A Comparative Analysis of Machine Learning Classifier of Anemia
Diagnosis Based on Complete Blood Count (CBC) Data

A Comparative Analysis of Machine Learning Classifier of Anemia
Diagnosis Based on Complete Blood Count (CBC) Data

Anemia classification, Complete Blood Count (CBC), Machine Learning, XGBoost, SVM, Naive Bayes, Medical Diagnosis

Anemia is a prevalent hematological condition that requires accurate and timely diagnosis to ensure effective treatment. This study aims to
compare the performance of several machine learning algorithms Random Forest, Support Vector Machine (SVM), Naive Bayes, and XGBoost
in classifying different types of anemia based on Complete Blood Count (CBC) data. The dataset includes three diagnostic categories: Healthy,
Normocytic hypochromic anemia, and Normocytic normochromic anemia. After preprocessing and normalization, each model was evaluated
using accuracy, precision, recall, F1-score, and ROC-AUC. The results show that XGBoost achieved the highest overall performance with 99%
accuracy and a perfect AUC of 1.00, followed closely by SVM and Naive Bayes. Naive Bayes showed lower performance, particularly in
identifying normocytic normochromic anemia. These findings suggest that machine learning, especially ensemble-based models, holds strong
potential in supporting clinical diagnosis of anemia using CBC data.

Nadya Awalia Putri1,*
, Bayu Priya Mukti2

https://ijiis.org/index.php/IJIIS/article/view/286/169

Amikom Purwokerto University

desember 2025

Fajar bagus W

PDF

English

Text