs with diameters greater than 12 mm, the sensitivity and FPR improved to 0.98 and 0.3, respectively. CONCLUSIONS The DCNN model developed can automatically detect and segment brain metastases with reasonable accuracy, high sensitivity, and low FPR on a multi-institutional data set with nonprespecified and highly variable magnetic resonance imaging sequences. For large BMs, the model achieved clinically relevant results. The model is robust and may be potentially used in real-world situations.
Collapse
Affiliation(s)
- Ying Liang
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, Wisconsin
| | - Karen Lee
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, Wisconsin
| | - Joseph A Bovi
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, Wisconsin
| | - Joshua D Palmer
- Department of Radiation Oncology, The James Cancer Hospital and Solove Research Institute at the Ohio State University, Columbus, Ohio
| | - Paul D Brown
- Department of Radiation Oncology, Mayo Clinic, Rochester, Minnesota
| | - Vinai Gondi
- Department of Radiation Oncology, Northwestern Medicine Cancer Center and Proton Center, Warrenville, Illinois
| | - Wolfgang A Tomé
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, New York
| | - Tammie L S Benzinger
- Department of Radiology, Washington University School of Medicine, St Louis, Missouri
| | | | - X Allen Li
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, Wisconsin.
| |
Collapse
|
11
|
Hasanov E, Yeboa DN, Tucker MD, Swanson TA, Beckham TH, Rini B, Ene CI, Hasanov M, Derks S, Smits M, Dudani S, Heng DYC, Brastianos PK, Bex A, Hanalioglu S, Weinberg JS, Hirsch L, Carlo MI, Aizer A, Brown PD, Bilen MA, Chang EL, Jaboin J, Brugarolas J, Choueiri TK, Atkins MB, McGregor BA, Halasz LM, Patel TR, Soltys SG, McDermott DF, Elder JB, Baskaya MK, Yu JB, Timmerman R, Kim MM, Mut M, Markert J, Beal K, Tannir NM, Samandouras G, Lang FF, Giles R, Jonasch E. An interdisciplinary consensus on the management of brain metastases in patients with renal cell carcinoma. CA Cancer J Clin 2022; 72:454-489. [PMID: 35708940 DOI: 10.3322/caac.21729] [Citation(s) in RCA: 15] [Impact Index Per Article: 7.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/08/2021] [Revised: 03/28/2022] [Accepted: 04/11/2022] [Indexed: 12/23/2022] Open
Abstract
Brain metastases are a challenging manifestation of renal cell carcinoma. We have a limited understanding of brain metastasis tumor and immune biology, drivers of resistance to systemic treatment, and their overall poor prognosis. Current data support a multimodal treatment strategy with radiation treatment and/or surgery. Nonetheless, the optimal approach for the management of brain metastases from renal cell carcinoma remains unclear. To improve patient care, the authors sought to standardize practical management strategies. They performed an unstructured literature review and elaborated on the current management strategies through an international group of experts from different disciplines assembled via the network of the International Kidney Cancer Coalition. Experts from different disciplines were administered a survey to answer questions related to current challenges and unmet patient needs. On the basis of the integrated approach of literature review and survey study results, the authors built algorithms for the management of single and multiple brain metastases in patients with renal cell carcinoma. The literature review, consensus statements, and algorithms presented in this report can serve as a framework guiding treatment decisions for patients. CA Cancer J Clin. 2022;72:454-489.
