AS-NeSt: A Novel 3D Deep Learning Model for Radiation Therapy Dose Distribution Prediction in Esophageal Cancer Treatment With Multiple Prescriptions

Abstract

Purpose: Implementing artificial intelligence (AI) technologies allows for the accurate prediction of radiotherapy dose distributions, enhancing treatment planning efficiency. However, esophageal cancers present unique challenges due to tumor complexity and diverse prescription types. Additionally, limited data availability hampers the effectiveness of existing AI models. This study developed a deep learning model, trained on a diverse dataset of esophageal cancer prescriptions, to improve dose prediction accuracy. Methods and Materials: We retrospectively collected data from 530 esophageal cancer patients, including single-target and simultaneous integrated boost (SIB) prescriptions, for model building. The proposed Asymmetric ResNeSt (AS-NeSt) model features novel 3D ResNeSt blocks and an asymmetric architecture. We constructed a loss function targeting global and local doses and validated the model's performance against existing alternatives. Model assisted experiments were used to validate its clinical benefits. Results: The AS-NeSt model maintained an absolute prediction error below 5% for each dosimetric metric. The average dice similarity coefficient (DSC) for isodose volumes was 0.93. The model achieved an average relative prediction error of 2.02%, statistically lower than HD-Unet (4.17%), DoseNet (2.35%), and DCNN (3.65%). It also demonstrated significantly fewer parameters and shorter prediction times. Clinically, the AS-NeSt model raised physicians’ ability to accurately pre-assess appropriate treatment methods before planning from 95.24% to 100%, reduced planning time by over 61% for junior dosimetrists and 52% for senior dosimetrists, and decreased both inter- and intra-dosimetrist discrepancies by more than 50%. Conclusion: The AS-NeSt model, developed with innovative 3D ResNeSt blocks and an asymmetric encoder-decoder structure, has been validated using clinical esophageal cancer patient data. It accurately predicts 3D dose distributions for various prescriptions, including SIB, showing potential to improve the management of esophageal cancer treatment in a clinical setting.