RESPIRATORY SOUND CLASSIFICATION USING MACHINE LEARNING

Respiratory diseases are a major global health challenge, ranking among the leading cause for mortality anddisability worldwide. Conditions such as chronic obstructive pulmonary disease (COPD), asthma,tuberculosis, and lung cancer contribute significantly to global mortality, with COPD alone responsible foran estimated 3.91 million deaths annually. The burden is particularly severe in low-income regions withlimited healthcare infrastructure, where socioeconomic factors and a shortage of medical professionalsfurther hinder timely and accurate diagnoses. To tackle these issues, innovations in Artificial Intelligenceprovide effective strategies to minimize misdiagnoses and facilitate precise treatments. A cost-efficient anduser-friendly algorithm was created to analyze respiratory sounds, ensuring compatibility with variousdevices. The methodology integrates machine learning approaches, utilizing Gammatone CepstralCoefficients (GTCC) within a Convolutional Neural Network (CNN) and a CNN-LSTM hybrid modelincorporating GTCC and STFC features. Four datasets were prepared for classifying respiratory audio,covering healthy versus pathological classification, rale/rhonchus/normal sound classification, singularrespiratory sound type classification, and audio type classification with all sound types.