本計畫結合呼吸音辦識技術，利用深度學習模型進行異常呼吸音的自動判讀和風險預測。
此技術能識別各類異常呼吸音（如喘鳴音、爆裂音），並對慢性阻塞性肺疾病（COPD）、肺炎、哮喘等疾病提供早期預警與診斷建議。
The project integrates respiratory sound recognition technology, using deep learning models for automatic interpretation of abnormal respiratory sounds and risk prediction.
This system can identify abnormal respiratory sounds such as wheezes and crackles, providing early warnings and diagnostic suggestions for conditions like Chronic Obstructive Pulmonary Disease (COPD), pneumonia, and asthma.