Roya Norouzi Kandalan1, Dan Nguyen2, Nima Hassan Rezaeian3, Ana M Barragán-Montero4, Sebastiaan Breedveld5, Kamesh Namuduri6, Steve Jiang7, Mu-Han Lin8. 1. Medical Artificial Intelligence and Automation (MAIA) Laboratory, Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, USA; Department of Electrical Engineering, University of North Texas, Denton, USA. 2. Medical Artificial Intelligence and Automation (MAIA) Laboratory, Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, USA. 3. Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, USA. 4. Molecular Imaging, Radiotherapy and Oncology (MIRO), UCLouvain, Brussels, Belgium. 5. Department of Radiation Oncology, Erasmus University Medical Center - Cancer Institute, Rotterdam, The Netherlands. 6. Department of Electrical Engineering, University of North Texas, Denton, USA. 7. Medical Artificial Intelligence and Automation (MAIA) Laboratory, Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, USA. Electronic address: Steve.Jiang@UTSouthwestern.edu. 8. Medical Artificial Intelligence and Automation (MAIA) Laboratory, Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, USA. Electronic address: Mu-Han.Lin@UTSouthwestern.edu.
Abstract
PURPOSE: This work aims to study the generalizability of a pre-developed deep learning (DL) dose prediction model for volumetric modulated arc therapy (VMAT) for prostate cancer and to adapt the model, via transfer learning with minimal input data, to three different internal treatment planning styles and one external institution planning style. METHODS: We built the source model with planning data from 108 patients previously treated with VMAT for prostate cancer. For the transfer learning, we selected patient cases planned with three different styles, 14-29 cases per style, in the same institution and 20 cases treated in a different institution to adapt the source model to four target models in total. We compared the dose distributions predicted by the source model and the target models with the corresponding clinical plan dose used for patient treatments and quantified the improvement in the prediction quality for the target models over the source model using the Dice similarity coefficients (DSC) of 0% to 100% isodose volumes and the dose-volume-histogram (DVH) parameters of the planning target volume and the organs-at-risk. RESULTS: The source model accurately predicts dose distributions for plans generated in the same source style, but performs sub-optimally for the three different internal and one external target styles, with the mean DSC ranging between 0.81-0.94 and 0.82-0.91 for the internal and the external styles, respectively. With transfer learning, the target model predictions improved the mean DSC to 0.88-0.95 and 0.92-0.96 for the internal and the external styles, respectively. Target model predictions significantly improved the accuracy of the DVH parameter predictions to within 1.6%. CONCLUSION: We demonstrated the problem of model generalizability for DL-based dose prediction and the feasibility of using transfer learning to solve this problem. With 14-29 cases per style, we successfully adapted the source model into several different practice styles. This indicates a realistic way forward to widespread clinical implementation of DL-based dose prediction.
PURPOSE: This work aims to study the generalizability of a pre-developed deep learning (DL) dose prediction model for volumetric modulated arc therapy (VMAT) for prostate cancer and to adapt the model, via transfer learning with minimal input data, to three different internal treatment planning styles and one external institution planning style. METHODS: We built the source model with planning data from 108 patients previously treated with VMAT for prostate cancer. For the transfer learning, we selected patient cases planned with three different styles, 14-29 cases per style, in the same institution and 20 cases treated in a different institution to adapt the source model to four target models in total. We compared the dose distributions predicted by the source model and the target models with the corresponding clinical plan dose used for patient treatments and quantified the improvement in the prediction quality for the target models over the source model using the Dice similarity coefficients (DSC) of 0% to 100% isodose volumes and the dose-volume-histogram (DVH) parameters of the planning target volume and the organs-at-risk. RESULTS: The source model accurately predicts dose distributions for plans generated in the same source style, but performs sub-optimally for the three different internal and one external target styles, with the mean DSC ranging between 0.81-0.94 and 0.82-0.91 for the internal and the external styles, respectively. With transfer learning, the target model predictions improved the mean DSC to 0.88-0.95 and 0.92-0.96 for the internal and the external styles, respectively. Target model predictions significantly improved the accuracy of the DVH parameter predictions to within 1.6%. CONCLUSION: We demonstrated the problem of model generalizability for DL-based dose prediction and the feasibility of using transfer learning to solve this problem. With 14-29 cases per style, we successfully adapted the source model into several different practice styles. This indicates a realistic way forward to widespread clinical implementation of DL-based dose prediction.
Authors: Binbin Wu; Francesco Ricchetti; Giuseppe Sanguineti; Michael Kazhdan; Patricio Simari; Robert Jacques; Russell Taylor; Todd McNutt Journal: Int J Radiat Oncol Biol Phys Date: 2010-08-26 Impact factor: 7.038
Authors: Binbin Wu; Francesco Ricchetti; Giuseppe Sanguineti; Misha Kazhdan; Patricio Simari; Ming Chuang; Russell Taylor; Robert Jacques; Todd McNutt Journal: Med Phys Date: 2009-12 Impact factor: 4.071
Authors: Ana Barragán-Montero; Umair Javaid; Gilmer Valdés; Dan Nguyen; Paul Desbordes; Benoit Macq; Siri Willems; Liesbeth Vandewinckele; Mats Holmström; Fredrik Löfman; Steven Michiels; Kevin Souris; Edmond Sterpin; John A Lee Journal: Phys Med Date: 2021-05-09 Impact factor: 2.685
Authors: Jianhui Ma; Dan Nguyen; Ti Bai; Michael Folkerts; Xun Jia; Weiguo Lu; Linghong Zhou; Steve Jiang Journal: Med Phys Date: 2021-07-11 Impact factor: 4.506
Authors: David J Sher; Andrew Godley; Yang Park; Colin Carpenter; Marc Nash; Hasti Hesami; Xinran Zhong; Mu-Han Lin Journal: Clin Transl Radiat Oncol Date: 2021-05-20