• C Siva Balaji Yadav
• B Harika
• A Chandrakala
• B Babu Kumar
• Dhurgesh V C, K Mounisha

Blood group detection is crucial in medical diagnosis, blood transfusion safety, and emergency medicine. This paper discusses a deep learning-based method for discrimination of blood groups based on fingerprint features. The study analyzes several deep learning architectures - Convolutional Neural Networks (CNNs), MobileNet, ResNet with Recurrent Neural Networks (RNNs), Vision Transformer models, in order to evaluate their performance on a publicly released fingerprint-based blood group dataset. The method adopted involves data preprocessing and using techniques of data augmentation to provide the model with robustness and generalization. Feature extraction has been taken from a baseline CNN, whereas MobileNet was focused on computation as being lightweight and efficient. ResNet+RNN used residual learning to integrate sequential patterns of recognition into an architecture to boost the classification accuracy. Lastly, the attention mechanism of Vision Transformer model uses intricate details about fingerprint significantly enhancing the blood group classification. Experimental results show that Vision Transformer outperforms other architectures, achieving state-of-the-art accuracy and indicating that it is indeed focusing on the fingerprint features that matter. The approach proposed here brings about a novel and efficient solution for the identification of blood groups using the latest machine learning techniques. The study contributes to the growing area of data-driven healthcare research and offers a scalable framework for similar classification tasks.

Fingerprint patterns, blood group classification, deep learning, CNN, MobileNet, ResNet, RNN, Vision Transformer, classification accuracy, healthcare research.

"INNOVATIVE BLOOD GROUP DETECTION: A DEEP LEARNING FRAMEWORK USING FINGERPRINT ANALYSIS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 4, page no.f361-f366, April-2025, Available :http://www.jetir.org/papers/JETIR2504547.pdf

"INNOVATIVE BLOOD GROUP DETECTION: A DEEP LEARNING FRAMEWORK USING FINGERPRINT ANALYSIS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 4, page no. ppf361-f366, April-2025, Available at : http://www.jetir.org/papers/JETIR2504547.pdf