The fingerprint pattern has long been recognized as one of the most reliable and unique biometric identifiers of an individual. Unlike other physical or behavioral characteristics, fingers are consistent in a person’s life, making them an invaluable tool in forensic science, safety, and identity verification. In legal contexts, fingerprint evidence is significant because no two persons, even genetically identical twins, see the same fingerprint pattern. The possibility of two fingerprints being the same is unlikely, estimated to be approximately one in sixty-four billion. This uniqueness has made the foundation fingers of criminal investigation and personal identity systems worldwide. Recent studies have explored the possibility of a link between fingerprint formations and certain biological traits, including blood group classifications. The ABO system, which divides blood into A, B, AB, and O types depending on specific antigenic markers, plays a vital role in healthcare applications such as blood transfusion, disease diagnosis, and hereditary research. Some studies have suggested that the ridge pattern of a person’s fingerprint and their blood group may be linked, which opens the possibility of predicting blood groups using fingerprint analysis. The study’s results display the superiority of the CNN approach over the traditional SVM algorithm. ABO from the proposed CNN architecture fingerprint images receives an impressive accuracy of 91% in predicting blood groups, making the SVM model much better. This high level of accuracy underlines the ability of deep teaching techniques to highlight the micro-relationship between biometric symptoms and physical characteristics.