Non-invasive A,B and O Blood Group Identification from Ocular Images Using a Hybrid Multi-modal Deep Learning Approach

Key Points

• Blood group identification was successfully achieved using ocular images, enhancing the efficiency of non-invasive methods.
• The system demonstrated an accuracy of over 90% in blood group classification, validating its effectiveness in practical applications.
• Utilizing a hybrid approach, deep learning models processed various ocular image datasets for optimal performance.
• This method highlights the potential of applying image processing and biometrics for safer blood group determination.