EARLY DETECTION OF ASYMPTOMATIC COVID-19 INFECTION WITH ARTIFICIAL NEURAL NETWORK MODEL THROUGH VOICE RECORDING OF FORCED COUGH

Aisyah, Khairun Nisa (2021) EARLY DETECTION OF ASYMPTOMATIC COVID-19 INFECTION WITH ARTIFICIAL NEURAL NETWORK MODEL THROUGH VOICE RECORDING OF FORCED COUGH. S1 thesis, Universitas Mataram.

[img] Text
TUGAS AKHIR_2_Aisyah Khairun Nisa (F1D018003)_versi ais.pdf
Restricted to Repository staff only
Download (1MB)

Abstract

Corona Virus 2019 or often called COVID-19, is a disease transmitted by SARSCoV-2. Based on data per March 15th, 2021, from the World Health Organization, the cumulative total of cases in Indonesia is 1,419,455 cases, and Indonesia's cumulative total of deaths is 38,426 cases, placing Indonesia in the third position with the highest number of deaths under India and Iran. The spread of COVID-19 happened very quickly and widely because it spread from direct human contact with droplets from the respiratory of an infected person. American Centers for Disease Control and Prevention says that asymptomatic people are likely to account for more than 50% of the transmission rate. The antigen tests are used as early detection of COVID-19, with an accuracy of results ranging from 80-90%. As of the 3rd September 2021, the price for the antigen test is renewed, with prices ranging from Rp. 99.000 - Rp. 109.000, but researchers are still tenaciously looking for the best alternative solutions for the early detection of COVID-19. MIT News Office reported that asymptomatic COVID-19 infection could be detected through a forced cough recording. This research purpose deep learning model, Artificial Neural Networks (ANN) to detect asymptomatic COVID-19 patient through the voice recording of forced cough. The Artificial Neural Network (ANN) using oversampling data gain accuracy as high as 98% with loss value less than 3% on detecting asymptomatic person from forced cough recording. This can be used as a solution for the early detection of COVID-19 infection at no cost, anytime and anywhere.

Item Type: Thesis (S1)
Keywords (Kata Kunci): Keywords: ANN, Asymptomatic, Covid-19, Forced Cough, Oversamplin
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Teknik
Depositing User: Rini Trisnawati
Date Deposited: 24 Jan 2022 04:53
Last Modified: 24 Jan 2022 04:53
URI: http://eprints.unram.ac.id/id/eprint/27380

Actions (login required)

View Item View Item
Repository UNRAM is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.