Heart scan AI could predict sudden cardiac death risk in common heart condition

Copied word for word from the study's registry entry, so the wording is the study team's rather than ours.

Inclusion Criteria: * Age ≥ 18 years at the time of enrollment * Diagnosis of hypertrophic cardiomyopathy (HCM) according to current ESC guideline criteria, defined as left ventricular wall thickness unexplained solely by loading conditions * Availability of a clinically indicated cardiac magnetic resonance (CMR) examination performed according to standard protocols and of sufficient diagnostic quality for radiomic analysis * Written informed consent for participation in the study and for data processing, when required by applicable regulations and local center procedures For the subgroup of patients with obstructive HCM treated with mavacamten only: * Initiation of mavacamten therapy according to clinical indication * Availability of baseline CMR and follow-up CMR performed at a later time point Exclusion Criteria: * Age \< 18 years * Absence of a diagnosis of hypertrophic cardiomyopathy according to current ESC guideline criteria * Presence of phenocopies of hypertrophic cardiomyopathy or other structural cardiac diseases that may interfere with phenotypic characterization of HCM, including but not limited to cardiac amyloidosis, Fabry disease, infiltrative or storage cardiomyopathies, and other forms of secondary left ventricular hypertrophy not consistent with HCM * Inadequate quality of cardiac magnetic resonance (CMR) images for radiomic analysis, including motion artifacts, low spatial resolution, incomplete acquisitions, or lack of technical adequacy of required sequences * Absence of required CMR sequences for the planned analyses (in particular cine and/or late gadolinium enhancement \[LGE\] sequences) * Inability to achieve reliable myocardial segmentation in relevant sequences due to technical or anatomical reasons * Prior septal reduction therapy (surgical myectomy or alcohol septal ablation), when such intervention substantially alters myocardial morphology and prevents meaningful comparison with native phenotype radiomic analysis