Non-Invasive Blood Group Detection using Fingerprint Patterns with Artificial Intelligence and Machine Learning

The work proposes a novel, non-invasive method for identifying blood groups using fingerprint patterns through the application of artificial intelligence and machine learning techniques. The framework investigates the relationship between dermatoglyphic characteristics—such as ridge flow, minutiae patterns, texture features—and standard blood group categories (A, B, O, and AB). High-resolution fingerprint images are collected and subjected to preprocessing steps, including noise reduction, contrast enhancement, and ridge normalization, to improve the quality and consistency of feature extraction. Subsequently, a combination of statistical, structural, and textural features is extracted and utilized to train supervised machine learning models for blood group classification. The proposed system is evaluated using a well-structured dataset, where it demonstrates encouraging levels of prediction accuracy while minimizing reliance on traditional invasive blood testing methods. Overall, this approach provides a fast, cost-efficient, and user-friendly solution for preliminary blood group estimation, with potential applications in healthcare screening, emergency response scenarios, and biometric-based medical support systems.