Learning-based dose prediction for pancreatic stereotactic body radiation therapy using dual pyramid adversarial network.
Treatment planning for pancreatic cancer stereotactic body radiation therapy (SBRT) is very challenging owing to vast spatial variations and close proximity of many organs-at-risk. Recently, deep learning (DL)-based methods have been applied in dose prediction tasks of various treatment sites with the aim of relieving planning challenges. Limited investigations, however, have been reported on DL-based dose prediction for pancreatic cancer SBRT. This study aims to further current knowledge in DL-based dose prediction tasks by implementing and demonstrating the feasibility of a new dual pyramid networks (DPNs) integrated DL-model for predicting dose distributions of pancreatic SBRT. The proposed framework is composed of four parts: CT-only feature pyramid network (FPN), contour-only FPN, late fusion network and adversarial network. Combination of mean absolute error, gradient difference error, histogram matching, and adversarial loss is used for supervision. The performance of proposed model was demonstrated for pancreatic cancer SBRT plans for doses prescribed between 33 and 50 Gy over five fractions to up to three planning target volumes (PTV). Five-fold cross-validation was performed on 30 patients and another 20 patients were used as holdout tests. Predicted plans were compared with clinically approved plans through dose-volume parameters. The proposed framework was able to predict 87% and 91% of clinically relevant dose parameters for cross-validation sets and holdout sets, respectively, without any significant differences (P > 0.05). For the parameters with significant differences (P < 0.05), the predicted doses to the duodenum and spinal cord were lower in comparison to the clinical plans. The proposed model was also able to predict the intentional hotspots as feature characteristics of SBRT plans. Overall, the proposed model was able to predict doses to cases with both single and multiple PTVs. In conclusion, the DPN integrated DL-model demonstrated good overall plan quality and dose prediction accuracy for pancreatic cancer SBRT.