Collapse
Affiliation(s)
- Elshad Hasanov
- Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - Debra Nana Yeboa
- Department of Radiation Oncology, Division of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - Mathew D Tucker
- Department of Medicine, Division of Hematology and Oncology, Vanderbilt University Medical Center, Nashville, Tennessee
| | - Todd A Swanson
- Department of Radiation Oncology, Division of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - Thomas Hendrix Beckham
- Department of Radiation Oncology, Division of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - Brian Rini
- Department of Medicine, Division of Hematology and Oncology, Vanderbilt University Medical Center, Nashville, Tennessee
| | - Chibawanye I Ene
- Department of Neurosurgery, Division of Surgery, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - Merve Hasanov
- Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - Sophie Derks
- Department of Radiology and Nuclear Medicine, Erasmus Medical Center, University Medical Center Rotterdam, Rotterdam, the Netherlands
| | - Marion Smits
- Department of Radiology and Nuclear Medicine, Erasmus Medical Center, University Medical Center Rotterdam, Rotterdam, the Netherlands
| | - Shaan Dudani
- Division of Oncology/Hematology, William Osler Health System, Brampton, Ontario, Canada
| | - Daniel Y C Heng
- Tom Baker Cancer Center, University of Calgary, Calgary, Alberta, Canada
| | - Priscilla K Brastianos
- Division of Neuro-Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
| | - Axel Bex
- The Royal Free London National Health Service Foundation Trust, London, United Kingdom
- University College London Division of Surgery and Interventional Science, London, United Kingdom
- Department of Urology, The Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, the Netherlands
| | - Sahin Hanalioglu
- Department of Neurosurgery, Hacettepe University Faculty of Medicine, Ankara, Turkey
| | - Jeffrey S Weinberg
- Department of Neurosurgery, Division of Surgery, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - Laure Hirsch
- Department of Medical Oncology, Cochin University Hospital, Public Assistance Hospital of Paris, Paris, France
- Lank Center for Genitourinary Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts
| | - Maria I Carlo
- Genitourinary Oncology Service, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, New York
| | - Ayal Aizer
- Department of Radiation Oncology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts
| | - Paul David Brown
- Department of Radiation Oncology, Mayo Clinic, Rochester, Minnesota
| | - Mehmet Asim Bilen
- Department of Hematology and Medical Oncology, Emory University School of Medicine, Atlanta, Georgia
- Winship Cancer Institute of Emory University, Atlanta, Georgia
| | - Eric Lin Chang
- Department of Radiation Oncology, University of Southern California, Keck School of Medicine, California, Los Angeles
| | - Jerry Jaboin
- Department of Radiation Medicine, Oregon Health & Science University, Portland, Oregon
| | - James Brugarolas
- Kidney Cancer Program, Simmons Comprehensive Cancer Center, The University of Texas Southwestern Medical Center, Dallas, Texas
- Division of Hematology/Oncology, Department of Internal Medicine, The University of Texas Southwestern Medical Center, Dallas, Texas
| | - Toni K Choueiri
- Lank Center for Genitourinary Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts
| | - Michael B Atkins
- Lombardi Comprehensive Cancer Center, MedStar Georgetown University Hospital, Washington, DC
| | - Bradley A McGregor
- Lank Center for Genitourinary Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts
| | - Lia M Halasz
- Department of Radiation Oncology, University of Washington, Seattle, Washington
| | - Toral R Patel
- Kidney Cancer Program, Simmons Comprehensive Cancer Center, The University of Texas Southwestern Medical Center, Dallas, Texas
- Department of Neurosurgery, The University of Texas Southwestern Medical Center, Dallas, Texas
| | - Scott G Soltys
- Department of Radiation Oncology, Stanford Cancer Institute, Stanford, California
| | - David F McDermott
- Division of Medical Oncology, Beth Israel Deaconess Medical Center, Boston, Massachusetts
| | - James Bradley Elder
- Department of Neurological Surgery, Ohio State University Wexner Medical Center, Columbus, Ohio
| | - Mustafa K Baskaya
- Department of Neurological Surgery, University of Wisconsin-Madison, School of Medicine and Public Health, Madison, Wisconsin
| | - James B Yu
- Department of Therapeutic Radiology, Yale University School of Medicine, New Haven, Connecticut
| | - Robert Timmerman
- Kidney Cancer Program, Simmons Comprehensive Cancer Center, The University of Texas Southwestern Medical Center, Dallas, Texas
- Department of Radiation Oncology, The University of Texas Southwestern Medical Center, Dallas, Texas
| | - Michelle Miran Kim
- Department of Radiation Oncology, University of Michigan, Ann Arbor, Michigan
| | - Melike Mut
- Department of Neurosurgery, Hacettepe University Faculty of Medicine, Ankara, Turkey
| | - James Markert
- Department of Neurosurgery, The University of Alabama at Birmingham, Birmingham, Alabama
| | - Kathryn Beal
- Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, New York
| | - Nizar M Tannir
- Department of Genitourinary Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - George Samandouras
- Victor Horsley Department of Neurosurgery, The National Hospital for Neurology and Neurosurgery, Queen Square, London, United Kingdom
- University College London Queen Square Institute of Neurology, University College London, Queen Square, London, United Kingdom
| | - Frederick F Lang
- Department of Neurosurgery, Division of Surgery, The University of Texas MD Anderson Cancer Center, Houston, Texas
| | - Rachel Giles
