Tied Prototype Model for Few-Shot Medical Image Segmentation

TL;DR

The paper introduces the Tied Prototype Model (TPM), a novel approach for few-shot medical image segmentation that improves upon existing methods by modeling multiple prototypes and adaptively handling background variability.

Contribution

TPM reformulates ADNet with tied prototypes, enabling multi-prototype and multi-class segmentation, and incorporates adaptive thresholds based on class priors for better accuracy.

Findings

TPM outperforms previous methods in segmentation accuracy.

The model effectively handles background variability.

Adaptive thresholds improve segmentation performance.

Abstract

Common prototype-based medical image few-shot segmentation (FSS) methods model foreground and background classes using class-specific prototypes. However, given the high variability of the background, a more promising direction is to focus solely on foreground modeling, treating the background as an anomaly -- an approach introduced by ADNet. Yet, ADNet faces three key limitations: dependence on a single prototype per class, a focus on binary classification, and fixed thresholds that fail to adapt to patient and organ variability. To address these shortcomings, we propose the Tied Prototype Model (TPM), a principled reformulation of ADNet with tied prototype locations for foreground and background distributions. Building on its probabilistic foundation, TPM naturally extends to multiple prototypes and multi-class segmentation while effectively separating non-typical background features. Notably, both extensions lead to improved segmentation accuracy. Finally, we leverage naturally occurring class priors to define an ideal target for adaptive thresholds, boosting segmentation performance. Taken together, TPM provides a fresh perspective on prototype-based FSS for medical image segmentation. The code can be found at https://github.com/hjk92g/TPM-FSS.