CLASSIFYING LIVER DISEASE WITH BOOSTING MACHINE LEARNING APPROACHES

Liver diseases pose a significant global health challenge due to their impact on metabolic function and the difficulty of early detection. Traditional diagnostic methods such as liver biopsy have limitations due to their invasive nature and high costs. This research examines the application of advan...

Full description

Saved in:
Bibliographic Details
Main Authors: Erol Özçekiç, Ümit Yılmaz
Format: Article
Language:English
Published: Eskişehir Osmangazi University 2025-08-01
Series:Eskişehir Osmangazi Üniversitesi Mühendislik ve Mimarlık Fakültesi Dergisi
Subjects:
Online Access:https://dergipark.org.tr/tr/download/article-file/4396614
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849228960783663104
author Erol Özçekiç
Ümit Yılmaz
author_facet Erol Özçekiç
Ümit Yılmaz
author_sort Erol Özçekiç
collection DOAJ
description Liver diseases pose a significant global health challenge due to their impact on metabolic function and the difficulty of early detection. Traditional diagnostic methods such as liver biopsy have limitations due to their invasive nature and high costs. This research examines the application of advanced machine learning techniques such as Gradient Boosting, AdaBoost, XGBoost and CatBoost for classification of liver diseases using a publicly available dataset of 1700 clinical records. Statistical analyses identified key predictors such as age, body mass index (BMI), lifestyle factors, and liver function tests, which were used to train and evaluate the models. The performance of the models was evaluated using metrics such as accuracy, precision, recall and AUC-ROC. The CatBoost model showed the highest performance with an accuracy of 93.82%, while also producing the most consistent results with precision (91.97%), recall (96.62%), F1 score (94.25%) and AUC-ROC (95.64%). These results highlight the potential of machine learning-based approaches to improve diagnostic accuracy and reduce reliance on invasive procedures. The proposed framework can contribute to improving patient outcomes and optimizing healthcare resources by providing a foundation for real-time clinical decision support systems.
format Article
id doaj-art-c48ed8d723a44d84b9caee98ac34c758
institution Kabale University
issn 2630-5712
language English
publishDate 2025-08-01
publisher Eskişehir Osmangazi University
record_format Article
series Eskişehir Osmangazi Üniversitesi Mühendislik ve Mimarlık Fakültesi Dergisi
spelling doaj-art-c48ed8d723a44d84b9caee98ac34c7582025-08-22T11:23:57ZengEskişehir Osmangazi UniversityEskişehir Osmangazi Üniversitesi Mühendislik ve Mimarlık Fakültesi Dergisi2630-57122025-08-013321882189210.31796/ogummf.1591951122CLASSIFYING LIVER DISEASE WITH BOOSTING MACHINE LEARNING APPROACHESErol Özçekiç0https://orcid.org/0000-0002-1896-6853Ümit Yılmaz1https://orcid.org/0000-0003-4268-8598BALIKESİR ÜNİVERSİTESİ, BALIKESİR MESLEK YÜKSEKOKULU, BİLGİSAYAR TEKNOLOJİLERİ BÖLÜMÜ, BİLGİSAYAR PROGRAMCILIĞI PR.BALIKESİR ÜNİVERSİTESİ, BİGADİÇ MESLEK YÜKSEKOKULU, YÖNETİM VE ORGANİZASYON BÖLÜMÜ, LOJİSTİK PR.Liver diseases pose a significant global health challenge due to their impact on metabolic function and the difficulty of early detection. Traditional diagnostic methods such as liver biopsy have limitations due to their invasive nature and high costs. This research examines the application of advanced machine learning techniques such as Gradient Boosting, AdaBoost, XGBoost and CatBoost for classification of liver diseases using a publicly available dataset of 1700 clinical records. Statistical analyses identified key predictors such as age, body mass index (BMI), lifestyle factors, and liver function tests, which were used to train and evaluate the models. The performance of the models was evaluated using metrics such as accuracy, precision, recall and AUC-ROC. The CatBoost model showed the highest performance with an accuracy of 93.82%, while also producing the most consistent results with precision (91.97%), recall (96.62%), F1 score (94.25%) and AUC-ROC (95.64%). These results highlight the potential of machine learning-based approaches to improve diagnostic accuracy and reduce reliance on invasive procedures. The proposed framework can contribute to improving patient outcomes and optimizing healthcare resources by providing a foundation for real-time clinical decision support systems.https://dergipark.org.tr/tr/download/article-file/4396614karaciğer hastalığımakine öğrenmesitanısınıflandırmaboostingliver diseasemachine learningdiagnosisclassificationboosting
spellingShingle Erol Özçekiç
Ümit Yılmaz
CLASSIFYING LIVER DISEASE WITH BOOSTING MACHINE LEARNING APPROACHES
Eskişehir Osmangazi Üniversitesi Mühendislik ve Mimarlık Fakültesi Dergisi
karaciğer hastalığı
makine öğrenmesi
tanı
sınıflandırma
boosting
liver disease
machine learning
diagnosis
classification
boosting
title CLASSIFYING LIVER DISEASE WITH BOOSTING MACHINE LEARNING APPROACHES
title_full CLASSIFYING LIVER DISEASE WITH BOOSTING MACHINE LEARNING APPROACHES
title_fullStr CLASSIFYING LIVER DISEASE WITH BOOSTING MACHINE LEARNING APPROACHES
title_full_unstemmed CLASSIFYING LIVER DISEASE WITH BOOSTING MACHINE LEARNING APPROACHES
title_short CLASSIFYING LIVER DISEASE WITH BOOSTING MACHINE LEARNING APPROACHES
title_sort classifying liver disease with boosting machine learning approaches
topic karaciğer hastalığı
makine öğrenmesi
tanı
sınıflandırma
boosting
liver disease
machine learning
diagnosis
classification
boosting
url https://dergipark.org.tr/tr/download/article-file/4396614
work_keys_str_mv AT erolozcekic classifyingliverdiseasewithboostingmachinelearningapproaches
AT umityılmaz classifyingliverdiseasewithboostingmachinelearningapproaches