[Development of a Deep Learning Model for Judging Late Gadolinium-enhancement in Cardiac MRI]

Area of Science:

• Cardiovascular Imaging
• Artificial Intelligence in Medicine
• Medical Image Analysis

Context:

• Late gadolinium-enhancement (LGE) imaging is crucial for assessing myocardial scar in cardiac magnetic resonance imaging (MRI).
• Accurate interpretation of LGE images is vital for patient management.
• Deep learning offers potential for automating image analysis tasks.

Purpose:

• To evaluate the efficacy of a deep learning model in identifying contrast-enhanced myocardium on cardiac MRI LGE images.
• To assess the model's performance in distinguishing between healthy and infarcted myocardial tissue.

Summary:

• A convolutional neural network model was trained on 3312 LGE cardiac MRI images derived from 174 patient scans.
• Data augmentation techniques were employed to enhance the training dataset.
• The model achieved a specificity of 100.0% and an accuracy of 96.7% on test data.

Impact:

• The developed deep learning model shows high prediction accuracy for contrast-enhanced myocardium detection.
• This technology has the potential to aid cardiologists in LGE image interpretation.
• Automated analysis could improve efficiency and consistency in cardiac MRI assessments.