Deep learning-based segmentation of T1 and T2 cardiac MRI maps for automated disease detection

2025-07-01

AI Summary

Traditional cardiac T1/T2 mapping analysis suffers from high inter-observer variability in manual segmentation and loss of myocardial tissue heterogeneity due to reliance on single-mean intensity thresholds. To address this, we propose a deep learning–driven fully automatic framework for simultaneous myocardium and left ventricular blood pool segmentation. Crucially, we introduce— for the first time in cardiac MRI mapping—a quantile-based statistical characterization (e.g., median, interquartile range) of myocardial relaxation distributions, replacing conventional mean-based summarization. These quantile features are integrated with spatial and intensity descriptors and classified via random forest. In disease detection, our method achieves an F1-score of 92.7% (p < 0.001), significantly outperforming mean-only baselines. The segmentation model attains a Dice coefficient of 85.4%, exceeding inter-observer agreement. By moving beyond the mean-centric paradigm, our approach enables fine-grained, quantitative pathological assessment of myocardial tissue and provides a generalizable technical pipeline for advanced cardiac MRI analysis.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Qualitative Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Objectives Parametric tissue mapping enables quantitative cardiac tissue characterization but is limited by inter-observer variability during manual delineation. Traditional approaches relying on average relaxation values and single cutoffs may oversimplify myocardial complexity. This study evaluates whether deep learning (DL) can achieve segmentation accuracy comparable to inter-observer variability, explores the utility of statistical features beyond mean T1/T2 values, and assesses whether machine learning (ML) combining multiple features enhances disease detection. Materials & Methods T1 and T2 maps were manually segmented. The test subset was independently annotated by two observers, and inter-observer variability was assessed. A DL model was trained to segment left ventricle blood pool and myocardium. Average (A), lower quartile (LQ), median (M), and upper quartile (UQ) were computed for the myocardial pixels and employed in classification by applying cutoffs or in ML. Dice similarity coefficient (DICE) and mean absolute percentage error evaluated segmentation performance. Bland-Altman plots assessed inter-user and model-observer agreement. Receiver operating characteristic analysis determined optimal cutoffs. Pearson correlation compared features from model and manual segmentations. F1-score, precision, and recall evaluated classification performance. Wilcoxon test assessed differences between classification methods, with p < 0.05 considered statistically significant. Results 144 subjects were split into training (100), validation (15) and evaluation (29) subsets. Segmentation model achieved a DICE of 85.4%, surpassing inter-observer agreement. Random forest applied to all features increased F1-score (92.7%, p < 0.001). Conclusion DL facilitates segmentation of T1/ T2 maps. Combining multiple features with ML improves disease detection.

Problem

• Automate cardiac MRI segmentation to reduce manual variability
• Enhance disease detection using multiple statistical features
• Compare deep learning accuracy with inter-observer variability

Innovation

• Deep learning segments T1/T2 cardiac MRI maps
• Multiple statistical features enhance disease detection
• Random forest improves classification performance significantly