Staged Voxel-Level Deep Reinforcement Learning for 3D Medical Image Segmentation with Noisy Annotations

YuYang Fu, XiuZhen Guo, Ji Shi · 2026

Read Staged Voxel-Level Deep Reinforcement Learning for 3D Medical Image Segmentation with Noisy Annotations by YuYang Fu, XiuZhen Guo, Ji Shi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-03875.txt.

Description

Yuyang Fu, Xiuzhen Guo, and Ji Shi design a medical image segmentation setup that stays robust when its training annotations contain errors. Accurate segmentation of organs in 3D scans usually demands large, carefully labeled datasets, but complex anatomy and varied annotators make noisy labels common. Their setup treats segmentation as a staged, voxel-level decision process trained with deep reinforcement learning.

The dynamic iterative strategy mimics a human annotator who corrects labeling mistakes using prior knowledge as work proceeds. Casting noisy annotations as a voxel-dependent problem lets the system down-weight or repair bad labels with no manual help. Robust training matters clinically: a segmentation model that quietly absorbs label errors can misjudge organ size, so methods that resist annotation noise widen where automated contouring can be trusted.

Catalog record

CREATORYuYang Fu, XiuZhen Guo, Ji Shi
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE46 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-03875

Browse the full library · About the Tower of Babel