- International Kidney Cancer Coalition, Duivendrecht, the Netherlands
| | - Eric Jonasch
- Department of Genitourinary Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas
| |
Collapse
|
12
|
Establishment and Validation of CyberKnife Irradiation in a Syngeneic Glioblastoma Mouse Model. Cancers (Basel) 2021; 13:cancers13143416. [PMID: 34298631 PMCID: PMC8303959 DOI: 10.3390/cancers13143416] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/14/2021] [Revised: 07/01/2021] [Accepted: 07/05/2021] [Indexed: 12/12/2022] Open
Abstract
Simple Summary Stereotactic radiosurgery (SRS) provides precise high-dose irradiation of intracranial tumors. However, its radiobiological mechanisms are not fully understood. This study aims to establish CyberKnife SRS on an intracranial glioblastoma tumor mouse model and assesses the early radiobiological effects of radiosurgery. Following exposure to a single dose of 20 Gy, the tumor volume was evaluated using MRI scans, whereas cellular proliferation and apoptosis, tumor vasculature, and immune response were evaluated using immunofluorescence staining. The mean tumor volume was significantly reduced by approximately 75% after SRS. The precision of irradiation was verified by the detection of DNA damage consistent with the planned dose distribution. Our study provides a suitable mouse model for reproducible and effective irradiation and further investigation of radiobiological effects and combination therapies of intracranial tumors using CyberKnife. Abstract CyberKnife stereotactic radiosurgery (CK-SRS) precisely delivers radiation to intracranial tumors. However, the underlying radiobiological mechanisms at high single doses are not yet fully understood. Here, we established and evaluated the early radiobiological effects of CK-SRS treatment at a single dose of 20 Gy after 15 days of tumor growth in a syngeneic glioblastoma-mouse model. Exact positioning was ensured using a custom-made, non-invasive, and trackable frame. One superimposed target volume for the CK-SRS planning was created from the fused tumor volumes obtained from MRIs prior to irradiation. Dose calculation and delivery were planned using a single-reference CT scan. Six days after irradiation, tumor volumes were measured using MRI scans, and radiobiological effects were assessed using immunofluorescence staining. We found that CK-SRS treatment reduced tumor volume by approximately 75%, impaired cell proliferation, diminished tumor vasculature, and increased immune response. The accuracy of the delivered dose was demonstrated by staining of DNA double-strand breaks in accordance with the planned dose distribution. Overall, we confirmed that our proposed setup enables the precise irradiation of intracranial tumors in mice using only one reference CT and superimposed MRI volumes. Thus, our proposed mouse model for reproducible CK-SRS can be used to investigate radiobiological effects and develop novel therapeutic approaches.
Collapse
|
13
|
Jünger ST, Hoyer UCI, Schaufler D, Laukamp KR, Goertz L, Thiele F, Grunz JP, Schlamann M, Perkuhn M, Kabbasch C, Persigehl T, Grau S, Borggrefe J, Scheffler M, Shahzad R, Pennig L. Fully Automated MR Detection and Segmentation of Brain Metastases in Non-small Cell Lung Cancer Using Deep Learning. J Magn Reson Imaging 2021; 54:1608-1622. [PMID: 34032344 DOI: 10.1002/jmri.27741] [Citation(s) in RCA: 22] [Impact Index Per Article: 7.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/15/2021] [Revised: 05/12/2021] [Accepted: 05/12/2021] [Indexed: 12/14/2022] Open
Abstract
BACKGROUND Non-small cell lung cancer (NSCLC) is the most common tumor entity spreading to the brain and up to 50% of patients develop brain metastases (BMs). Detection of BMs on MRI is challenging with an inherent risk of missed diagnosis. PURPOSE To train and evaluate a deep learning model (DLM) for fully automated detection and 3D segmentation of BMs in NSCLC on clinical routine MRI. STUDY TYPE Retrospective. POPULATION Ninety-eight NSCLC patients with 315 BMs on pretreatment MRI, divided into training (66 patients, 248 BMs) and independent test (17 patients, 67 BMs) and control (15 patients, 0 BMs) cohorts. FIELD STRENGTH/SEQUENCE T1 -/T2 -weighted, T1 -weighted contrast-enhanced (T1 CE; gradient-echo and spin-echo sequences), and FLAIR at 1.0, 1.5, and 3.0 T from various vendors and study centers. ASSESSMENT A 3D convolutional neural network (DeepMedic) was trained on the training cohort using 5-fold cross-validation and evaluated on the independent test and control sets. Three-dimensional voxel-wise manual segmentations of BMs by a neurosurgeon and a radiologist on T1 CE served as the reference standard. STATISTICAL TESTS Sensitivity (recall) and false positive (FP) findings per scan, dice similarity coefficient (DSC) to compare the spatial overlap between manual and automated segmentations, Pearson's correlation coefficient (r) to evaluate the relationship between quantitative volumetric measurements of segmentations, and Wilcoxon rank-sum test to compare the volumes of BMs. A P value <0.05 was considered statistically significant. RESULTS In the test set, the DLM detected 57 of the 67 BMs (mean volume: 0.99 ± 4.24 cm3 ), resulting in a sensitivity of 85.1%, while FP findings of 1.5 per scan were observed. Missed BMs had a significantly smaller volume (0.05 ± 0.04 cm3 ) than detected BMs (0.96 ± 2.4 cm3 ). Compared with the reference standard, automated segmentations achieved a median DSC of 0.72 and a good volumetric correlation (r = 0.95). In the control set, 1.8 FPs/scan were observed. DATA CONCLUSION Deep learning provided a high detection sensitivity and good segmentation performance for BMs in NSCLC on heterogeneous scanner data while yielding a low number of FP findings. Level of Evidence 3 Technical Efficacy Stage 2.
