This study proposes an interpretable fuzzy rule–based classifier for phenotype classification in rare inherited cardiovascular diseases, focusing on hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), arrhythmogenic right ventricular cardiomyopathy (ARVC), long QT syndrome (LQTS), and Brugada syndrome. The model integrates multimodal data routinely available in clinical practice, including 12-lead ECG intervals and patterns, echocardiographic and cardiac magnetic resonance–derived structural and functional parameters (such as LV wall thickness, LVEDD, and LVEF), targeted genetic findings, and key demographic and clinical variables. Fuzzy linguistic variables and membership functions are defined over these features, and a rule base combining expert knowledge and data-driven rules is trained using a cross-entropy loss with class weighting to address phenotype imbalance. Benchmark comparisons are performed against logistic regression, support vector machines, random forests, and a small neural network using accuracy, macro-F1, ROC–AUC, and per-class sensitivity and precision. The fuzzy classifier achieves competitive or superior performance overall and demonstrates improved detection of less prevalent phenotypes such as ARVC and Brugada syndrome, while preserving performance on common phenotypes like HCM and DCM. At the same time, rule-level explanations and clinically meaningful membership functions provide transparent, traceable decision pathways, supporting clinical acceptability and trust in the model’s predictions.