Mine yOur owN Anatomy: Revisiting Medical Image Segmentation with Extremely Limited Labels

by   Chenyu You, et al.

Recent studies on contrastive learning have achieved remarkable performance solely by leveraging few labels in the context of medical image segmentation. Existing methods mainly focus on instance discrimination and invariant mapping. However, they face three common pitfalls: (1) tailness: medical image data usually follows an implicit long-tail class distribution. Blindly leveraging all pixels in training hence can lead to the data imbalance issues, and cause deteriorated performance; (2) consistency: it remains unclear whether a segmentation model has learned meaningful and yet consistent anatomical features due to the intra-class variations between different anatomical features; and (3) diversity: the intra-slice correlations within the entire dataset have received significantly less attention. This motivates us to seek a principled approach for strategically making use of the dataset itself to discover similar yet distinct samples from different anatomical views. In this paper, we introduce a novel semi-supervised medical image segmentation framework termed Mine yOur owN Anatomy (MONA), and make three contributions. First, prior work argues that every pixel equally matters to the model training; we observe empirically that this alone is unlikely to define meaningful anatomical features, mainly due to lacking the supervision signal. We show two simple solutions towards learning invariances - through the use of stronger data augmentations and nearest neighbors. Second, we construct a set of objectives that encourage the model to be capable of decomposing medical images into a collection of anatomical features in an unsupervised manner. Lastly, our extensive results on three benchmark datasets with different labeled settings validate the effectiveness of our proposed MONA which achieves new state-of-the-art under different labeled settings.


page 2

page 7

page 9

page 10

page 23

page 24


Bootstrapping Semi-supervised Medical Image Segmentation with Anatomical-aware Contrastive Distillation

Contrastive learning has shown great promise over annotation scarcity pr...

Rethinking Semi-Supervised Medical Image Segmentation: A Variance-Reduction Perspective

For medical image segmentation, contrastive learning is the dominant pra...

Self-Ensembling Contrastive Learning for Semi-Supervised Medical Image Segmentation

Deep learning has demonstrated significant improvements in medical image...

A biologically constrained model of the whole basal ganglia addressing the paradoxes of connections and selection

The basal ganglia nuclei form a complex network of nuclei often assumed ...

ACTION++: Improving Semi-supervised Medical Image Segmentation with Adaptive Anatomical Contrast

Medical data often exhibits long-tail distributions with heavy class imb...

Exploring Smoothness and Class-Separation for Semi-supervised Medical Image Segmentation

Semi-supervised segmentation remains challenging in medical imaging sinc...

Slice and Dice: A Physicalization Workflow for Anatomical Edutainment

During the last decades, anatomy has become an interesting topic in educ...

Please sign up or login with your details

Forgot password? Click here to reset