Communicative Reinforcement Learning Agents for Landmark Detection in Brain Images

TL;DR

This paper introduces a communicative multi-agent reinforcement learning system for automatic landmark detection in 3D brain images, demonstrating improved accuracy over single-agent methods in MRI and ultrasound datasets.

Contribution

The novel C-MARL system enables multiple agents to learn explicit and implicit communication for improved landmark detection in medical images.

Findings

Multi-agent communication improves detection accuracy.

The approach outperforms single-agent methods.

Effective in both MRI and ultrasound datasets.

Abstract

Accurate detection of anatomical landmarks is an essential step in several medical imaging tasks. We propose a novel communicative multi-agent reinforcement learning (C-MARL) system to automatically detect landmarks in 3D brain images. C-MARL enables the agents to learn explicit communication channels, as well as implicit communication signals by sharing certain weights of the architecture among all the agents. The proposed approach is evaluated on two brain imaging datasets from adult magnetic resonance imaging (MRI) and fetal ultrasound scans. Our experiments show that involving multiple cooperating agents by learning their communication with each other outperforms previous approaches using single agents.