Multimodal Volume-Aware Detection and Segmentation for Brain Metastases Radiosurgery

TL;DR

This paper introduces a deep learning system that automates detection and segmentation of brain metastases in multimodal images, improving SRS treatment planning by incorporating lesion volume information.

Contribution

It presents a novel volume-aware Dice loss and ensemble neural networks for enhanced detection and segmentation of small, multiple brain metastases.

Findings

Surpasses current benchmark levels in detection accuracy.

Demonstrates reliable AI-assisted SRS planning for multiple metastases.

Improves segmentation of small brain metastases.

Abstract

Stereotactic radiosurgery (SRS), which delivers high doses of irradiation in a single or few shots to small targets, has been a standard of care for brain metastases. While very effective, SRS currently requires manually intensive delineation of tumors. In this work, we present a deep learning approach for automated detection and segmentation of brain metastases using multimodal imaging and ensemble neural networks. In order to address small and multiple brain metastases, we further propose a volume-aware Dice loss which optimizes model performance using the information of lesion size. This work surpasses current benchmark levels and demonstrates a reliable AI-assisted system for SRS treatment planning for multiple brain metastases.