Collapse
Affiliation(s)
- Stephanie T Jünger
- Department of General Neurosurgery, Center for Neurosurgery, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany.,Centre for Integrated Oncology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| | - Ulrike Cornelia Isabel Hoyer
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| | - Diana Schaufler
- Department of Internal Medicine, Center for Integrated Oncology Aachen Bonn Cologne Duesseldorf, Network Genomic Medicine, Lung Cancer Group Cologne, Faculty of Medicine and University Hospital of Cologne, University of Cologne, Cologne, Germany
| | - Kai Roman Laukamp
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| | - Lukas Goertz
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| | - Frank Thiele
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany.,Philips GmbH Innovative Technologies, Aachen, Germany
| | - Jan-Peter Grunz
- Department of Diagnostic and Interventional Radiology, University Hospital Würzburg, Würzburg, Germany
| | - Marc Schlamann
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| | - Michael Perkuhn
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany.,Philips GmbH Innovative Technologies, Aachen, Germany
| | - Christoph Kabbasch
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| | - Thorsten Persigehl
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| | - Stefan Grau
- Department of General Neurosurgery, Center for Neurosurgery, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany.,Centre for Integrated Oncology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| | - Jan Borggrefe
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany.,Department of Radiology, Neuroradiology and Nuclear Medicine, Johannes Wesling University Hospital, Ruhr University Bochum, Bochum, Germany
| | - Matthias Scheffler
- Department of Internal Medicine, Center for Integrated Oncology Aachen Bonn Cologne Duesseldorf, Network Genomic Medicine, Lung Cancer Group Cologne, Faculty of Medicine and University Hospital of Cologne, University of Cologne, Cologne, Germany
| | - Rahil Shahzad
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany.,Philips GmbH Innovative Technologies, Aachen, Germany
| | - Lenhard Pennig
- Institute for Diagnostic and Interventional Radiology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany
| |
Collapse
|
14
|
The Management of Brain Metastases-Systematic Review of Neurosurgical Aspects. Cancers (Basel) 2021; 13:cancers13071616. [PMID: 33807384 PMCID: PMC8036330 DOI: 10.3390/cancers13071616] [Citation(s) in RCA: 18] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/21/2021] [Revised: 03/23/2021] [Accepted: 03/26/2021] [Indexed: 02/07/2023] Open
Abstract
Simple Summary In this comprehensive review, we focused on the neurosurgical treatment as an integrative part of the challenging multidisciplinary management of cerebral metastases, a neuro-oncologic entity, which has been observed to have an increased incidence over the last years. In selected cases, the surgical removal of the space-occupying mass reduces the intracranial pressure, normalizes the metabolic environment, reduces the symptom burden, and allows for the intensification of local and systemic adjuvant treatment. In detail, we discuss the incidence of brain metastases, the role of surgical resection, as well as the evolution of current neurosurgical techniques, the surgical morbidity and mortality of single and multiple lesions, and we enlighten the role of surgery for recurrent tumors. Abstract The multidisciplinary management of patients with brain metastases (BM) consists of surgical resection, different radiation treatment modalities, cytotoxic chemotherapy, and targeted molecular treatment. This review presents the current state of neurosurgical technology applied to achieve maximal resection with minimal morbidity as a treatment paradigm in patients with BM. In addition, we discuss the contribution of neurosurgical resection on functional outcome, advanced systemic treatment strategies, and enhanced understanding of the tumor biology.
