Print2type: Revolutionizing Blood Group Detection Through Fingerprint Analysis and ML

This paper presents Print2Type, a novel machine learning-based system that predicts an individual’s blood group using fingerprint analysis. Traditional blood typing requires invasive sampling, specialized reagents, and laboratory equipment, whereas our approach leverages image processing and classification algorithms to identify subtle biometric patterns that correlate with blood type, offering a non-invasive, rapid, and scalable solution for blood group detection. Using a dataset of 6,000 fingerprint images across eight blood groups, we developed a CNN model incorporating image enhancement and feature extraction techniques. Our system achieved 91 percentage test accuracy with sub-second inference time, demonstrating the feasibility of biometric-based blood group determination which is a comfortable alternative for clinical diagnostics, remote healthcare, and emergency situations where conventional blood typing is challenging or unavailable.