Fully Differentiable Bidirectional Dual-Task Synergistic Learning for Semi-Supervised 3D Medical Image Segmentation

TL;DR

This paper introduces a fully differentiable bidirectional dual-task learning framework that enhances semi-supervised 3D medical image segmentation by enabling online bidirectional collaboration between related tasks, leading to improved performance.

Contribution

The proposed DBiSL framework uniquely integrates bidirectional task collaboration with differentiable design, advancing semi-supervised learning in medical image segmentation.

Findings

Achieves state-of-the-art results on benchmark datasets

Demonstrates the effectiveness of bidirectional task synergy

Provides a new architectural foundation for dual-task semi-supervised learning

Abstract

Semi-supervised learning relaxes the need of large pixel-wise labeled datasets for image segmentation by leveraging unlabeled data. The scarcity of high-quality labeled data remains a major challenge in medical image analysis due to the high annotation costs and the need for specialized clinical expertise. Semi-supervised learning has demonstrated significant potential in addressing this bottleneck, with pseudo-labeling and consistency regularization emerging as two predominant paradigms. Dual-task collaborative learning, an emerging consistency-aware paradigm, seeks to derive supplementary supervision by establishing prediction consistency between related tasks. However, current methodologies are limited to unidirectional interaction mechanisms (typically regression-to-segmentation), as segmentation results can only be transformed into regression outputs in an offline manner, thereby failing to fully exploit the potential benefits of online bidirectional cross-task collaboration. Thus, we propose a fully Differentiable Bidirectional Synergistic Learning (DBiSL) framework, which seamlessly integrates and enhances four critical SSL components: supervised learning, consistency regularization, pseudo-supervised learning, and uncertainty estimation. Experiments on two benchmark datasets demonstrate our method's state-of-the-art performance. Beyond technical contributions, this work provides new insights into unified SSL framework design and establishes a new architectural foundation for dual-task-driven SSL, while offering a generic multitask learning framework applicable to broader computer vision applications. The code will be released on github upon acceptance.