Anemia Level Detection and Prediction in Pregnant Women Using Machine Learning

Abstract

Anemia during pregnancy is a significant public health concern, leading to adverse maternal and fetal outcomes. Early detection and accurate prediction of anemia levels can help in timely interventions and better healthcare management. This study presents a machine learning-based approach for anemia detection and prediction in pregnant women by utilizing both structured and unstructured data. The structured data includes clinical parameters such as hemoglobin levels, serum ferritin, red blood cell count, and dietary intake, while the unstructured data consists of conjunctiva, fingernail, and palm images analyzed using deep learning techniques. By integrating these multimodal data sources, the proposed model enhances the accuracy of anemia detection compared to conventional diagnostic methods. The study leverages image processing for non-invasive anemia assessment and applies machine learning algorithms to structured data for comprehensive risk prediction. Experimental results demonstrate the model’s effectiveness in classifying anemia levels, offering a robust tool for clinical decision-making and maternal health monitoring.