Related Experiment Video
Updated: Dec 15, 2025
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
43.3K
Automatic left ventricle segmentation in short-axis MRI using deep convolutional neural networks and central-line
Lipeng Xie1, Yi Song2, Qiang Chen3
1School of Information and Communication Engineering, University of Electronic Science and Technology of China, China.
Computers in Biology and Medicine
|July 14, 2020
Summary
This study introduces an automated method for segmenting the left ventricle (LV) in cardiac MRI using a convolutional neural network (CNN) and level set approach. The novel technique improves diagnostic speed and accuracy for cardiovascular diseases.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Cardiovascular Medicine
Background:
- Accurate left ventricle (LV) segmentation in cardiac magnetic resonance imaging (MRI) is crucial for diagnosing cardiovascular diseases.
- Manual segmentation is time-consuming and prone to inter-observer variability.
Purpose of the Study:
- To develop an automated LV segmentation method to reduce diagnostic time.
- To improve the accuracy and consistency of LV segmentation in cardiac MRI.
Main Methods:
- Integration of a convolutional neural network (CNN) for myocardial central-line detection.
- A novel central-line guided level set approach (CGLS) for myocardium delineation.
- Incorporation of the myocardial central-line as a constraint in the level set energy formulation.
Main Results:
- Achieved epicardium and endocardium perpendicular distances of 1.74 mm and 2.06 mm on the MICCAI 2009 dataset.
- Obtained LV and myocardium Dice metrics of 0.955 and 0.853 at end-diastole on the ACDC MICCAI 2017 dataset.
- Demonstrated superior performance compared to state-of-the-art methods, with good agreement to manual segmentation.
Conclusions:
- The proposed automated LV segmentation method effectively reduces segmentation time.
- The CGLS approach enhances anatomical accuracy and consistency in LV segmentation.
- This method shows significant potential for clinical application in cardiovascular disease diagnosis.
