Transfer Learning Approaches for Lung Sound Detection

lung disease is now the biggest cause of death in the whole world. For the most part, lung disease is found in its last stages after it has already progressed to a serious state. One on either hand, early detection of lung disease may aid in therapy. In today's society, technological advancements are critical to the delivery of healthcare services. To record the sounds emitted by the patients' lungs, we use electronic stethoscopes. The sounds made by the lungs give useful information for lung diagnosis. Study in the medical profession is now concentrating on the significance of diagnosing lung disease using lung acoustics, which is a current research topic. Transfer learning is critical for success in the medical system. In this study, we provide a range of transfer learning algorithms for lung sound classification, including ALEXNET, MOBILENET, VGGNET, and RESNET, among others. For the new data set, the WAVGAN model will be used to produce a new data set, which will be used to perform recognition of respiratory system noises with in Transfer learning model. With their high accuracy in classifying lung sounds, our transfer learning models might one day be used to diagnose lung disorders. Transfer learning strategies and their advantages and disadvantages will be discussed in this article. There are other dangers to consider. As a means of distinguishing between the four distinct lung sounds, in addition, it recommends future directions for research into lung sound identification.