Comparative Study of AI-based Anemia Prediction using Eye Conjunctiva Images and Blood Tests

Anemia is a condition characterized by low levels of hemoglobin in the blood or low levels of blood in the body. It adversely impacts the overall well-being of individuals and is associated with higher mortality rates. Anemia is usually detected through medical history and blood tests, although new studies show that it can be detected early from eye images. Using AI to detect anemia with blood tests and eye images instead of relying on traditional analysis methods allows for the identification of high-risk populations, earlier diagnoses, and increased accuracy. This paper explores the use of a python-based machine learning model in the prediction of anemia. The datasets used in this project were curated by Eureka diagnostic center in Lucknow, India and Medical College Hospital in Bangladesh. The primary outcome of this study is a model that predicts whether a patient has anemia, both with the blood test dataset and the eye image dataset. The AdaBoostClassifier, LGBMClassifier, ExtraTreesClassifier had a 100% accuracy in predicting anemia in the blood test dataset, while LGBMClassifier and Perceptron had a 97.61% accuracy for the eye image dataset.