1
|
Strickler EAT, Thomas J, Thomas JP, Benjamin B, Shamsuddin R. Exploring a global interpretation mechanism for deep learning networks when predicting sepsis. Sci Rep 2023; 13:3067. [PMID: 36810645 PMCID: PMC9945464 DOI: 10.1038/s41598-023-30091-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/15/2022] [Accepted: 02/15/2023] [Indexed: 02/24/2023] Open
Abstract
The purpose of this study is to identify additional clinical features for sepsis detection through the use of a novel mechanism for interpreting black-box machine learning models trained and to provide a suitable evaluation for the mechanism. We use the publicly available dataset from the 2019 PhysioNet Challenge. It has around 40,000 Intensive Care Unit (ICU) patients with 40 physiological variables. Using Long Short-Term Memory (LSTM) as the representative black-box machine learning model, we adapted the Multi-set Classifier to globally interpret the black-box model for concepts it learned about sepsis. To identify relevant features, the result is compared against: (i) features used by a computational sepsis expert, (ii) clinical features from clinical collaborators, (iii) academic features from literature, and (iv) significant features from statistical hypothesis testing. Random Forest was found to be the computational sepsis expert because it had high accuracies for solving both the detection and early detection, and a high degree of overlap with clinical and literature features. Using the proposed interpretation mechanism and the dataset, we identified 17 features that the LSTM used for sepsis classification, 11 of which overlaps with the top 20 features from the Random Forest model, 10 with academic features and 5 with clinical features. Clinical opinion suggests, 3 LSTM features have strong correlation with some clinical features that were not identified by the mechanism. We also found that age, chloride ion concentration, pH and oxygen saturation should be investigated further for connection with developing sepsis. Interpretation mechanisms can bolster the incorporation of state-of-the-art machine learning models into clinical decision support systems, and might help clinicians to address the issue of early sepsis detection. The promising results from this study warrants further investigation into creation of new and improvement of existing interpretation mechanisms for black-box models, and into clinical features that are currently not used in clinical assessment of sepsis.
Collapse
Affiliation(s)
- Ethan A T Strickler
- Physics and Mathematics, East Central University, PO Box 385, Ada, OK, 74820, USA
| | - Joshua Thomas
- Department of Internal Medicine, Rush University Medical Center, 1700 W Van Buren St, 5th Floor, Chicago, IL, 60612, USA
| | - Johnson P Thomas
- Oklahoma State University, 201 Math and Science Building, Stillwater, OK, 74078, USA
| | - Bruce Benjamin
- School of Biomedical Sciences, Center for Health Sciences, 1111 W. 17th st., Tulsa, OK, 74107, USA
| | - Rittika Shamsuddin
- Oklahoma State University, 212 Math and Science Building, Stillwater, OK, 74078, USA.
| |
Collapse
|
2
|
Torshabi A. Investigation the efficacy of fuzzy logic implementation at image-guided radiotherapy. JOURNAL OF MEDICAL SIGNALS & SENSORS 2022; 12:163-170. [PMID: 35755973 PMCID: PMC9215832 DOI: 10.4103/jmss.jmss_76_20] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/04/2020] [Revised: 04/05/2021] [Accepted: 10/24/2021] [Indexed: 11/04/2022]
|
3
|
Mylonas A, Booth J, Nguyen DT. A review of artificial intelligence applications for motion tracking in radiotherapy. J Med Imaging Radiat Oncol 2021; 65:596-611. [PMID: 34288501 DOI: 10.1111/1754-9485.13285] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/05/2021] [Accepted: 06/29/2021] [Indexed: 11/28/2022]
Abstract
During radiotherapy, the organs and tumour move as a result of the dynamic nature of the body; this is known as intrafraction motion. Intrafraction motion can result in tumour underdose and healthy tissue overdose, thereby reducing the effectiveness of the treatment while increasing toxicity to the patients. There is a growing appreciation of intrafraction target motion management by the radiation oncology community. Real-time image-guided radiation therapy (IGRT) can track the target and account for the motion, improving the radiation dose to the tumour and reducing the dose to healthy tissue. Recently, artificial intelligence (AI)-based approaches have been applied to motion management and have shown great potential. In this review, four main categories of motion management using AI are summarised: marker-based tracking, markerless tracking, full anatomy monitoring and motion prediction. Marker-based and markerless tracking approaches focus on tracking the individual target throughout the treatment. Full anatomy algorithms monitor for intrafraction changes in the full anatomy within the field of view. Motion prediction algorithms can be used to account for the latencies due to the time for the system to localise, process and act.
