Blood Group Detection Using Fingerprint

Blood Group Detection Using Fingerprint

 

 

Dr. Sweta Raut

Department computer Science and Engineering

Jhulelal Institute Of Technology

Nagpur, India

s.raut@jitnagpurgmail.com

Miss. Gunjan Mankar

Department Computer Science and Engineering

Jhulelal Institute Of Technology

Nagpur, India gunjanmankar47@gmial.com

Miss. Sharayu Nistana

Department Computer Science and Engineering

Jhulelal Institute Of Technology

Nagpur, India

Sharayunistana14@gmail.com

Miss. Vasundhara Shivhare

Department Computer Science and Engineering

Jhulelal Institute Of Technology

Nagpur, India

shivharevasundhara@gmail.com

Miss. Shreya Sayare

Department Computer Science and Engineering

Jhulelal Institute Of Technology Nagpur, India shreyasayare@gmail.com

Miss. Saloni Deorankar

Department Computer Science and Engineering

Jhulelal Institute Of Technology

Nagpur, India

 deorankarp@gmail.com

 

 

Abstract— Blood group identification is crucial for safe medical procedures, but traditional techniques are based on invasive blood draws, which are risky and inconvenient. This research presents a new, non-invasive method to identify ABO and Rh blood groups from fingerprints. By taking advantage of the biochemical residues in fingerprint sweat, which include blood group-specific antigens, we integrate Fourier-transform infrared (FTIR) spectroscopy and machine learning to identify blood groups. 200 participants with known blood groups provided spectroscopic analysis fingerprint samples to distinguish spectral biomarkers. Supervised learning algorithms based on support vector machines (SVM) and convolutional neural networks (CNN) were then trained to align spectral patterns against blood group antigens. A total accuracy rate of 94.5% was achieved through the suggested methodology, with the precision rate crossing 92% for ABO group discrimination and 96% for Rh factor classification. Comparative verification established its reliability over conventional serological tests. Environmental contamination and inter-individual variability of sweat were overcome with sophisticated preprocessing and feature selection. This non-invasive method exhibits great promise for rapid, cost-efficient blood group screening in clinical, forensic, and resource-scarce contexts. By fusing biometric and  biochemical information, this study opens the way for portable, fingerprint-based diagnostic devices, maximizing patient comfort and operational effectiveness within healthcare systems

Keywords— Fingerprint recognition, Blood group

      identification, Pattern recognition, Image processing

,Machine learning, Artificial intelligence, Healthcare

technology, Medical diagnostics, Personalized medicine, Fingerprint analysis, Vein pattern recognition, Dermatoglyphics, Biometric authentication, Health informatics