Biometric-Based Blood Group Detection Using Deep Learning

Abstract

In emergency medical situations, rapidly determining a person\'s blood type is crucial, especially when prompt and safe blood transfusion is necessary. Conventional methods like serological testing, although dependable, often involve drawing blood and using laboratory equipment, which can delay results and limit accessibility in emergencies or remote areas lacking medical facilities.To address these issues, this research introduces a novel and non-inavasive approach that uses fingerprint patterns and deep learning to determine blood groups. A Convolutional Neural Network (CNN), known for its ability to recognize patterns in images, is used to examine fingerprint ridges and accurately classify both ABO and Rh blood groups.The system was developed and tested using a well-organized and accurately labeled dataset, allowing the model to learn effectively. The results revealed that the method performs with high accuracy, showing its potential as a practical alternative to traditional blood testing techniques. This approach is not only fast and easy to use but also adaptable for use in fieldwork, emergency healthcare, and mobile diagnostics where traditional testing is not feasible.Beyond emergency use, the fingerprint-based system could also be applied in health ID cards, biometric authentication systems, and automated hospital processes, providing healthcare professionals with quick access to critical patient information. This work contributes meaningfully to the development of AI-driven healthcare solutions by offering a unique integration of biometrics and medical diagnostics for enhanced patient care.

Introduction

Blood group identification is critical for safe blood transfusions, organ transplants, and emergency care but traditional blood typing methods are invasive, slow, and require lab facilities. To address these limitations, this research proposes a non-invasive, rapid system that predicts blood groups from fingerprint images using deep learning.

Leveraging Convolutional Neural Networks (CNNs), the system analyzes fingerprint patterns—linked genetically to blood groups—to classify individuals into all eight common ABO and Rh blood types. A curated dataset of fingerprint images labeled with blood groups was preprocessed and used to train a custom CNN model, achieving over 99% accuracy in classification.

The model was evaluated with metrics such as precision, recall, and F1-score, and demonstrated strong performance across all blood groups. An interactive, cloud-based interface was developed to allow easy fingerprint image upload and real-time blood group prediction, making the solution suitable for remote, emergency, or resource-limited settings.

This approach offers a promising, scalable, and non-invasive alternative for rapid blood group identification, with potential applications in field hospitals, rural healthcare, and emergency diagnostics.

Conclusion

This study presents a novel and non-invasive approach to blood group classification using fingerprint biometrics combined with deep learning techniques. By leveraging a custom-designed Convolutional Neural Network (CNN), the system effectively analyzes fingerprint ridge patterns and accurately predicts the ABO and Rh blood group types. The model was trained on a balanced dataset and achieved high classification performance, with validation accuracy exceeding 93% and precision and F1-scores surpassing conventional approaches. In addition to strong quantitative results, the system is supported by a practical user interface that enables real-time predictions, making it accessible and useful in emergency care, rural diagnostics, and point-of-care scenarios where traditional blood testing may not be feasible. The integration of fingerprint recognition with artificial intelligence in this context represents a significant advancement toward smart, contactless, and scalable healthcare solutions. Overall, the proposed system demonstrates that fingerprint-based biometric traits, when interpreted using deep learning models, hold great promise for enhancing the speed, accessibility, and reliability of medical diagnostics—offering a compelling alternative to traditional, lab-based blood typing methods.

References

[1] Vijaykumar Patil1.D. R. Ingle1 “An association between fingerprint patterns with blood group and lifestyle based diseases: a review” [1] [2] T Nihar1, K Yeswanth1 and K Prabhakar1 “Blood group determination using fingerprint”[2] [3] Zhu Wen , Songtong Han , Yongmin Yu , Xuemin Xiang , Shenzheng Lin , Xiaoling Xu, “Empowering robust biometric authentication: The fusion of deep learning and security image analysis ”[3] [4] Shervin Minaee, Amirali Abdolrashidi, Hang Su, Mohammed Bennamoun, David Zhang, “Biometrics Recognition Using Deep Learning: A Survey”[4] [5] Haruna Chiroma,“Deep Learning Algorithms based Fingerprint Authentication: Systematic Literature Review”[5] [6] Fahman Saeed ,Muhammad Hussain andHatimA.Aboalsamh, “Automatic Fingerprint Classification Using Deep Learning Technology (DeepFKTNet)[6] [7] S. A. Shaban and D. L. Elsheweikh, “ Blood Group Classification System Based on Image Processing Techniques”[7] [8] Anand Upadhyay 1, Jyotsna Anthal 2, Thangavel ,“ Identification of Human Blood Group Detection using Support Vector Machine and Image Processing”[8] [9] C.Sivamurugan, B.Perumal, Yelchuri Siddarthha, Vavilala Krishna Murthi, “Enhanced Blood Group Prediction with Fingerprint Images using Deep Learning”[9] [10] Louai A. Maghrabi(member, ieee), Mohammed Altwijri, Sami Saeed Binyamin, Fouadshoie Alallah, Diaa Hamed, and Mahmoud Ragab, “Secure Biometric Identification Using Orca Predators Algorithm With Deep Learning: Retinal Iris Image Analysis”[10] [11] Gh. Mohd. Bhat, M. Arif Mukhdoomi, Bahir Ahmed Shah, Mohd Saleem Ittoo “Dermatoglyphics: in health and disease - a review” in International Journal of Research in Medical Sciences, [11] [12] JAIN, L. C., HALICI, U., HAYASHI, I., LEE, S. B. and TSUTSUI, S,“Intelligent Biometric Techniques in Fingerprint and Face Recognition”, [12] [13] McBean RS, Hyland CA, Flower RL, “Approaches to determination of a full profile of blood group genotypes: single nucleotide variant mapping and massively parallel sequencing”, [13] [14] Ravindran G, Joby T, Pravin M, Pandiyan P, “Determination and classification of blood types using image processing techniques”, [14] [15] Fayrouz NE, Farida N, Irshad A H, “Relation between fingerprints and different blood groups”, [15] [16] Noor Eldin Fayrouz, Noor Farida, A.H. Irshad, “Relation between fingerprints and different blood groups”, [16] [17] Kanchan T, Chattopadhyay S, “Distribution of fingerprint patterns among medical students”, J Indian Acad Forens Med 28:65–68, 2006[17] [18] Karim JK, Mohammed AL, Saleem A, “Dermatoglyphics study of fingerprintspattern’s variations of a group of type II diabetic mellitus patients in erbil City”, [18]. [19] Keerthana D, Ranganathan L, “Design and development of blood sample analyzer using intelligent machine vision techniques”,[19]. [20] N. Zaeri, \"Minutiae-based Fingerprint Extraction and Recognition\", in Biometrics. London, United Kingdom”[20].

Copyright

Copyright © 2025 Vedulla Tejaswi, Dr. A Krishna Mohan. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.