# TractOracle: towards an anatomically-informed reward function for RL-based tractography

arXiv: https://arxiv.org/abs/2403.17845 · HTML: https://mlsanity.com/paper/2403.17845/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in RL for clinical treatment and surgery, one of 15 directions in Deep reinforcement learning methods.
- Builds on 5 of its 32 references that are in the map.
- First posted on arXiv 26 Mar 2024.

## Placed in

- Deep reinforcement learning methods / [RL for clinical treatment and surgery](https://mlsanity.com/deeprl/rl-for-clinical-treatment-and-surgery/)

## Builds on (most cited first, 5 of 5)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Soft Actor-Critic Algorithms and Applications](https://mlsanity.com/paper/1812.05905/) (2018)
- [Filtering in tractography using autoencoders (FINTA)](https://mlsanity.com/paper/2010.04007/) (2020)
- [DeepTract: A Probabilistic Deep Learning Framework for White Matter Fiber Tractography](https://mlsanity.com/paper/1812.05129/) (2018)
- [What Matters in Reinforcement Learning for Tractography](https://mlsanity.com/paper/2305.09041/) (2023)

## Built on by (most cited first, 5 of 5)

- [Deep Learning-Based Diffusion MRI Tractography: Integrating Spatial and Anatomical Information](https://mlsanity.com/paper/2503.03329/) (2025)
- [DDTracking: A Deep Generative Framework for Diffusion MRI Tractography with Streamline Local-Global Spatiotemporal Modeling](https://mlsanity.com/paper/2508.04568/) (2025)
- [Exploring the robustness of TractOracle methods in RL-based tractography](https://mlsanity.com/paper/2507.11486/) (2025)
- [GenTract: Generative Global Tractography](https://mlsanity.com/paper/2511.13183/) (2025)
- [TractRLFusion: A GPT-Based Multi-Critic Policy Fusion Framework for Fiber Tractography](https://mlsanity.com/paper/2601.13897/) (2026)

## Abstract (arXiv)

Reinforcement learning (RL)-based tractography is a competitive alternative to machine learning and classical tractography algorithms due to its high anatomical accuracy obtained without the need for any annotated data. However, the reward functions so far used to train RL agents do not encapsulate anatomical knowledge which causes agents to generate spurious false positives tracts. In this paper, we propose a new RL tractography system, TractOracle, which relies on a reward network trained for streamline classification. This network is used both as a reward function during training as well as a mean for stopping the tracking process early and thus reduce the number of false positive streamlines. This makes our system a unique method that evaluates and reconstructs WM streamlines at the same time. We report an improvement of true positive ratios by almost 20% and a reduction of 3x of false positive ratios on one dataset and an increase between 2x and 7x in the number true positive streamlines on another dataset.