Machine Learning for Classifying Liver Diseases

Ahmed, Omar; Shiba, Rana; Magdy, Philopateer

doi:10.21608/iiis.2024.342004

Machine Learning for Classifying Liver Diseases

Authors

¹ Department of Artificial Intillegence Misr University for Science and Technology,Cairo, Egypt

² Department of Artificial Intillegence Misr University for Science and Technology ,Cairo, Egypt

³ Department of Computer Science Misr University for Science and Technology ,Cairo, Egypt

10.21608/iiis.2024.342004

Abstract

Hepatitis C poses a significant public health challenge, often progressing to life-threatening chronic infections without early symptoms, making it difficult to detect and diagnose This study promptly provides a solution by developing a machine learning model aimed at the early detection of hepatitis C-related liver diseases A comprehensive analysis was done using data from the captured UCI Machine Learning Repository, with laboratory internal values and patient demographics and apply 5 algorithms for liver diseases classification.
The study's results were good, achieving a validation accuracy of 97% and testing accuracy of 95%. For detailed analysis, we used a confusion matrix to reveal additional insights into the model performance, as it achieved a Precision score of 99%, Recall score of 99%, and F1-score of 99% as well.
These findings present the promising potential of machine learning for early disease classification, which can help in medical applications, minimize human error, and participate in improved detection of Liver diseases.

Keywords