MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202641117855 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering And Technology on October 01, 2026, for An Artificial Intelligence-Based System And Method For Blood Group Identification Using Fingerprint Image Analysis.
Inventor includes Dr. A. Kousar Nikhath.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: ABSTRACT [0020]The proposed invention enables deep-learning-based system to identify and predict various blood groups from the given fingerprint images. Image is collected and preprocessed by grayscale conversion, histogram equalization, resizing, and normalization. Five convolutional neural network architectures ResNet50, MobileNetV2, DenseNet121, EfficientNetB0, and a custom CNN are trained and comparatively evaluated. The custom CNN achieves the highest test accuracy of 89.07 percent. The best-performing model is integrated into a FastAPI-based prediction service that accepts uploaded fingerprint images through a web interface, applies the same preprocessing pipeline, and returns the predicted blood group together with a confidence score. The resulting non-invasive, automated approach offers a practical alternative to conventional serological testing for applications in forensic investigations, biometric authentication, and rapid medical triage.
Disclaimer: Curated by HT Syndication.