Collapse
|
15
|
Alyahyawi A, Dimitriadis A, Nisbet A, Bradley D. GeB flat fibre TL dosimeters for in-vivo measurements in radiosurgery. Radiat Phys Chem Oxf Engl 1993 2021. [DOI: 10.1016/j.radphyschem.2020.108973] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/20/2022]
|
16
|
Positioning accuracy of a single-isocenter multiple targets SRS treatment: A comparison between Varian TrueBeam CBCT and Brainlab ExacTrac. Phys Med 2020; 80:267-273. [PMID: 33221708 DOI: 10.1016/j.ejmp.2020.10.022] [Citation(s) in RCA: 8] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/30/2020] [Revised: 10/21/2020] [Accepted: 10/26/2020] [Indexed: 11/21/2022] Open
Abstract
PURPOSE This study compared the positioning accuracy between cone-beam CT (CBCT) and ExacTrac (ETX) for a single-isocenter multiple target stereotactic radiosurgery (SRS) on two TrueBeam STx systems. METHODS A single-isocenter treatment plan was simulated on an anthropomorphic head phantom with six spherical steel ball bearings (BBs). One of the BBs was chosen to be the isocenter. The five off-isocenter targets were located at various distances from the isocenter. MV portal images were generated to evaluate the deviations between the expected and the real center of the targets after CBCT and ETX positioning, respectively. RESULTS The evaluation of the positioning accuracy for the isocenter target showed that CBCT and ETX positioning provided comparable, sub-millimetric results. Deviations in positioning accuracy were also calculated for all other targets, also showing comparable results for CBCT and ETX. Moreover, our study showed that the deviation between CBCT and ETX positioning were in better agreement for TBSTx1 and deviated slightly higher on TBSTx2 (maximum: 1.23 mm at S/I direction), due to a less perfect alignment between the CBCT coordinate system and the ETX coordinate system on TBSTx2 compared to TBSTx1. This study also showed a correlation between the target positioning accuracy and the distance to the isocenter. CONCLUSION The positioning accuracy of ETX and CBCT for targets located at isocenter and off-isocenter locations was compared on two treatment machines and found comparable. Our study highlights the importance of a proper calibration procedure, to ensure correct alignment between the CBCT, ETX and machine coordinate systems.
Collapse
|
17
|
Sayan M, Mustafayev TZ, Balmuk A, Mamidanna S, Kefelioglu ESS, Gungor G, Chundury A, Ohri N, Karaarslan E, Ozyar E, Atalar B. Management of symptomatic radiation necrosis after stereotactic radiosurgery and clinical factors for treatment response. Radiat Oncol J 2020; 38:176-180. [PMID: 33012145 PMCID: PMC7533401 DOI: 10.3857/roj.2020.00171] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/25/2020] [Revised: 06/26/2020] [Accepted: 07/06/2020] [Indexed: 01/31/2023] Open
Abstract
PURPOSE Approximately 10% of patients who received brain stereotactic radiosurgery (SRS) develop symptomatic radiation necrosis (RN). We sought to determine the effectiveness of treatment options for symptomatic RN, based on patient-reported outcomes. MATERIALS AND METHODS We conducted a retrospective review of 217 patients with 414 brain metastases treated with SRS from 2009 to 2018 at our institution. Symptomatic RN was determined by appearance on serial magnetic resonance images (MRIs), MR spectroscopy, requirement of therapy, and development of new neurological complaints without evidence of disease progression. Therapeutic interventions for symptomatic RN included corticosteroids, bevacizumab and/or surgical resection. Patient-reported therapeutic outcomes were graded as complete response (CR), partial response (PR), and no response. RESULTS Twenty-six patients experienced symptomatic RN after treatment of 50 separate lesions. The mean prescription dose was 22 Gy (range, 15 to 30 Gy) in 1 to 5 fractions (median, 1 fraction). Of the 12 patients managed with corticosteroids, 6 patients (50%) reported CR and 4 patients (33%) PR. Of the 6 patients managed with bevacizumab, 3 patients (50%) reported CR and 1 patient (18%) PR. Of the 8 patients treated with surgical resection, all reported CR (100%). Other than surgical resection, age ≥54 years (median, 54 years; range, 35 to 81 years) was associated with CR (odds ratio = 8.40; 95% confidence interval, 1.27-15.39; p = 0.027). CONCLUSION Corticosteroids and bevacizumab are commonly utilized treatment modalities with excellent response rate. Our results suggest that patient's age is associated with response rate and could help guide treatment decisions for unresectable symptomatic RN.
