Hepatitis C Diagnosis Using Computational Intelligence Techniques


Cedolin M., GENEVOIS M., Canbulat Z.

International Conference on Intelligent and Fuzzy Systems, INFUS 2024, Çanakkale, Turkey, 16 - 18 July 2024, vol.1090 LNNS, pp.29-36 identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 1090 LNNS
  • Doi Number: 10.1007/978-3-031-67192-0_4
  • City: Çanakkale
  • Country: Turkey
  • Page Numbers: pp.29-36
  • Keywords: Health Data, Hepatitis C, Machine Learning, Risk Analysis
  • Galatasaray University Affiliated: Yes

Abstract

The global impact of chronic Hepatitis C virus (HCV) infections, estimating around 58 million affected individuals worldwide with 1.5 million new cases annually. In 2019, approximately 290,000 deaths were attributed to Hepatitis C, as per the World Health Organization (WHO). Despite the availability of numerous laboratory tests for liver disease assessment, patients often show normal range laboratory values, and there are asymptomatic cases. The passage highlights the necessity for a sophisticated system to diagnose the disease accurately and efficiently, emphasizing the support of machine learning in this endeavor. It mentions the use of HCV data from the “UCI Machine Learning Repository” and the application of six computational intelligence algorithms, namely, logistic regression, random forest, k-NN, SVM, light-GBM, and XG-Boost. After data normalization, imputation, and oversampling processes light-GBM showed the best performance with 98,9% accuracy.