Data to Diagnosis Machine Learning Models for Accurate Anemia Classifications

Anaemia classification presents significant challenges due to limited labeled medical data and complex feature interrelations. This study introduces a hybridframeworkthatcombinesTransferLearning with Exterme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) to enhance the accuracy of anaemia subtype classification. A pretrained convolutional neural network (CNN) model is fine-tuned using a haematological dataset comprising [insert number] patient samples, encompassing features such as haemoglobin levels, red blood cell indices, and additional clinical biomarkers. The CNN extracts deep, informative features, which are subsequently processed by XGBoost and LightGBM classifiers. Thismethodology effectively addresses the constraints of small and imbalanced medical datasets while capturing intricate feature relationships. The proposed approach achievedaclassificationaccuracyof[insertaccuracy]%, surpassingtraditionalmodelslikelogisticregressionand random forest. Statistical validation, including /insert statistical tests, e.g., <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$k$</tex>-fold cross-validation], confirmed therobustnessand significance of theresults. This dualmodel framework enables precise classification of anaemia subtypes-such as iron-deficiency anaemia, vitamin-deficiency anaemia, and haemolytic anaemiafacilitating early diagnosis and improved patient outcomes.