Related Experiment Video
Updated: Jun 26, 2026
10:25
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
MRI acute/sub-acute ischemic stroke segmentation with deep learning: A comprehensive review
Mohammad Alshurbaji1, Maregu Assefa2, Said Boumaraf1
1Department of Computer Science, Khalifa University of Science and Technology, Abu Dhabi, United Arab Emirates.
International Review of Cell and Molecular Biology
|June 24, 2026
Summary
Deep learning models show promise for segmenting ischemic stroke lesions in MRI scans. This review analyzes recent advancements, challenges, and future directions for improved stroke lesion segmentation accuracy.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Medicine
- Neurology
Background:
- Ischemic stroke lesion segmentation from MRI is crucial for diagnosis and treatment.
- Deep learning (DL) offers advanced capabilities for medical image analysis.
- Accurate segmentation remains a challenge due to lesion variability and image quality.
Purpose of the Study:
- To provide a comprehensive review of DL methodologies for ischemic stroke lesion segmentation in MRI.
- To analyze advancements, challenges, and research gaps in the field since 2020.
- To establish a performance benchmark and serve as a reference for researchers.
Main Methods:
- Systematic review of studies from 2020 onwards utilizing DL for stroke lesion segmentation.
- Analysis of various DL models, MRI modalities, datasets, and preprocessing techniques.
- Compilation of evaluation metrics, loss functions, and augmentation strategies.
Main Results:
- Identified key advancements in DL model effectiveness for stroke lesion segmentation.
- Documented a wide range of datasets, MRI modalities, and technical approaches.
- Highlighted common challenges and identified research gaps.
Conclusions:
- Deep learning models have significantly advanced stroke lesion segmentation from MRI.
- Further research is needed to address existing challenges and enhance model accuracy.
- This review provides a comprehensive resource for future research and development.
