Dose Prediction with Deep Learning for Prostate Cancer Radiation Therapy: Model Adaptation to Different Treatment Planning Practices

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.

[1]  Kaiming He,et al.  Group Normalization , 2018, ECCV.

[2]  Steve B. Jiang,et al.  A feasibility study for predicting optimal radiation therapy dose distributions of prostate cancer patients from patient anatomy using deep learning , 2017, Scientific Reports.

[3]  M Monz,et al.  Pareto navigation—algorithmic foundation of interactive multi-criteria IMRT planning , 2008, Physics in medicine and biology.

[4]  Russell H. Taylor,et al.  Patient geometry-driven information retrieval for IMRT treatment plan quality control. , 2009, Medical physics.

[5]  Satomi Shiraishi,et al.  Knowledge-based prediction of three-dimensional dose distributions for external beam radiotherapy. , 2015, Medical physics.

[6]  Ben J M Heijmen,et al.  Automatically configuring the reference point method for automated multi-objective treatment planning , 2019, Physics in medicine and biology.

[7]  Ke Sheng,et al.  Deterministic direct aperture optimization using multiphase piecewise constant segmentation , 2017, Medical physics.

[8]  Sasa Mutic,et al.  Predicting dose-volume histograms for organs-at-risk in IMRT planning. , 2012, Medical physics.

[9]  Russell H. Taylor,et al.  Data-driven approach to generating achievable dose-volume histogram objectives in intensity-modulated radiotherapy planning. , 2011, International journal of radiation oncology, biology, physics.

[10]  Steve B. Jiang,et al.  Three-Dimensional Dose Prediction for Lung IMRT Patients with Deep Neural Networks: Robust Learning from Heterogeneous Beam Configurations , 2018, ArXiv.

[11]  David L Craft,et al.  Approximating convex pareto surfaces in multiobjective radiotherapy planning. , 2006, Medical physics.

[12]  T. Bortfeld,et al.  Improved planning time and plan quality through multicriteria optimization for intensity-modulated radiotherapy. , 2012, International journal of radiation oncology, biology, physics.

[13]  Ben J M Heijmen,et al.  Automatic configuration of the reference point method for fully automated multi‐objective treatment planning applied to oropharyngeal cancer , 2020, Medical physics.

[14]  Russell H. Taylor,et al.  Using overlap volume histogram and IMRT plan data to guide and automate VMAT planning: a head-and-neck case study. , 2013, Medical physics.

[15]  Terry M. Peters,et al.  Medical Image Computing and Computer Assisted Intervention – MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part IV , 2019, MICCAI.

[16]  Sebastiaan Breedveld,et al.  Multi-criteria optimization and decision-making in radiotherapy , 2019 .

[17]  Y. Ge,et al.  Quantitative analysis of the factors which affect the interpatient organ-at-risk dose sparing variation in IMRT plans. , 2012, Medical physics.

[18]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[19]  Frederik Maes,et al.  Feasibility of CT-Only 3D Dose Prediction for VMAT Prostate Plans Using Deep Learning , 2019, AIRT@MICCAI.

[20]  Binbin Wu,et al.  Improved robotic stereotactic body radiation therapy plan quality and planning efficacy for organ-confined prostate cancer utilizing overlap-volume histogram-driven planning methodology. , 2014, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[21]  D. Low,et al.  Experience-based quality control of clinical intensity-modulated radiotherapy planning. , 2011, International Journal of Radiation Oncology, Biology, Physics.