Deep Reinforcement Learning for Small Bowel Path Tracking using Different Types of Annotations

TL;DR

This paper introduces a deep reinforcement learning approach for small bowel path tracking that effectively uses datasets with various annotation types, reducing the need for costly ground-truth path annotations.

Contribution

It presents a novel environment design enabling training with both segmentation and path annotations, improving usability and reducing annotation costs.

Findings

Method successfully utilizes weakly annotated data.

Achieves accurate small bowel path tracking.

Reduces reliance on costly ground-truth paths.

Abstract

Small bowel path tracking is a challenging problem considering its many folds and contact along its course. For the same reason, it is very costly to achieve the ground-truth (GT) path of the small bowel in 3D. In this work, we propose to train a deep reinforcement learning tracker using datasets with different types of annotations. Specifically, we utilize CT scans that have only GT small bowel segmentation as well as ones with the GT path. It is enabled by designing a unique environment that is compatible for both, including a reward definable even without the GT path. The performed experiments proved the validity of the proposed method. The proposed method holds a high degree of usability in this problem by being able to utilize the scans with weak annotations, and thus by possibly reducing the required annotation cost.