ScarNet: A Novel Foundation Model for Automated Myocardial Scar Quantification from LGE in Cardiac MRI

2025-01-02

AI Summary

Manual segmentation of myocardial scar in late gadolinium enhancement MRI (LGE-MRI) is time-consuming and subjective, while existing automated methods suffer from limited accuracy on low-contrast, ill-defined boundary images. To address this, we propose a hybrid architecture integrating the MedSAM vision transformer encoder with a custom spatial-channel joint-attention U-Net decoder, specifically designed to model weak myocardial-scar boundary features under expert annotation supervision. Evaluated on an independent test set of 184 patients, our method achieves a Dice coefficient of 0.912 (IQR: 0.863–0.944), mean volume bias of −0.63%, and a low coefficient of variation (4.3%), significantly outperforming both MedSAM and nnU-Net. This work represents the first application of MedSAM to cardiac scar segmentation and introduces structural innovations enabling highly accurate and robust fully automatic quantitative analysis of left ventricular scar.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Background: Late Gadolinium Enhancement (LGE) imaging is the gold standard for assessing myocardial fibrosis and scarring, with left ventricular (LV) LGE extent predicting major adverse cardiac events (MACE). Despite its importance, routine LGE-based LV scar quantification is hindered by labor-intensive manual segmentation and inter-observer variability. Methods: We propose ScarNet, a hybrid model combining a transformer-based encoder from the Medical Segment Anything Model (MedSAM) with a convolution-based U-Net decoder, enhanced by tailored attention blocks. ScarNet was trained on 552 ischemic cardiomyopathy patients with expert segmentations of myocardial and scar boundaries and tested on 184 separate patients. Results: ScarNet achieved robust scar segmentation in 184 test patients, yielding a median Dice score of 0.912 (IQR: 0.863--0.944), significantly outperforming MedSAM (median Dice = 0.046, IQR: 0.043--0.047) and nnU-Net (median Dice = 0.638, IQR: 0.604--0.661). ScarNet demonstrated lower bias (-0.63%) and coefficient of variation (4.3%) compared to MedSAM (bias: -13.31%, CoV: 130.3%) and nnU-Net (bias: -2.46%, CoV: 20.3%). In Monte Carlo simulations with noise perturbations, ScarNet achieved significantly higher scar Dice (0.892 pm 0.053, CoV = 5.9%) than MedSAM (0.048 pm 0.112, CoV = 233.3%) and nnU-Net (0.615 pm 0.537, CoV = 28.7%). Conclusion: ScarNet outperformed MedSAM and nnU-Net in accurately segmenting myocardial and scar boundaries in LGE images. The model exhibited robust performance across diverse image qualities and scar patterns.

Problem

• Late Gadolinium Enhancement (LGE)
• Automated Scar Size Calculation
• Cardiac MRI

Innovation

• ScarNet
• Cardiac MRI
• Automated Scar Quantification