文章基本信息

标题：THE CLASSIFICATION OF POSSIBLE CORONAVIRUS TREATMENTS ON A SINGLE HUMAN CELL USING DEEP LEARNING AND MACHINE LEARNING APPROACHES
本地全文：下载
作者：NOUR ELDEEN KHALIFA ; GUNASEKARAN MANOGARAN ; MOHAMED HAMED N 等
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2021
卷号：99
期号：21
语种：English
出版社：Journal of Theoretical and Applied
摘要：Every major healthcare system is now under the throes of the Coronavirus disease outbreak as it is operating at its maximum capacity. There is an absolute need to establish an appropriate cure for this virus as quickly and efficiently as possible. Advances in deep learning models may play a critical role in SARS-2 discovery by locating a possible treatment. This article's objective is to demonstrate the machine learning and deep learning models approaches for classifying prospective coronavirus treatment on a single human cell. A partial dataset of RXRX.ai which is a publicly available dataset is used in this research. This work targeted to implement a strategy for automatically identifying a single human cell depending on the type of treatment and its concentration level. Throughout this study, we present a DCNN model along with an image processing approach. The systematic approach comprises translating the original dataset's numerical attributes to the image domain, and then incorporating them into DCNN model. In comparison to standard machine learning techniques including such Ensemble, Decision Tree and Support Vector Machine, the experimental findings indicate that the suggested DCNN model for treatment classification (32 categories) obtained a testing accuracy of 98.05 percent. The (Ensemble) algorithm achieves 98.5 percent for the accuracy test in treatment concentration level prognosis, whereas the suggested DCNN model reached 98.2 percent. The classification of treatments and assessing their concentration levels are considerably accurate due to the performance indicators obtained from the experiments.
关键词：Deep Learning;Machine Learning;Coronavirus;COVID-