Submit Your Manuscript Online

Detection of Lung Disease Using MFCC and CNN

	International Journal of Scientific Research and Engineering Development (IJSRED)	Full Text:
	Published Issue : Volume-3 Issue-6
	Year of Publication : 2020
	Unique Identification Number : IJSRED-V3I6P39
	Authors : V.Subapriya, Dr.K.L.Shunmuganathan, Kypa Sandhil, Naresh Mayya, Gopireddy Tejeswarreddy
	: Click Here

Abstract :

Lungs sounds has been used as trustworthy tool for diagnosis of various diseases for centuries. The growth in science and technology has developed different tools for identification of different types of diseases related to lungs. However we cannot fully depend on these tools for identification of various lungs related problems as using these tools is more or less subjective. In a study conducted among the physicians in US it was found that the doctors were unable to characterize and identify only 70% to 80% of wheeze sounds in a series of recording. In developed and under developing countries there is a scarcity of trained doctors. In such cases the accuracy of sound identification is much worse which will lead to misdiagnose of pulmonary diseases and their proper lungs sounds. Hence there is a need for technique which will classify various abnormal sounds such as wheeze and crackles against the normal one. In this paper we have proposed a system to classify the lung sounds into normal,wheeze, crackle and both that are taken by an electronic stethoscope. The proposed method achieves an accuracy of 80% and 66% for training and testing respectively.