Identification of ABO Using Fingerprint Pattern Analysis

Blood group identification is a crucial requirement in medical diagnosis, emergency healthcare services, and blood transfusion processes. Conventional blood group determination methods are invasive, time-consuming, and require laboratory infrastructure. This paper presents a non-invasive and automated approach for predicting human blood groups using fingerprint pattern analysis and deep learning techniques. Fingerprint images are preprocessed and classified using a Convolutional Neural Network (CNN) to predict ABO and Rh blood groups. The trained model is saved in blood_group_classifier_split.keras and blood_group_classifier_split.h5 formats for training, validation, and deployment. The system is implemented with a web-based interface for real-time prediction. Experimental results show high accuracy and confidence scores, demonstrating that fingerprint pattern analysis can serve as an efficient alternative for blood group prediction in academic and healthcare research applications.