TractRLFusion: A GPT-Based Multi-Critic Policy Fusion Framework for Fiber Tractography

TL;DR

TractRLFusion introduces a GPT-based multi-critic policy fusion framework that significantly improves white matter tract reconstruction accuracy and reliability in brain connectivity mapping, advancing non-invasive neuroimaging techniques.

Contribution

The paper presents a novel GPT-based policy fusion framework with a two-stage training process and multi-critic fine-tuning, enhancing tractography performance over existing methods.

Findings

Outperforms classical and DRL methods in accuracy

Demonstrates robustness across multiple datasets

Improves anatomical reliability of tract reconstructions

Abstract

Tractography plays a pivotal role in the non-invasive reconstruction of white matter fiber pathways, providing vital information on brain connectivity and supporting precise neurosurgical planning. Although traditional methods relied mainly on classical deterministic and probabilistic approaches, recent progress has benefited from supervised deep learning (DL) and deep reinforcement learning (DRL) to improve tract reconstruction. A persistent challenge in tractography is accurately reconstructing white matter tracts while minimizing spurious connections. To address this, we propose TractRLFusion, a novel GPT-based policy fusion framework that integrates multiple RL policies through a data-driven fusion strategy. Our method employs a two-stage training data selection process for effective policy fusion, followed by a multi-critic fine-tuning phase to enhance robustness and generalization. Experiments on HCP, ISMRM, and TractoInferno datasets demonstrate that TractRLFusion outperforms individual RL policies as well as state-of-the-art classical and DRL methods in accuracy and anatomical reliability.