A haematological data-driven approach for the classification and early detection of Anaemia in women using machine learning techniques: a comparative analysis of Random Forest and Logistic Regression models

View at publisher Summarise with VIRAT
Authors: Raju Anitha, Praveena Mandapati, V Vinay Kumar, M V V A L Sunitha, Palamakula Ramesh Babu, A Aruna Kumari
Cited by: 0
References: 15

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Anaemia remains a pressing public health concern, especially among women, often going undiagnosed in its early stages due to subtle symptoms and limited access to advanced diagnostics. The purpose of this study is to close that gap by predicting anaemia using routinely available Complete Blood Count (CBC) values utilising machine learning techniques, notably Random Forest and Logistic Regression. After preprocessing steps like cleaning missing values, normalising numeric features, and encoding categorical data, Anaemia classification was carried out based on the World Health Organization (WHO)- defined threshold for women (HGB < 12 g/dL). Machine learning models have been trained on these marked data to classify individuals as anaemic or non-anaemic. Random Forest and logistical regression algorithms were evaluated by metrics such as accuracy, accuracy, appeal, F1-Score and Roc-AUC. The Random Forest model achieved the highest performance, accurately identifying Anaemia cases with over 90% accuracy. Logistic Regression also demonstrated competitive results, exhibiting strong interpretability and simplicity. Feature importance analysis highlighted Haemoglobin, MCV, and PCV as the most influential predictors in the model’s decision-making process. A comparative analysis was conducted with existing models, demonstrating that the Random Forest algorithm provided the most accurate results.

Fields of Study & Subjects

Artificial Intelligence in Healthcare Digital Imaging for Blood Diseases Data Mining and Machine Learning Applications

Publication details

Journal / source
Australian Journal of Electrical & Electronics Engineering
Pages
1–10
Publication date
20 March 2026
Language
EN