Collapse
Affiliation(s)
- Mutlay Sayan
- Department of Radiation Oncology, Rutgers Cancer Institute of New Jersey, New Brunswick, NJ, USA
| | | | - Aykut Balmuk
- School of Medicine, Mehmet Ali Aydinlar Acibadem University, Istanbul, Turkey
| | - Swati Mamidanna
- Department of Radiation Oncology, Rutgers Cancer Institute of New Jersey, New Brunswick, NJ, USA
| | | | - Gorkem Gungor
- Institute of Health Sciences, Mehmet Ali Aydinlar Acibadem University, Istanbul, Turkey
| | - Anupama Chundury
- Department of Radiation Oncology, Rutgers Cancer Institute of New Jersey, New Brunswick, NJ, USA
| | - Nisha Ohri
- Department of Radiation Oncology, Rutgers Cancer Institute of New Jersey, New Brunswick, NJ, USA
| | - Ercan Karaarslan
- Department of Radiology, School of Medicine, Mehmet Ali Aydinlar Acibadem University, Istanbul, Turkey
| | - Enis Ozyar
- School of Medicine, Mehmet Ali Aydinlar Acibadem University, Istanbul, Turkey
| | - Banu Atalar
- School of Medicine, Mehmet Ali Aydinlar Acibadem University, Istanbul, Turkey
| |
Collapse
|
18
|
Granina E, Fehniger J, Kondziolka D, Silverman J, Downey A, Placantonakis D, Muggia F. Endometrial adenocarcinoma presenting as a suprasellar mass: lessons to be learned. Ecancermedicalscience 2020; 14:1083. [PMID: 32863877 PMCID: PMC7434505 DOI: 10.3332/ecancer.2020.1083] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/15/2020] [Indexed: 12/03/2022] Open
Abstract
A 66-year-old woman with a history of stage IA mixed endometrioid and serous endometrial cancer presented to our centre with 2 weeks of worsening headaches nearly 4 years after her initial surgery. At admission, she manifested bitemporal hemianopsia, difficulty walking and clinical and laboratory findings of panhypopituitarism, including diabetes insipidus. Magnetic resonance imaging of the brain revealed a 2.7 cm sellar/suprasellar mass compressing the optic chiasm and infiltrating the pituitary stalk. Computerised tomography documented mediastinal, lung, adrenal and liver involvement, including a 2.5 cm palpable left supraclavicular node that on excisional biopsy demonstrated metastatic endometrial adenocarcinoma. Due to the advanced stage of her cancer as well as the presence of multiple metastases, including lung and hepatic metastases causing post-obstructive pneumonia and coagulopathy, the sellar/suprasellar mass was treated with fractionated radiosurgery rather than surgical excision.
Collapse
Affiliation(s)
- Evgenia Granina
- Department of Internal Medicine, NYU Langone Health, New York, NY 10016, USA
| | - Julia Fehniger
- Department of Gynecologic Oncology, NYU Langone Health, New York, NY 10016, USA
| | | | - Joshua Silverman
- Department of Radiation Oncology, NYU Langone Health, New York, NY 10016, USA
| | - Andrea Downey
- Department of Pathology, NYU Langone Health, New York, NY 10016, USA
| | | | - Franco Muggia
- Department of Medical Oncology, NYU Langone Health, New York, NY 10016, USA
| |
Collapse
|
19
|
Sawlani V, Patel MD, Davies N, Flintham R, Wesolowski R, Ughratdar I, Pohl U, Nagaraju S, Petrik V, Kay A, Jacob S, Sanghera P, Wykes V, Watts C, Poptani H. Multiparametric MRI: practical approach and pictorial review of a useful tool in the evaluation of brain tumours and tumour-like lesions. Insights Imaging 2020; 11:84. [PMID: 32681296 PMCID: PMC7367972 DOI: 10.1186/s13244-020-00888-1] [Citation(s) in RCA: 42] [Impact Index Per Article: 10.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/22/2020] [Accepted: 06/24/2020] [Indexed: 12/17/2022] Open
Abstract
MRI has a vital role in the assessment of intracranial lesions. Conventional MRI has limited specificity and multiparametric MRI using diffusion-weighted imaging, perfusion-weighted imaging and magnetic resonance spectroscopy allows more accurate assessment of the tissue microenvironment. The purpose of this educational pictorial review is to demonstrate the role of multiparametric MRI for diagnosis, treatment planning and for assessing treatment response, as well as providing a practical approach for performing and interpreting multiparametric MRI in the clinical setting. A variety of cases are presented to demonstrate how multiparametric MRI can help differentiate neoplastic from non-neoplastic lesions compared to conventional MRI alone.
Collapse
Affiliation(s)
- Vijay Sawlani
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK.
