Newswise – Researchers at Huazhong University of Science and Technology and Tongji Medical College conducted a study titled “Machine Learning Modeling Identifies Hypertrophic Cardiomyopathy Subtypes with Genetic Characteristics.” This study was published in Frontiers of medicineVolumes 17 and Issue 4.
Previous studies have revealed that patients with hypertrophic cardiomyopathy (HCM) exhibit differences in symptoms severity and prognosis, indicating potential HCM subtypes among these patients. Here, 793 patients with HCM were recruited at a mean follow-up of 32.78 ± 27.58 months to identify potential HCM subtypes by performing consensus clustering based on echocardiographic function. Furthermore, this study proposed a systematic method to demonstrate the phenotype and genotype relationship of each HCM subtype using machine learning modeling and interaction network detection techniques, based on machine learning modeling and whole exome sequencing data. Another independent cohort of 414 patients in HCM was recruited to replicate the findings. As a result, two subtypes characterized by different clinical outcomes were identified in HCM. Subtype 2 patients showed asymmetric septal hypertrophy associated with stable course, while subtype 1 patients showed left ventricular systolic dysfunction and aggressive progression. Machine learning modeling based on individual systemic data identified 46 genes with mutational burden that could accurately predict subtype trends. Furthermore, patients from another cohort were predicted to be predicted as subtype 1 by the 46 gene model. By employing echocardiography and genetic screening for 46 genes, this study categorized HCM into two subtypes with distinct clinical outcomes.
This research was supported by the National Key R&D Program of China, the National Natural Science Foundation of China, Shanghai City Science and Technology Major Project, and the Basic Research Fund of Central University. For more information, the complete paper is available as follows: https://journal.hep.com.cn/fmd/en/10.1007/S11684-023-0982-1.
