This study presents our contribution to the murmur classification task from the 2022 George B. Moody PhysioNet Challenge. This task consisted of classifying murmurs as present, absent or unknown using patients' heart sound recordings and demographic data. Models were evaluated using a weighted accuracy biased towards the present and unknown.
Two models are designed and implemented. The first model is a Dual Bayesian ResNet (DBRes), where each patient's heart sound recording is segmented into overlapping log mel spectrograms. These spectrograms undergo two binary classifications: present versus unknown or absent, and unknown versus present or absent. These classifications are aggregated to give a patient's final classification. The second model is the output of DBRes integrated with demographic data and signal features using XGBoost.
DBRes achieved our best weighted accuracy of 0.768 on the hidden test set. On our local test set, integrating the demographic data and signal features improved DBRes's accuracy from 0.762 to 0.820. However, this decreased DBRes's weighted accuracy from 0.780 to 0.749.
Our results demonstrate that log mel spectrograms are an effective representation of heart sound recordings, Bayesian networks provide strong supervised classification performance, and treating the ternary classification as two binary classifications increases performance on the weighted accuracy.