DETECTION OF BLOOD GROUP FROM FINGERPRINT USING CNN

Abstract: The prediction of a person's blood group based on their fingerprints represents an innovative approach that merges biometric identification with medical data. This project explores the use of Convolution Neural Networks (CNNs), a powerful class of deep learning algorithms, to predict an individual’s blood type from fingerprint images. The CNN is trained on a large dataset containing fingerprint images and their corresponding blood group labels. The model learns to recognize unique patterns, such as ridge and valley configurations, that may have subtle correlations with the individual’s blood type. According to preliminary findings, the CNN-based technology can attain encouraging accuracy levels, indicating a potential substitute for conventional blood group identification techniques. Enhancing model accuracy, growing the dataset, and resolving possible ethical and privacy issues with the use of biometric data will be the main goals of future study. While the primary goal of this research is to explore the feasibility of blood group prediction using fingerprints, the findings could have broader implications in fields like security, healthcare, and forensic investigations.

By learning visual patterns and agglutination characteristics present in the samples, the CNN model can classify blood types with high accuracy. This approach significantly reduces diagnostic time and human dependency, offering a scalable and efficient alternative to conventional methods. The results demonstrate the potential of AI-powered diagnostic tools in transforming healthcare delivery, particularly in remote or resource-limited settings. Future work will focus on expanding the dataset, refining the model, and validating the system in real- world clinical environments.

Keywords: Blood Group Prediction, Fingerprint Recognition, Convolution Neural Network (CNN), Deep Learning, Biometric Identification, Medical Image Classification, Pattern Recognition.