# DoseGAN: a generative adversarial network for synthetic dose prediction using attention-gated discrimination and generation

## TL;DR

This paper introduces DoseGAN, a new AI model that improves radiation dose prediction for cancer treatment by focusing on important anatomical areas.

## Contribution

The novel contribution is an attention-gated GAN architecture that enhances dose prediction accuracy in stereotactic body radiotherapy.

## Key findings

- DoseGAN outperformed existing methods in predicting realistic volumetric dosimetry.
- The model showed statistically significant improvements in PTV and rectum dose metrics.
- Attention mechanisms helped focus learning on relevant anatomical regions.

## Abstract

Deep learning algorithms have recently been developed that utilize patient anatomy and raw imaging information to predict radiation dose, as a means to increase treatment planning efficiency and improve radiotherapy plan quality. Current state-of-the-art techniques rely on convolutional neural networks (CNNs) that use pixel-to-pixel loss to update network parameters. However, stereotactic body radiotherapy (SBRT) dose is often heterogeneous, making it difficult to model using pixel-level loss. Generative adversarial networks (GANs) utilize adversarial learning that incorporates image-level loss and is better suited to learn from heterogeneous labels. However, GANs are difficult to train and rely on compromised architectures to facilitate convergence. This study suggests an attention-gated generative adversarial network (DoseGAN) to improve learning, increase model complexity, and reduce network redundancy by focusing on relevant anatomy. DoseGAN was compared to alternative state-of-the-art dose prediction algorithms using heterogeneity index, conformity index, and various dosimetric parameters. All algorithms were trained, validated, and tested using 141 prostate SBRT patients. DoseGAN was able to predict more realistic volumetric dosimetry compared to all other algorithms and achieved statistically significant improvement compared to all alternative algorithms for the V100 and V120 of the PTV, V60 of the rectum, and heterogeneity index.