- University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK.
| | - Markand Dipankumar Patel
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
- University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
| | - Nigel Davies
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Robert Flintham
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Roman Wesolowski
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Ismail Ughratdar
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Ute Pohl
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Santhosh Nagaraju
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Vladimir Petrik
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Andrew Kay
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Saiju Jacob
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
- University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
| | - Paul Sanghera
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
| | - Victoria Wykes
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
- University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
| | - Colin Watts
- University Hospitals Birmingham NHS Foundation Trust, Queen Elizabeth Hospital Birmingham, Mindelsohn Way, Edgbaston, Birmingham, B15 2TH, UK
- University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
| | - Harish Poptani
- Centre for Pre-Clinical Imaging, Department of Cellular and Molecular Physiology, University of Liverpool, Crown Street, Liverpool, L69 3BX, UK
| |
Collapse
|
20
|
Ren D, Cheng H, Wang X, Vishnoi M, Teh BS, Rostomily R, Chang J, Wong ST, Zhao H. Emerging treatment strategies for breast cancer brain metastasis: from translational therapeutics to real-world experience. Ther Adv Med Oncol 2020; 12:1758835920936151. [PMID: 32655700 PMCID: PMC7328353 DOI: 10.1177/1758835920936151] [Citation(s) in RCA: 17] [Impact Index Per Article: 4.3] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/24/2020] [Accepted: 05/21/2020] [Indexed: 01/08/2023] Open
Abstract
Systemic therapies for primary breast cancer have made great progress over the past two decades. However, oncologists confront an insidious and particularly difficult problem: in those patients with metastatic breast cancer, up to 50% of human epidermal growth factor 2 (HER2)-positive and 25-40% of triple-negative subtypes, brain metastases (BM) kill most of them. Fortunately, standard- of-care treatments for BM have improved rapidly, with a decline in whole brain radiation therapy and use of fractionated stereotactic radiosurgery as well as targeted therapies and immunotherapies. Meanwhile, advances in fundamental understanding of the basic biological processes of breast cancer BM (BCBM) have led to many novel experimental therapeutic strategies. In this review, we describe the most recent clinical treatment options and emerging experimental therapeutic strategies that have the potential to combat BCBM.
Collapse
Affiliation(s)
- Ding Ren
- Outpatient Department, PLA Navy NO.905 Hospital,
Shanghai, P.R. China
| | - Hao Cheng
- Department of Orthopedics, Tongji Hospital,
Wuhan, P.R. China
| | - Xin Wang
- Department of Systems Medicine and
Bioengineering, Houston Methodist Cancer Center, Weill Cornell Medicine,
Houston, TX, USA
| | - Monika Vishnoi
- Department of Neurosurgery, Houston Methodist
Hospital, Weill Cornell Medicine, Houston, TX, USA
| | - Bin S. Teh
- Department of Radiation Oncology, Houston
Methodist Hospital, Weill Cornell Medicine, Houston, TX, USA
| | - Robert Rostomily
- Department of Neurosurgery, Houston Methodist
Hospital, Weill Cornell Medicine, Houston, TX, USA
| | - Jenny Chang
- Houston Methodist Cancer Center, Weill Cornell
Medicine, Houston, TX, USA
| | - Stephen T. Wong
- Department of Systems Medicine and
Bioengineering, Houston Methodist Cancer Center, Weill Cornell Medicine,
6670 Bertner Ave, Houston, TX 77030, USA
| | - Hong Zhao
- Department of Systems Medicine and
Bioengineering, Houston Methodist Cancer Center, Weill Cornell Medicine,
6670 Bertner Ave, Houston, TX 77030, USA
| |
Collapse
|
21
|
Bousabarah K, Ruge M, Brand JS, Hoevels M, Rueß D, Borggrefe J, Große Hokamp N, Visser-Vandewalle V, Maintz D, Treuer H, Kocher M. Deep convolutional neural networks for automated segmentation of brain metastases trained on clinical data. Radiat Oncol 2020; 15:87. [PMID: 32312276 PMCID: PMC7171921 DOI: 10.1186/s13014-020-01514-6] [Citation(s) in RCA: 57] [Impact Index Per Article: 14.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/29/2020] [Accepted: 03/13/2020] [Indexed: 12/22/2022] Open
Abstract
INTRODUCTION Deep learning-based algorithms have demonstrated enormous performance in segmentation of medical images. We collected a dataset of multiparametric MRI and contour data acquired for use in radiosurgery, to evaluate the performance of deep convolutional neural networks (DCNN) in automatic segmentation of brain metastases (BM). METHODS A conventional U-Net (cU-Net), a modified U-Net (moU-Net) and a U-Net trained only on BM smaller than 0.4 ml (sU-Net) were implemented. Performance was assessed on a separate test set employing sensitivity, specificity, average false positive rate (AFPR), the dice similarity coefficient (DSC), Bland-Altman analysis and the concordance correlation coefficient (CCC). RESULTS A dataset of 509 patients (1223 BM) was split into a training set (469 pts) and a test set (40 pts). A combination of all trained networks was the most sensitive (0.82) while maintaining a specificity 0.83. The same model achieved a sensitivity of 0.97 and a specificity of 0.94 when considering only lesions larger than 0.06 ml (75% of all lesions). Type of primary cancer had no significant influence on the mean DSC per lesion (p = 0.60). Agreement between manually and automatically assessed tumor volumes as quantified by a CCC of 0.87 (95% CI, 0.77-0.93), was excellent. CONCLUSION Using a dataset which properly captured the variation in imaging appearance observed in clinical practice, we were able to conclude that DCNNs reach clinically relevant performance for most lesions. Clinical applicability is currently limited by the size of the target lesion. Further studies should address if small targets are accurately represented in the test data.
