Design android application to determine pneumonia levels based on Dempster Shafer method

Endah Purwanti, Debrina Rizka Pangesti, Arief Bakhtiar, La Febry Andira Rose Cynthia

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Pneumonia is a respiratory disease with a high prevalence in Indonesia. Determination of pneumonia levels is done by radiology examination, physical examination and clinical history. If the determination of the level of pneumonia is not right, it will affect the treatment and subsequent treatment. The purpose of this study is to design an android application that can be used to determine pneumonia levels. The device was built using 9 input parameters in the form of standard inspection results for the diagnosis of this disease. The classification method uses the Dempster-Shafer algorithm which has been equipped with a belief value for each parameter. The device output consists of 2 classes, namely mild pneumonia or severe pneumonia. Through the device performance tests obtained sensitivity, specificity and accuracy of 85%, 94.29% and 90.90%.

Original languageEnglish
Title of host publication2nd International Conference on Physical Instrumentation and Advanced Materials 2019
EditorsHerri Trilaksana, Sulaiman Wadi Harun, Cameron Shearer, Moh Yasin
PublisherAmerican Institute of Physics Inc.
ISBN (Electronic)9780735440562
DOIs
Publication statusPublished - 9 Dec 2020
Event2nd International Conference on Physical Instrumentation and Advanced Materials, ICPIAM 2019 - Surabaya, Indonesia
Duration: 22 Oct 2019 → …

Publication series

NameAIP Conference Proceedings
Volume2314
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

Conference2nd International Conference on Physical Instrumentation and Advanced Materials, ICPIAM 2019
Country/TerritoryIndonesia
CitySurabaya
Period22/10/19 → …

Fingerprint

Dive into the research topics of 'Design android application to determine pneumonia levels based on Dempster Shafer method'. Together they form a unique fingerprint.

Cite this