VI Seminar #82:Tied Prototype Model for Few-Shot Medical Image Segmentation

Presented by Hyeongji Kim, Postdoc Candidate at UiT The Arctic University of Norway

The presentation is based on a presentation given at MICCAI 2025 in South Korea.

Full paper: https://papers.miccai.org/miccai-2025/0929-Paper2931.html

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.

In this talk, I will briefly explain the prototype-based FFS approach and present our work, the Tied Prototype Model (TPM), which overcomes the limitations of ADNet and provides a new perspective on prototype learning for medical image segmentation.

‍

‍