• Version
• Download 0
• File Size 517.84 KB
• File Count 1
• Create Date 21 July 2025
• Last Updated 21 July 2025

Anemia Prediction Using Machine Learning Algorithms: A Comparative Analysis

1Hari Prakash G

Department of Biotechnology

KIT Kalaignarkarunanidhi Institute of Technology

Coimbatore, India

2Shanmugabharath V

Department of Biotechnology

KIT Kalaignarkarunanidhi Institute of Technology

Coimbatore, India

3Gokulakrishnan M

Department of Biotechnology

KIT Kalaignarkarunanidhi Institute of Technology

Coimbatore, India

Ms. Kamali M L (Corresponding Author)

Department of Biotechnology

KIT Kalaignarkarunanidhi Institute of Technology

Coimbatore, India

Abstract - A prevalent public health concern that impacts billions of individuals globally is anemia, especially in nations with low and middle incomes. Timely intervention and better patient outcomes depend on early diagnosis and detection. Using clinical blood test data, this study attempts to create and assess different machine learning (ML) models for anaemia prediction. A dataset of 1,421 patient records was analyzed using five supervised machine learning algorithms: Random Forest, Gradient Boosting, Logistic Regression, Support Vector Machine (SVM), and K-Nearest Neighbours (KNN). Accuracy and ROC-AUC scores were used to assess performance. The models were interpreted using SHapley Additive exPlanations (SHAP). The findings showed that ensemble models outperformed other models with 100% accuracy and AUC. These results show how ML can be used in clinical diagnostic support systems.

Keywords: Anemia prediction, machine learning, SHAP, Random Forest, Gradient Boosting