Advanced Medical Images Recognition and Diagnosis of Respiratory System Viruses

Résumé

Abstract Respiratory infections are a confusing and time-consuming task of constantly looking at clinical pictures of patients. Therefore, there is a need to develop and improve the respiratory case prediction model for Covid-19 and Viral Pneumonia as soon as possible to control the spread of disease. Deep learning makes it possible to discover that respiratory viruses such as Covid-19 and Viral Pneumonia can be effectively acquired using its classification tools such as CNN (Convolutional Neural Network). MFCC (Mel Frequency Cepstral Coefficients) is a common and effective method of signal processing. In this research, MFCC - CNN's learning model is proposed to speed up the prediction process that assists medical professionals. MFCC is used to extract image features related to the presence of Covid-19 and Viral Pneumonia or not. Prediction is done using a convolutional neural network. This makes the time-consuming process easier and faster with more accurate results for radiologists and this reduces the spread of the virus and saves lives. Experimental results show that using a CT image converted to Mel-frequency cepstral spectrogram as input to CNN can achieve high accuracy results; with the classification of validation data of 100% accuracy of the appropriate Covid-19 and Viral Pneumonia categories and images with the normal healthy (NON COVID) label. Therefore, it can probably be used to detect whether Covid-19 orViral Pneumonia are present in the CT images. The work here provides evidence of the idea that high accuracy can be achieved with a moderate dataset, which can have a significant impact on this area.

Citer ce document

Accès au document

Ce document est hébergé sur son dépôt institutionnel d'origine.

Auteur(s)

Statistiques

Consultations : 1

Téléchargements : 0