Open AccessDeteksi Coronavirus Disease Pada X-Ray Dan CT-Scan Paru Menggunakan Convolutional Neural Network
Author(s) -
Muhammad Ridho Fauzi,
Puspa Eosina,
Dewi Primasari
Publication year - 2021
Publication title -
juss (jurnal sains dan sistem informasi)
Language(s) - English
Resource type - Journals
ISSN - 2614-8277
DOI - 10.22437/juss.v3i2.10888
Subject(s) - convolutional neural network , confusion matrix , covid-19 , confusion , computer science , artificial intelligence , viral pneumonia , pneumonia , coronavirus , pattern recognition (psychology) , medicine , deep learning , disease , pathology , infectious disease (medical specialty) , psychology , psychoanalysis
In early 2020, countries in the world were shocked by the outbreak of a new virus, namely SARS-CoV-2 and the disease was named Coronavirus 2019 (Covid-19). It is known that the virus originated in Wuhan, China and was discovered at the end of December 2019. Based on data on July 18, 2020, there are more than 180 countries that have contracted Covid-19 with a total of 13,824,739 confirmed cases since December 31, 2019. Based on data on positive cases of Covid- 19 above, the average patient has several clinical symptoms, one of which is having difficulty breathing due to a large pneumonia infiltrate in the lungs. Therefore, it is necessary to implement an automatic pulmonary diagnosis system as an alternative to prevent the increasingly widespread spread of Covid-19. Covid-19 can be detected in the lungs through digital image processing of chest X-ray using the Convolutional Neural Network (CNN) algorithm. CNN is a Deep Learning method that functions to identify digital images. In this study, three different scenarios were used. This scenario aims to find the best model using hyperparameter tunnning. The results of ROC analysis and confusion matrix show that in scenarios I, II and III get 94%, 95% and 93% accuracy.