Meta Corrupted Pixels Mining for Medical Image Segmentation

TL;DR

This paper introduces Meta Corrupted Pixels Mining (MCPM), a novel method that automatically identifies and down-weights corrupted annotations in medical image segmentation, improving training on datasets with noisy labels.

Contribution

The paper proposes a meta mask network that estimates pixel importance to handle corrupted annotations, enhancing segmentation accuracy in noisy medical datasets.

Findings

Outperforms state-of-the-art methods on LIDC-IDRI and LiTS datasets.

Effectively identifies and down-weights corrupted pixels during training.

Improves segmentation performance with noisy annotations.

Abstract

Deep neural networks have achieved satisfactory performance in piles of medical image analysis tasks. However the training of deep neural network requires a large amount of samples with high-quality annotations. In medical image segmentation, it is very laborious and expensive to acquire precise pixel-level annotations. Aiming at training deep segmentation models on datasets with probably corrupted annotations, we propose a novel Meta Corrupted Pixels Mining (MCPM) method based on a simple meta mask network. Our method is targeted at automatically estimate a weighting map to evaluate the importance of every pixel in the learning of segmentation network. The meta mask network which regards the loss value map of the predicted segmentation results as input, is capable of identifying out corrupted layers and allocating small weights to them. An alternative algorithm is adopted to train the segmentation network and the meta mask network, simultaneously. Extensive experimental results on LIDC-IDRI and LiTS datasets show that our method outperforms state-of-the-art approaches which are devised for coping with corrupted annotations.