This research paper aims to contribute to the field of non-invasive medical diagnostics. Blood group identification is a crucial aspect of medical diagnostics, emergency healthcare, and forensic investigations, as it plays a vital role in blood transfusions, organ transplants, and forensic identification. Traditional blood typing methods rely on serological testing, which involves chemical reagents, specialized laboratory equipment, and trained medical personnel. While these methods are accurate, they can be time-consuming, invasive, and impractical in remote or emergency scenarios. To overcome these challenges, this study presents a novel, non-invasive approach for blood group detection using fingerprint images and deep learning techniques. The research explores the hypothesis that there exists a correlation between fingerprint ridge patterns and blood groups, enabling automated classification without requiring a blood sample. A fingerprint dataset is utilized to train a deep learning-based model for blood group prediction. A Convolutional Neural Network (CNN) is employed to extract high-level features from fingerprint images, which are further analysed using supervised machine learning classifiers. The dataset undergoes a series of preprocessing steps, including noise reduction, contrast enhancement, edge detection, and feature scaling, to optimize the input for deep learning models. The CNN model is trained on labelled fingerprint images, and its performance is evaluated using key classification metrics such as accuracy, precision, recall, and F1-score. This research paves the way for further advancements in AIdriven medical diagnostics and bioinformatics, promoting innovation in healthcare technology.