Curriculum-Guided Myocardial Scar Segmentation for Ischemic and Non-ischemic Cardiomyopathy

2026-03-30

AI Summary

This work addresses the challenge of unreliable myocardial scar segmentation in late gadolinium-enhanced cardiac magnetic resonance (LGE-CMR) images, which arises from low contrast, suboptimal imaging conditions, and inconsistent annotations. To overcome these issues, the authors propose a curriculum learning–based progressive training strategy that begins with high-confidence, clearly defined scar regions and gradually incorporates lower-confidence or ambiguous samples. This approach effectively mitigates label uncertainty and weak scar representations, alleviating the limitations of conventional methods in modeling sparse or diffuse scars. Experimental results demonstrate that the proposed strategy significantly improves both segmentation accuracy and inter-case consistency, with particularly notable gains over standard training baselines in cases featuring small scar burdens or diffuse scar distributions.

Abstract

Identification and quantification of myocardial scar is important for diagnosis and prognosis of cardiovascular diseases. However, reliable scar segmentation from Late Gadolinium Enhancement Cardiac Magnetic Resonance (LGE-CMR) images remains a challenge due to variations in contrast enhancement across patients, suboptimal imaging conditions such as post contrast washout, and inconsistencies in ground truth annotations on diffuse scars caused by inter observer variability. In this work, we propose a curriculum learning-based framework designed to improve segmentation performance under these challenging conditions. The method introduces a progressive training strategy that guides the model from high-confidence, clearly defined scar regions to low confidence or visually ambiguous samples with limited scar burden. By structuring the learning process in this manner, the network develops robustness to uncertain labels and subtle scar appearances that are often underrepresented in conventional training pipelines. Experimental results show that the proposed approach enhances segmentation accuracy and consistency, particularly for cases with minimal or diffuse scar, outperforming standard training baselines. This strategy provides a principled way to leverage imperfect data for improved myocardial scar quantification in clinical applications. Our code is publicly available on GitHub.

Problem

• myocardial scar segmentation
• Late Gadolinium Enhancement Cardiac Magnetic Resonance
• inter-observer variability
• diffuse scar
• image annotation inconsistency

Innovation

• curriculum learning
• myocardial scar segmentation
• LGE-CMR
• diffuse scar
• robust segmentation