This study introduces a novel machine learning-based discriminant model aimed at improving the detection of sleep-disordered breathing (SDB) in patients with hypertrophic cardiomyopathy (HCM). The model was developed using nocturnal oximetry data from a multicentre study involving 369 HCM patients, of whom 61.8% exhibited a high Oxygen Desaturation Index (ODI), indicative of moderate to severe SDB. The model, tested on a separate set of patients, demonstrated a high area under the receiver operating characteristic curve (AUC) of 0.86, with a sensitivity of 0.91 and specificity of 0.68. It significantly outperformed a previous logistic regression model, showing a higher predictive accuracy for SDB presence. This advancement is crucial as it offers a more effective tool for the early detection of SDB in HCM patients, potentially leading to better-targeted interventions and prevention of adverse cardiovascular events associated with SDB.