Collapse
Affiliation(s)
- Adam Mylonas
- ACRF Image X Institute, Faculty of Medicine and Health, The University of Sydney, Sydney, New South Wales, Australia.,School of Biomedical Engineering, University of Technology Sydney, Sydney, New South Wales, Australia
| | - Jeremy Booth
- Northern Sydney Cancer Centre, Royal North Shore Hospital, St Leonards, New South Wales, Australia.,Institute of Medical Physics, School of Physics, The University of Sydney, Sydney, New South Wales, Australia
| | - Doan Trang Nguyen
- ACRF Image X Institute, Faculty of Medicine and Health, The University of Sydney, Sydney, New South Wales, Australia.,School of Biomedical Engineering, University of Technology Sydney, Sydney, New South Wales, Australia.,Northern Sydney Cancer Centre, Royal North Shore Hospital, St Leonards, New South Wales, Australia
| |
Collapse
|
4
|
Field M, Hardcastle N, Jameson M, Aherne N, Holloway L. Machine learning applications in radiation oncology. PHYSICS & IMAGING IN RADIATION ONCOLOGY 2021; 19:13-24. [PMID: 34307915 PMCID: PMC8295850 DOI: 10.1016/j.phro.2021.05.007] [Citation(s) in RCA: 21] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 11/06/2020] [Revised: 05/19/2021] [Accepted: 05/22/2021] [Indexed: 12/23/2022]
Abstract
Machine learning technology has a growing impact on radiation oncology with an increasing presence in research and industry. The prevalence of diverse data including 3D imaging and the 3D radiation dose delivery presents potential for future automation and scope for treatment improvements for cancer patients. Harnessing this potential requires standardization of tools and data, and focused collaboration between fields of expertise. The rapid advancement of radiation oncology treatment technologies presents opportunities for machine learning integration with investments targeted towards data quality, data extraction, software, and engagement with clinical expertise. In this review, we provide an overview of machine learning concepts before reviewing advances in applying machine learning to radiation oncology and integrating these techniques into the radiation oncology workflows. Several key areas are outlined in the radiation oncology workflow where machine learning has been applied and where it can have a significant impact in terms of efficiency, consistency in treatment and overall treatment outcomes. This review highlights that machine learning has key early applications in radiation oncology due to the repetitive nature of many tasks that also currently have human review. Standardized data management of routinely collected imaging and radiation dose data are also highlighted as enabling engagement in research utilizing machine learning and the ability integrate these technologies into clinical workflow to benefit patients. Physicists need to be part of the conversation to facilitate this technical integration.
