Two Deep Learning Approaches for Automated Segmentation of Left Ventricle in Cine Cardiac MRI
Image generated by Gemini AI
Researchers have developed two deep learning models, LNU-Net and IBU-Net, for left ventricle segmentation in short-axis cine MRI images. LNU-Net enhances the U-Net with layer normalization, while IBU-Net combines instance and batch normalization. Tested on a dataset of 805 MRI images from 45 patients, both models significantly improved segmentation accuracy, outperforming existing methods in terms of the dice coefficient and average perpendicular distance. This advancement could enhance clinical diagnostics and quantification in cardiology.
Novel Deep Learning Architectures Enhance Left Ventricle Segmentation in MRI
Recent advancements in deep learning have led to the development of two new architectures, LNU-Net and IBU-Net, aimed at improving left ventricle (LV) segmentation from cine cardiac MRI images. Both models demonstrate significant improvements in segmentation accuracy compared to traditional methods.
Researchers designed LNU-Net based on layer normalization and IBU-Net utilizing instance-batch normalization. Each model includes an up-sampling path to ensure precise localization, crucial for accurate segmentation outcomes.
Evaluation and Results
The models were evaluated on a dataset of 805 MRI images from 45 patients. Both LNU-Net and IBU-Net exceeded the performance metrics of existing state-of-the-art segmentation methods, achieving higher dice coefficients and reduced average perpendicular distances, indicating enhanced accuracy in delineating the left ventricle.
Related Topics:
📰 Original Source: https://arxiv.org/abs/2601.00794v1
All rights and credit belong to the original publisher.