Convolutional neural networks based efficient approach for classification of lung diseases
- PDF / 1,137,530 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 31 Downloads / 143 Views
(2020) 8:4 Demir et al. Health Inf Sci Syst https://doi.org/10.1007/s13755-019-0091-3
RESEARCH
Convolutional neural networks based efficient approach for classification of lung diseases Fatih Demir1* , Abdulkadir Sengur1 and Varun Bajaj2
Abstract Treatment of lung diseases, which are the third most common cause of death in the world, is of great importance in the medical field. Many studies using lung sounds recorded with stethoscope have been conducted in the literature in order to diagnose the lung diseases with artificial intelligence-compatible devices and to assist the experts in their diagnosis. In this paper, ICBHI 2017 database which includes different sample frequencies, noise and background sounds was used for the classification of lung sounds. The lung sound signals were initially converted to spectrogram images by using time–frequency method. The short time Fourier transform (STFT) method was considered as time–frequency transformation. Two deep learning based approaches were used for lung sound classification. In the first approach, a pre-trained deep convolutional neural networks (CNN) model was used for feature extraction and a support vector machine (SVM) classifier was used in classification of the lung sounds. In the second approach, the pre-trained deep CNN model was fine-tuned (transfer learning) via spectrogram images for lung sound classification. The accuracies of the proposed methods were tested by using the ten-fold cross validation. The accuracies for the first and second proposed methods were 65.5% and 63.09%, respectively. The obtained accuracies were then compared with some of the existing results and it was seen that obtained scores were better than the other results. Keywords: Lung disease detection, Deep learning, Convolutional neural networks, Time-frequency images Introduction Respiratory system diseases affect people’s social, economic and health life significantly. For these reasons, a lot of researches are going on for early diagnosis and intervention in respiratory diseases. In this context, lung sound characteristics provide important clues in the diagnosis of respiratory abnormalities and infections. Auscultation is an effective technique in which physicians evaluate and diagnose the disease after using a stethoscope for lung disease. This method is both inexpensive and easy, and also it does not require internal intervention into the human body [1]. However, traditional stethoscopes may *Correspondence: [email protected] 1 Electrical and Electronics Engineering Dept., Technology Faculty, Firat University, Elazig, Turkey Full list of author information is available at the end of the article
be exposed to external noise sounds, weaken the sound components above 120 Hz, and cannot filter the audio frequencies of the body in auscultation and cannot create permanent recordings in monitoring of the disease course [1]. In addition, accurate diagnosis of diseases requires highly experienced medical staff. Therefore, it is important to use electronic instrumentation and syst
Data Loading...
