BioRidge: Unlocking Blood Group via Fingerprint Ridge Analysis

The method of identifying people using biometric information has grown in popularity for both security and medical applications. Fingerprint recognition is unique among biometric modalities. Medical diagnostics depend on blood group determinations. Traditional techniques require time-consuming, invasive procedures. This study proposes an invasive-free blood group identification system that utilizes Convolutional Neural Networks (CNNs) to analyze fingerprint images and achieve accurate predictions. This technology enables the integration of biometric and physiological data collection, especially useful in emergency and resource-constrained settings. The ResNet34 design proves to be the most effective choice for blood group detection.