Predicting Infectious State of Hepatitis C Virus Affected Patient's Applying Machine Learning Methods

Khair Ahammed, Md Shahriare Satu, Md Imran Khan, Md Whaiduzzaman

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

23 Citations (Scopus)

Abstract

Hepatitis C virus is a major cause for happening liver disease all over the world. However, many tools have been build that try to reduce the influence of this virus. In this work, a machine learning based model has been proposed that can classify hepatitis C virus infected patient's stages of liver. We gathered the instances of liver fibrosis disease of Egyptian patients from UCI machine learning repository. To balance instances of multiple categories, synthetic minority oversampling methodology has been used that increases synthetic instances of patients. Later, we applied different feature selection methods to identify significant features of hepatitis C virus in this dataset. Various classifiers has been employed to categorize patients into balanced primary, feature selected and primary HCV instances. After analyzing this results, KNN shows the best 94.40% accuracy than any other classifiers. This result has been useful to scrutinize and take decision in hepatitis C virus infectious disease.

Original languageEnglish
Title of host publication2020 IEEE Region 10 Symposium, TENSYMP 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1371-1374
Number of pages4
ISBN (Electronic)9781728173665
DOIs
Publication statusPublished - 5 Jun 2020
Event2020 IEEE Region 10 Symposium, TENSYMP 2020 - Virtual, Dhaka, Bangladesh
Duration: 5 Jun 20207 Jun 2020

Publication series

Name2020 IEEE Region 10 Symposium, TENSYMP 2020

Conference

Conference2020 IEEE Region 10 Symposium, TENSYMP 2020
Country/TerritoryBangladesh
CityVirtual, Dhaka
Period5/06/207/06/20

Keywords

  • classification
  • feature selection
  • Hepatitis C virus
  • machine learning
  • SMOTE

Fingerprint

Dive into the research topics of 'Predicting Infectious State of Hepatitis C Virus Affected Patient's Applying Machine Learning Methods'. Together they form a unique fingerprint.

Cite this