SM2C: Boost the Semi-supervised Segmentation for Medical Image by using Meta Pseudo Labels and Mixed Images

TL;DR

SM2C enhances semi-supervised medical image segmentation by employing multi-strategy data augmentation techniques, leading to better generalization and improved performance on benchmark datasets.

Contribution

The paper introduces SM2C, a novel semi-supervised segmentation method using scaling-up images, multi-class mixing, and shape jittering to improve learning from limited labeled data.

Findings

Significant performance improvements over state-of-the-art methods.

Effective in handling complex organ and lesion shapes.

Proven on three benchmark datasets.

Abstract

Recently, machine learning-based semantic segmentation algorithms have demonstrated their potential to accurately segment regions and contours in medical images, allowing the precise location of anatomical structures and abnormalities. Although medical images are difficult to acquire and annotate, semi-supervised learning methods are efficient in dealing with the scarcity of labeled data. However, overfitting is almost inevitable due to the limited images for training. Furthermore, the intricate shapes of organs and lesions in medical images introduce additional complexity in different cases, preventing networks from acquiring a strong ability to generalize. To this end, we introduce a novel method called Scaling-up Mix with Multi-Class (SM2C). This method uses three strategies - scaling-up image size, multi-class mixing, and object shape jittering - to improve the ability to learn semantic features within medical images. By diversifying the shape of the segmentation objects and enriching the semantic information within each sample, the SM2C demonstrates its potential, especially in the training of unlabelled data. Extensive experiments demonstrate the effectiveness of the SM2C on three benchmark medical image segmentation datasets. The proposed framework shows significant improvements over state-of-the-art counterparts.