Self-Supervised Few-Shot Learning for Ischemic Stroke Lesion Segmentation

TL;DR

This paper introduces a self-supervised few-shot learning method for ischemic stroke lesion segmentation that requires only one annotated sample, utilizing color-coded parametric maps to improve performance.

Contribution

It proposes a novel self-supervised training mechanism tailored for ischemic stroke lesion segmentation in a few-shot setting, reducing the need for extensive annotated data.

Findings

Achieves an average Dice score of 0.58 with a single annotated sample.

Demonstrates significant performance improvements over traditional methods in few-shot scenarios.

Utilizes color-coded parametric maps from CT Perfusion scans for effective learning.

Abstract

Precise ischemic lesion segmentation plays an essential role in improving diagnosis and treatment planning for ischemic stroke, one of the prevalent diseases with the highest mortality rate. While numerous deep neural network approaches have recently been proposed to tackle this problem, these methods require large amounts of annotated regions during training, which can be impractical in the medical domain where annotated data is scarce. As a remedy, we present a prototypical few-shot segmentation approach for ischemic lesion segmentation using only one annotated sample during training. The proposed approach leverages a novel self-supervised training mechanism that is tailored to the task of ischemic stroke lesion segmentation by exploiting color-coded parametric maps generated from Computed Tomography Perfusion scans. We illustrate the benefits of our proposed training mechanism, leading to considerable improvements in performance in the few-shot setting. Given a single annotated patient, an average Dice score of 0.58 is achieved for the segmentation of ischemic lesions.