TY - JOUR
T1 - Identification of treatment error types for lung cancer patients using convolutional neural networks and EPID dosimetry
AU - Wolfs, Cecile J. A.
AU - Canters, Richard A. M.
AU - Verhaegen, Frank
N1 - Funding Information:
This study was funded by Varian Medical Systems (project: Decision DGRT-I). The Titan Xp GPUs used for this research were donated by the NVIDIA Corporation. The authors would like to thank G. van Iersel and R. de Vries for their contribution to the mechanical error simulations, Dr. B. van der Heyden for technical support with GPU computing, Dr. E. de Jong for fruitful discussion, and Dr. C. Baltes and Dr. S. Scheib from Varian Medical Systems for commenting on the manuscript.
Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2020/12
Y1 - 2020/12
N2 - Background/Purpose: Electronic portal imaging device (EPID) dosimetry aims to detect treatment errors, potentially leading to treatment adaptation. Clinically used threshold classification methods for detecting errors lead to loss of information (from multi-dimensional EPID data to a few numbers) and cannot be used for identifying causes of errors. Advanced classification methods, such as deep learning, can use all available information. In this study, convolutional neural networks (CNNs) were trained to detect and identify error type and magnitude of simulated treatment errors in lung cancer patients. The purpose of this simulation study is to provide a proof-of-concept of CNNs for error identification using EPID dosimetry in an in vivo scenario.Materials and methods: Clinically realistic ranges of anatomical changes, positioning errors and mechanical errors were simulated for lung cancer patients. Predicted portal dose images (PDIs) containing errors were compared to error-free PDIs using the widely used gamma analysis. CNNs were trained to classify errors using 2D gamma maps. Three classification levels were assessed: Level 1 (main error type, e.g., anatomical change), Level 2 (error subtype, e.g., tumor regression) and Level 3 (error magnitude, e.g., >50% tumor regression).Results: CNNs showed good performance for all classification levels (training/test accuracy 99.5%/96.1%, 92.5%/86.8%, 82.0%/72.9%). For Level 3, overfitting became more apparent.Conclusion: This simulation study indicates that deep learning is a promising powerful tool for identifying types and magnitude of treatment errors with EPID dosimetry, providing additional information not currently available from EPID dosimetry. This is a first step towards rapid, automated models for identification of treatment errors using EPID dosimetry. (C) 2020 Elsevier B.V. All rights reserved.
AB - Background/Purpose: Electronic portal imaging device (EPID) dosimetry aims to detect treatment errors, potentially leading to treatment adaptation. Clinically used threshold classification methods for detecting errors lead to loss of information (from multi-dimensional EPID data to a few numbers) and cannot be used for identifying causes of errors. Advanced classification methods, such as deep learning, can use all available information. In this study, convolutional neural networks (CNNs) were trained to detect and identify error type and magnitude of simulated treatment errors in lung cancer patients. The purpose of this simulation study is to provide a proof-of-concept of CNNs for error identification using EPID dosimetry in an in vivo scenario.Materials and methods: Clinically realistic ranges of anatomical changes, positioning errors and mechanical errors were simulated for lung cancer patients. Predicted portal dose images (PDIs) containing errors were compared to error-free PDIs using the widely used gamma analysis. CNNs were trained to classify errors using 2D gamma maps. Three classification levels were assessed: Level 1 (main error type, e.g., anatomical change), Level 2 (error subtype, e.g., tumor regression) and Level 3 (error magnitude, e.g., >50% tumor regression).Results: CNNs showed good performance for all classification levels (training/test accuracy 99.5%/96.1%, 92.5%/86.8%, 82.0%/72.9%). For Level 3, overfitting became more apparent.Conclusion: This simulation study indicates that deep learning is a promising powerful tool for identifying types and magnitude of treatment errors with EPID dosimetry, providing additional information not currently available from EPID dosimetry. This is a first step towards rapid, automated models for identification of treatment errors using EPID dosimetry. (C) 2020 Elsevier B.V. All rights reserved.
KW - EPID dosimetry
KW - In vivo dosimetry
KW - Treatment verification
KW - Error identification
KW - Artificial intelligence
KW - Deep learning
KW - PORTAL DOSIMETRY
KW - DELIVERY
U2 - 10.1016/j.radonc.2020.09.048
DO - 10.1016/j.radonc.2020.09.048
M3 - Article
C2 - 33011206
SN - 0167-8140
VL - 153
SP - 243
EP - 249
JO - Radiotherapy and Oncology
JF - Radiotherapy and Oncology
ER -