A novel multi-task machine learning classifier for rare disease patterning using cardiac strain imaging data

• Published in: Scientific reports Vol. 14; no. 1; pp. 10672 - 9
• Main Authors: Siva, Nanda K., Singh, Yashbir, Hathaway, Quincy A., Sengupta, Partho P., Yanamala, Naveena
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 09.05.2024; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114; 639/166; 639/705; 692/308; 692/4019; Adult; Aged; Alternation learning; Cardiomyopathy; Cardiomyopathy, Restrictive - diagnostic imaging; Congestive heart failure; Constrictive pericarditis; Coronary artery disease; Echocardiography; Echocardiography - methods; Female; Heart diseases; Heart failure; Heart Failure - diagnostic imaging; Heart Ventricles - diagnostic imaging; Heart Ventricles - physiopathology; Homology; Humanities and Social Sciences; Humans; Learning algorithms; Machine Learning; Male; Middle Aged; multidisciplinary; Pericarditis; Pericarditis, Constrictive - diagnosis; Pericarditis, Constrictive - diagnostic imaging; Rare disease; Rare diseases; Rare Diseases - diagnostic imaging; Restrictive cardiomyopathy; Retrospective Studies; Science; Science (multidisciplinary); Restrictive cardiomyopathy; Constrictive pericarditis; Echocardiography; Rare disease; Machine learning
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-024-61201-4

To provide accurate predictions, current machine learning-based solutions require large, manually labeled training datasets. We implement persistent homology (PH), a topological tool for studying the pattern of data, to analyze echocardiography-based strain data and differentiate between rare diseases like constrictive pericarditis (CP) and restrictive cardiomyopathy (RCM). Patient population (retrospectively registered) included those presenting with heart failure due to CP (n = 51), RCM (n = 47), and patients without heart failure symptoms (n = 53). Longitudinal, radial, and circumferential strains/strain rates for left ventricular segments were processed into topological feature vectors using Machine learning PH workflow. In differentiating CP and RCM, the PH workflow model had a ROC AUC of 0.94 (Sensitivity = 92%, Specificity = 81%), compared with the GLS model AUC of 0.69 (Sensitivity = 65%, Specificity = 66%). In differentiating between all three conditions, the PH workflow model had an AUC of 0.83 (Sensitivity = 68%, Specificity = 84%), compared with the GLS model AUC of 0.68 (Sensitivity = 52% and Specificity = 76%). By employing persistent homology to differentiate the “pattern” of cardiac deformations, our machine-learning approach provides reasonable accuracy when evaluating small datasets and aids in understanding and visualizing patterns of cardiac imaging data in clinically challenging disease states.