Collapse
Affiliation(s)
- Matthew Field
- South Western Sydney Clinical School, Faculty of Medicine, University of New South Wales, Sydney, NSW, Australia.,Ingham Institute for Applied Medical Research, Sydney, NSW, Australia
| | - Nicholas Hardcastle
- Physical Sciences, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia.,Centre for Medical Radiation Physics, University of Wollongong, Wollongong, NSW, Australia
| | - Michael Jameson
- GenesisCare, Alexandria, NSW, Australia.,St Vincent's Clinical School, Faculty of Medicine, University of New South Wales, Australia
| | - Noel Aherne
- Mid North Coast Cancer Institute, NSW, Australia.,Rural Clinical School, Faculty of Medicine, University of New South Wales, Sydney, NSW, Australia
| | - Lois Holloway
- South Western Sydney Clinical School, Faculty of Medicine, University of New South Wales, Sydney, NSW, Australia.,Ingham Institute for Applied Medical Research, Sydney, NSW, Australia.,Cancer Therapy Centre, Liverpool Hospital, Sydney, NSW, Australia.,Centre for Medical Radiation Physics, University of Wollongong, Wollongong, NSW, Australia
| |
Collapse
|
5
|
Fast and accurate online sequential learning of respiratory motion with random convolution nodes for radiotherapy applications. Appl Soft Comput 2020. [DOI: 10.1016/j.asoc.2020.106528] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/13/2023]
|
6
|
Vicente E, Modiri A, Yu KC, Wibowo H, Yan Y, Timmerman R, Sawant A. Accounting for respiratory motion in small serial structures during radiotherapy planning: proof of concept in virtual bronchoscopy-guided lung functional avoidance radiotherapy. Phys Med Biol 2019; 64:225011. [PMID: 31665703 DOI: 10.1088/1361-6560/ab52a1] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/25/2022]
Abstract
Respiratory motion management techniques in radiotherapy (RT) planning are primarily focused on maintaining tumor target coverage. An inadequately addressed need is accounting for motion in dosimetric estimations in smaller serial structures. Accurate dose estimations in such structures are more sensitive to motion because respiration can cause them to move completely in or out of a high dose-gradient field. In this work, we study three motion management strategies (m1-m3) to find an accurate method to estimate the dosimetry in airways. To validate these methods, we generated a 'ground truth' digital breathing model based on a 4DCT scan from a lung stereotactic ablative radiotherapy (SAbR) patient. We simulated 225 breathing cycles with ±10% perturbations in amplitude, respiratory period, and time per respiratory phase. A high-resolution breath-hold CT (BHCT) was also acquired and used with a research virtual bronchoscopy software to autosegment 239 airways. Contours for planning target volume (PTV) and organs at risk (OARs) were defined on the maximum intensity projection of the 4DCT (CTMIP) and transferred to the average of the 10 4DCT phases (CTAVG). To design the motion management methods, the RT plan was recreated using different images and structure definitions. Methods m1 and m2 recreated the plan using the CTAVG image. In method m1, airways were deformed to the CTAVG. In m2, airways were deformed to each of the 4DCT phases, and union structures were transferred onto the CTAVG. In m3, the RT plan was recreated on each of the 10 phases, and the dose distribution from each phase was deformed to the BHCT and summed. Dose errors (mean [min, max]) in airways were: m1: 21% (0.001%, 93%); m2: 45% (0.1%, 179%); and m3: 4% (0.006%, 14%). Our work suggests that accurate dose estimation in moving small serial structures requires customized motion management techniques (like m3 in this work) rather than current clinical and investigational approaches.
