Early Prediction Model for Anemia in Infants Using Clinical Data from Perú Applying Supervised Machine Learning Algorithms

Anemia is a disease that is always present from early stages; children are the most prone to contracting it from birth if it's not diagnosed, which can lead to more severe illnesses. Therefore, a predictive model is proposed for the detection of anemia in infants using their clinical data. The Cross Industry Standard Process for Data Mining (CRISP-DM) methodology is employed, along with Analysis Services for the Extract, Transform, Load (ETL) process and four Machine Learning (ML) algorithms: Logistic Regression (LR), Decision Tree (DT), Support Vector Machine (SVM), and Random Forest (RF). All four phases of CRISP-DM were utilized: Business Understanding, Data Understanding, Data Preparation, Modeling, and Evaluation. The dataset used consists of 400,000 records related to the medical history of children from a healthcare institution in Peru, from which 27 variables were selected. The results showed that the DT algorithm performed better, achieving an AUC-ROC Curve greater than 93%, and the “Scikit Learn” library for ML in Python facilitated obtaining the results.