Wheezing Feature Classification using ANFIS for Asthma Diagnosis

One of the chronic obstructive pulmonary diseases brought on by bronchial spasms is asthma. Skilled physicians use auscultation to examine patients with asthma in order to diagnose the condition in its early stages, but this method is subjective and depends on the knowledge and experience of individual physicians. Therefore, in order to increase the accuracy of asthma diagnosis, an objective auscultation procedure to identify wheezing in asthma must be established. The majority of research has classified lung sounds using their spectral properties, and some have even done feature extraction. Nonetheless, their performance has fallen short of expectations. The purpose of this study is to diagnose asthma by combining the time domain and frequency domain to identify patterns of wheezing features. The wheezing feature pattern threshold was successfully discovered through the use of mean and entropy features in the feature extraction process, which combined the pass-band Butterworthband filter with the DWT (Discrete Wavelet Transform) algorithm. The Adaptive Neuro-Fuzzy Inference System (ANFIS) algorithm was suggested in this study to efficiently carry out classification. The experimental findings shown that the suggested approach demonstrated 94.94% accuracy, 89.87% specificity, and 100% sensitivity.