Collapse
Affiliation(s)
- Khaled Bousabarah
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Stereotactic and Functional Neurosurgery, Cologne, Germany.
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Diagnostic and Interventional Radiology, Cologne, Germany.
| | - Maximilian Ruge
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Stereotactic and Functional Neurosurgery, Cologne, Germany
| | - Julia-Sarita Brand
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Stereotactic and Functional Neurosurgery, Cologne, Germany
| | - Mauritius Hoevels
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Stereotactic and Functional Neurosurgery, Cologne, Germany
| | - Daniel Rueß
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Stereotactic and Functional Neurosurgery, Cologne, Germany
| | - Jan Borggrefe
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Diagnostic and Interventional Radiology, Cologne, Germany
| | - Nils Große Hokamp
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Diagnostic and Interventional Radiology, Cologne, Germany
| | - Veerle Visser-Vandewalle
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Stereotactic and Functional Neurosurgery, Cologne, Germany
| | - David Maintz
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Diagnostic and Interventional Radiology, Cologne, Germany
| | - Harald Treuer
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Stereotactic and Functional Neurosurgery, Cologne, Germany
| | - Martin Kocher
- University of Cologne, Faculty of Medicine and University Hospital Cologne, Department of Stereotactic and Functional Neurosurgery, Cologne, Germany
| |
Collapse
|
22
|
Editorial commentary to " 18F-Fluorocholine PET uptake correlates with pathologic evidence of recurrent tumor after stereotactic radiosurgery for brain metastases" by Grkovski and colleagues. Eur J Nucl Med Mol Imaging 2019; 47:1340-1341. [PMID: 31872279 DOI: 10.1007/s00259-019-04651-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/25/2022]
|
23
|
Boon IS, Au Yong TPT, Boon CS. Assessing the Role of Artificial Intelligence (AI) in Clinical Oncology: Utility of Machine Learning in Radiotherapy Target Volume Delineation. MEDICINES (BASEL, SWITZERLAND) 2018; 5:E131. [PMID: 30544901 PMCID: PMC6313566 DOI: 10.3390/medicines5040131] [Citation(s) in RCA: 46] [Impact Index Per Article: 7.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 11/30/2018] [Revised: 12/04/2018] [Accepted: 12/07/2018] [Indexed: 12/16/2022]
Abstract
The fields of radiotherapy and clinical oncology have been rapidly changed by the advances of technology. Improvement in computer processing power and imaging quality heralded precision radiotherapy allowing radiotherapy to be delivered efficiently, safely and effectively for patient benefit. Artificial intelligence (AI) is an emerging field of computer science which uses computer models and algorithms to replicate human-like intelligence and perform specific tasks which offers a huge potential to healthcare. We reviewed and presented the history, evolution and advancement in the fields of radiotherapy, clinical oncology and machine learning. Radiotherapy target delineation is a complex task of outlining tumour and organ at risks volumes to allow accurate delivery of radiotherapy. We discussed the radiotherapy planning, treatment delivery and reviewed how technology can help with this challenging process. We explored the evidence and clinical application of machine learning to radiotherapy. We concluded on the challenges, possible future directions and potential collaborations to achieve better outcome for cancer patients.
Collapse
Affiliation(s)
- Ian S Boon
- Department of Clinical Oncology, Leeds Cancer Centre, St James's Institute of Oncology, Leeds Teaching Hospitals NHS Trust, Leeds LS9 7TF, UK.
| | - Tracy P T Au Yong
- Department of Radiology, Worcestershire Acute Hospitals NHS Trust, Worcester WR5 1DD, UK.
| | - Cheng S Boon
- Worcestershire Oncology Centre, Worcestershire Acute Hospitals NHS Trust, Worcester WR5 1DD, UK.
| |
Collapse
|