Abstract:Semi-supervised segmentation is highly significant in 3D medical image segmentation. The typical solutions adopt a teacher-student dual-model architecture, and they constrain the two models' decision consistency on the same segmentation task. However, the scarcity of medical samples can lower the diversity of tasks, reducing the effectiveness of consistency constraint. The issue can further worsen as the weights of the models gradually become synchronized. In this work, we have proposed to construct diverse joint-tasks using masked image modelling for enhancing the reliability of the consistency constraint, and develop a novel architecture consisting of a single teacher but multiple students to enjoy the additional knowledge decoupled from the synchronized weights. Specifically, the teacher and student models 'see' varied randomly-masked versions of an input, and are trained to segment the same targets but reconstruct different missing regions concurrently. Such joint-task of segmentation and reconstruction can have the two learners capture related but complementary features to derive instructive knowledge when constraining their consistency. Moreover, two extra students join the original one to perform an inter-student learning. The three students share the same encoding but different decoding designs, and learn decoupled knowledge by constraining their mutual consistencies, preventing themselves from suboptimally converging to the biased predictions of the dictatorial teacher. Experimental on four medical datasets show that our approach performs better than six mainstream semi-supervised methods. Particularly, our approach achieves at least 0.61% and 0.36% higher Dice and Jaccard values, respectively, than the most competitive approach on our in-house dataset. The code will be released at http://github.com/zxmboshi/DDL.

An Efficient Semi-Supervised Framework with Multi-Task and Curriculum Learning for Medical Image Segmentation

MMSeg: A novel multi-task learning framework for class imbalance and label scarcity in medical image segmentation

Semisupervised Medical Image Segmentation through Prototype-Based Mutual Consistency Learning

Curriculum Consistency Learning and Multi-Scale Contrastive Constraint in Semi-Supervised Medical Image Segmentation

Semi-supervised Medical Image Segmentation Based on Multi-scale Knowledge Discovery and Multi-task Ensemble.

Multi-task Attention-Based Semi-supervised Learning for Medical Image Segmentation

Multi-task contrastive learning for semi-supervised medical image segmentation with multi-scale uncertainty estimation

Shape and boundary-aware multi-branch model for semi-supervised medical image segmentation

Dual-Task Mutual Learning for Semi-Supervised Medical Image Segmentation

Semi-supervised medical image segmentation via a tripled-uncertainty guided mean teacher model with contrastive learning

Semi-Supervised Medical Image Segmentation Based on Deep Consistent Collaborative Learning

Multi-Level Contrastive Student-Teacher Structure for Semi-Supervised Medical Image Segmentation

Transformation Consistent Self-ensembling Model for Semi-supervised Medical Image Segmentation

Semi-Supervised 3D Medical Image Segmentation Based on Dual-Task Consistent Joint Learning and Task-Level Regularization.

A Teacher-Student Framework for Semi-supervised Medical Image Segmentation From Mixed Supervision

Inherent Consistent Learning for Accurate Semi-supervised Medical Image Segmentation

Semi-MedSeq: Semi-supervised Semantic Segmentation for Medical Image Sequences.

Transformation-Consistent Self-Ensembling Model for Semisupervised Medical Image Segmentation

Robust Semi-supervised Multimodal Medical Image Segmentation via Cross Modality Collaboration

Tripled-Uncertainty Guided Mean Teacher Model for Semi-supervised Medical Image Segmentation

Robust Semi-supervised 3D Medical Image Segmentation with Diverse Joint-task Learning and Decoupled Inter-student Learning