Triple-View Feature Learning for Medical Image Segmentation

TL;DR

TriSegNet is a semi-supervised medical image segmentation framework that leverages triple-view feature learning and confidence-based label refinement to improve segmentation accuracy with limited labeled data.

Contribution

It introduces a novel triple-view architecture with confidence voting and tailored loss functions for semi-supervised medical image segmentation.

Findings

Outperforms existing semi-supervised methods on benchmark datasets.

Effective across ultrasound, CT, MRI, and histology images.

Improves segmentation confidence through multi-view learning.

Abstract

Deep learning models, e.g. supervised Encoder-Decoder style networks, exhibit promising performance in medical image segmentation, but come with a high labelling cost. We propose TriSegNet, a semi-supervised semantic segmentation framework. It uses triple-view feature learning on a limited amount of labelled data and a large amount of unlabeled data. The triple-view architecture consists of three pixel-level classifiers and a low-level shared-weight learning module. The model is first initialized with labelled data. Label processing, including data perturbation, confidence label voting and unconfident label detection for annotation, enables the model to train on labelled and unlabeled data simultaneously. The confidence of each model gets improved through the other two views of the feature learning. This process is repeated until each model reaches the same confidence level as its counterparts. This strategy enables triple-view learning of generic medical image datasets. Bespoke overlap-based and boundary-based loss functions are tailored to the different stages of the training. The segmentation results are evaluated on four publicly available benchmark datasets including Ultrasound, CT, MRI, and Histology images. Repeated experiments demonstrate the effectiveness of the proposed network compared against other semi-supervised algorithms, across a large set of evaluation measures.