Copyright © 2026 Authors retain the copyright of this article. This article 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.

@article{189616,
        author = {Rahul Patel},
        title = {Blood Group Detection Using Infrared Hand Images and Machine Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {7},
        pages = {6590-6599},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=189616},
        abstract = {Blood group identification is a fundamental require-ment in healthcare systems, particularly in emergency medicine, blood transfusion services, trauma care, and surgical procedures. Conventional blood group determination techniques rely on invasive laboratory-based serological tests that require blood extraction, chemical reagents, trained personnel, and controlled environments. These constraints limit their applicability in emer-gency scenarios, remote healthcare facilities, and point-of-care diagnostics.
This paper presents a non-invasive and automated blood group detection system using infrared hand images and deep learning techniques. The proposed system leverages thermal patterns and vascular characteristics captured through infrared imaging, com-bined with a convolutional neural network based on a pretrained VGG16 architecture. Due to the limited availability of real-world infrared blood group datasets, synthetic data generation and temperature-based infrared modeling are employed to expand the dataset and improve model generalization. The system classifies eight blood groups including Rh-positive and Rh-negative types. Extensive experiments are conducted on a dataset of 4000 infrared hand images using accuracy, precision, recall, and F1-score as evaluation metrics. The proposed approach achieves a test accuracy of 87.34%. Furthermore, a web-based application is developed to demonstrate real-time blood group prediction. The experimental results indicate that infrared imaging combined with deep learning offers a promising alternative to traditional invasive blood group detection methods.},
        keywords = {Blood group detection, infrared imaging, deep learning, convolutional neural networks, VGG16, non-invasive diagnostics, medical image processing},
        month = {December},
        }