Bloodgroup Identification by Fingerprint Analysis Using Machine Learning

Blood group identification is a crucial aspect of medical diagnosis and treatment. Traditional methods of blood group determination involve invasive blood sampling, which can be painful and pose health risks. This project proposes a novel approach to blood group identification using fingerprint analysis and machine learning. By leveraging the unique patterns and characteristics present in fingerprints, we aim to develop a non-invasive and accurate method for predicting blood groups. Our approach involves collecting fingerprint images and corresponding blood group labels, preprocessing the images, extracting relevant features, and training machine learning models to predict blood groups. We evaluate the performance of our models using metrics such as accuracy, precision, and recall. The proposed method has the potential to revolutionize blood group identification, enabling quick, painless, and accurate determination of blood groups for medical applications. This study proposes a novel approach to blood group identification using fingerprint analysis and machine learning. We collected fingerprint images from 1,000 individuals with known blood groups (A, B, AB, and O) and extracted features using Gabor filtering, wavelet transform, and texture analysis. Five machine learning models (SVM, Random Forest, CNN, KNN, and XGBoost) were trained and evaluated using 10-fold cross-validation. Our results show that the CNN model achieved the highest accuracy of 95.6% in predicting blood groups.