Collapse
Affiliation(s)
- Esther Vicente
- Department of Radiation Oncology, University of Maryland School of Medicine, Baltimore, MD 21201, United States of America. Author to whom correspondence should be addressed
| | | | | | | | | | | | | |
Collapse
|
7
|
Lin H, Zou W, Li T, Feigenberg SJ, Teo BKK, Dong L. A Super-Learner Model for Tumor Motion Prediction and Management in Radiation Therapy: Development and Feasibility Evaluation. Sci Rep 2019; 9:14868. [PMID: 31619736 PMCID: PMC6795883 DOI: 10.1038/s41598-019-51338-y] [Citation(s) in RCA: 18] [Impact Index Per Article: 3.6] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/22/2019] [Accepted: 09/25/2019] [Indexed: 12/25/2022] Open
Abstract
In cancer radiation therapy, large tumor motion due to respiration can lead to uncertainties in tumor target delineation and treatment delivery, thus making active motion management an essential step in thoracic and abdominal tumor treatment. In current practice, patients with tumor motion may be required to receive two sets of CT scans – the initial free-breathing 4-dimensional CT (4DCT) scan for tumor motion estimation and a second CT scan under appropriate motion management such as breath-hold or abdominal compression. The aim of this study is to assess the feasibility of a predictive model for tumor motion estimation in three-dimensional space based on machine learning algorithms. The model was developed based on sixteen imaging features extracted from non-4D diagnostic CT images and eleven clinical features extracted from the Electronic Health Record (EHR) database of 150 patients to characterize the lung tumor motion. A super-learner model was trained to combine four base machine learning models including the Random Forest, Multi-Layer Perceptron, LightGBM and XGBoost, the hyper-parameters of which were also optimized to obtain the best performance. The outputs of the super-learner model consist of tumor motion predictions in the Superior-Inferior (SI), Anterior-Posterior (AP) and Left-Right (LR) directions, and were compared against tumor motions measured in the free-breathing 4DCT scans. The accuracy of predictions was evaluated using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) through ten rounds of independent tests. The MAE and RMSE of predictions in the SI direction were 1.23 mm and 1.70 mm; the MAE and RMSE of predictions in the AP direction were 0.81 mm and 1.19 mm, and the MAE and RMSE of predictions in the LR direction were 0.70 mm and 0.95 mm. In addition, the relative feature importance analysis demonstrated that the imaging features are of great importance in the tumor motion prediction compared to the clinical features. Our findings indicate that a super-learner model can accurately predict tumor motion ranges as measured in the 4DCT, and could provide a machine learning framework to assist radiation oncologists in determining the active motion management strategy for patients with large tumor motion.
Collapse
Affiliation(s)
- Hui Lin
- University of Pennsylvania, Department of Radiation Oncology, Pennsylvania, 19104, United States
| | - Wei Zou
- University of Pennsylvania, Department of Radiation Oncology, Pennsylvania, 19104, United States
| | - Taoran Li
- University of Pennsylvania, Department of Radiation Oncology, Pennsylvania, 19104, United States
| | - Steven J Feigenberg
- University of Pennsylvania, Department of Radiation Oncology, Pennsylvania, 19104, United States
| | - Boon-Keng K Teo
- University of Pennsylvania, Department of Radiation Oncology, Pennsylvania, 19104, United States
| | - Lei Dong
- University of Pennsylvania, Department of Radiation Oncology, Pennsylvania, 19104, United States.
| |
Collapse
|
8
|
Real-time control of respiratory motion: Beyond radiation therapy. Phys Med 2019; 66:104-112. [PMID: 31586767 DOI: 10.1016/j.ejmp.2019.09.241] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/23/2019] [Revised: 09/23/2019] [Accepted: 09/26/2019] [Indexed: 12/16/2022] Open
Abstract
Motion management in radiation oncology is an important aspect of modern treatment planning and delivery. Special attention has been paid to control respiratory motion in recent years. However, other medical procedures related to both diagnosis and treatment are likely to benefit from the explicit control of breathing motion. Quantitative imaging - including increasingly important tools in radiology and nuclear medicine - is among the fields where a rapid development of motion control is most likely, due to the need for quantification accuracy. Emerging treatment modalities like focussed-ultrasound tumor ablation are also likely to benefit from a significant evolution of motion control in the near future. In the present article an overview of available respiratory motion systems along with ongoing research in this area is provided. Furthermore, an attempt is made to envision some of the most expected developments in this field in the near future.
Collapse
|
9
|
Knybel L, Penhaker M, Proto A, Otahal B, Nowakova J, Cvek J, Filipova B, Selamat A. Accuracy analysis of the dose delivery process while using the Xsight® Spine Tracking technology. Biomed Phys Eng Express 2018. [DOI: 10.1088/2057-1976/aae8d7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/11/2022]
|