Segmentation of Ischemic Stroke Lesions using Transfer Learning on Multi-sequence MRI

TL;DR

This paper introduces a transfer learning-based framework for automatic segmentation of ischemic stroke lesions across multiple MRI sequences, achieving high accuracy and efficiency compared to manual methods.

Contribution

It presents a novel multi-sequence MRI segmentation method using transfer learning with Res-Unet and a Majority Voting Classifier, improving accuracy over existing approaches.

Findings

Achieved a Dice score of 80.5% on ISLES 2015 dataset.

Demonstrated the benefits of transfer learning in medical image segmentation.

Integrated a Majority Voting Classifier for comprehensive 3D segmentation.

Abstract

The accurate understanding of ischemic stroke lesions is critical for efficient therapy and prognosis of stroke patients. Magnetic resonance imaging (MRI) is sensitive to acute ischemic stroke and is a common diagnostic method for stroke. However, manual lesion segmentation performed by experts is tedious, time-consuming, and prone to observer inconsistency. Automatic medical image analysis methods have been proposed to overcome this challenge. However, previous approaches have relied on hand-crafted features that may not capture the irregular and physiologically complex shapes of ischemic stroke lesions. In this study, we present a novel framework for quickly and automatically segmenting ischemic stroke lesions on various MRI sequences, including T1-weighted, T2-weighted, DWI, and FLAIR. The proposed methodology is validated on the ISLES 2015 Brain Stroke sequence dataset, where we trained our model using the Res-Unet architecture twice: first, with pre-existing weights, and then without, to explore the benefits of transfer learning. Evaluation metrics, including the Dice score and sensitivity, were computed across 3D volumes. Finally, a Majority Voting Classifier was integrated to amalgamate the outcomes from each axis, resulting in a comprehensive segmentation method. Our efforts culminated in achieving a Dice score of 80.5\% and an accuracy of 74.03\%, showcasing the efficacy of our segmentation approach.