Related Experiment Video
Updated: May 21, 2026
12:21
A Versatile Automated Platform for Micro-scale Cell Stimulation Experiments
Published on: August 6, 2013
Automated blood group classification using a digital microfluidics chip and vision transformer-based image analysis.
Syeda Sana Bukhari1, Aleena Khan2, Khansa2
1Faculty of Computer Science and Engineering, Ghulam Ishaq Khan Institute of Engineering Sciences and Technology, Topi, 23640, Pakistan.
Biomedical Microdevices
|May 20, 2026
Summary
This study introduces an AI-powered blood typing system using digital microfluidics and deep learning. The intelligent system accurately identifies blood groups, improving safety and reducing costs for medical diagnostics.
Area of Science:
- Biomedical Engineering
- Artificial Intelligence
- Hematology
Background:
- Accurate blood group identification is vital for safe transfusions and medical procedures.
- Traditional blood typing methods rely on visual agglutination, which can be subjective and error-prone, especially with small sample volumes.
Purpose of the Study:
- To develop an efficient and intelligent system for automated blood type detection.
- To improve the accuracy, reliability, and cost-effectiveness of blood group identification.
Main Methods:
- Integration of digital microfluidics with deep learning (Vision Transformer model) for agglutination pattern recognition.
- Automated antigen-based decision logic for ABO/Rh blood group assignment.
- Utilized droplet-based blood imaging for AI analysis.
Main Results:
- Achieved 100% accuracy in detecting antigen-antibody agglutination reactions.
- Perfect scores for precision, recall, specificity, and F1-score in blood group determination.
- Demonstrated reduction in reagent volume and testing costs.
Conclusions:
- The developed AI-driven microfluidic system offers a highly accurate and reliable method for blood typing.
- Future work aims to create a portable, paper-based device for low-cost, AI-assisted point-of-care blood typing.
