Related Experiment Video
Updated: Dec 18, 2025
06:48
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
Published on: January 7, 2019
9.3K
Mini Review: Deep Learning for Atrial Segmentation From Late Gadolinium-Enhanced MRIs
Kevin Jamart1, Zhaohan Xiong1, Gonzalo D Maso Talou1
1Auckland Bioengineering Institute, The University of Auckland, Auckland, New Zealand.
Frontiers in Cardiovascular Medicine
|June 13, 2020
Summary
Deep learning significantly improves atrial segmentation for cardiac arrhythmia diagnosis. This review analyzes AI methods for reliable and accurate 3D atrial reconstruction from MRI scans.
Area of Science:
- Medical imaging analysis
- Artificial intelligence in cardiology
Background:
- Manual segmentation of human atria for diagnosing atrial fibrillation is time-consuming and error-prone.
- Accurate atrial segmentation is crucial for effective treatment of cardiac arrhythmias.
Purpose of the Study:
- To review state-of-the-art deep learning approaches for segmenting human atria from late gadolinium-enhanced MRIs.
- To provide insights into reliable and high-performance deep learning methods for atrial segmentation.
Main Methods:
- In-depth review of deep learning architectures and techniques used in atrial segmentation.
- Analysis of approaches from the 2018 Atrial Segmentation Challenge.
Main Results:
- Deep learning methods achieved high accuracy (>90% Dice score) in atrial segmentation.
- Significant discrepancies exist between different deep learning approaches.
Conclusions:
- Further research is needed to determine optimal deep learning architectures for reliable and accurate atrial segmentation.
- Addressing current hindrances is key to advancing AI-driven cardiac imaging analysis.
