Development of a Blood Type Analyzer using Computer Vision and Machine Learning Techniques: A Review

摘要

In emergency cases, when the available time for blood transfusions is limited, blood type O (universal donor) is administered. However, sometimes, this can cause a transfusion reaction that can lead to the death of the patient receiving the transfusion. The equipment available on the market is not adequate for emergency scenarios (not portable and slow results). This paper presents the steps taken into consideration in the development of a blood type analyzer using computer vision and machine learning algorithms suitable for emergency situations (small size, lightweight, easy transportation, ease of use, fast results, high reliability and low cost). Several prototypes have been developed with the final version performing real world scenario experiments in hospitals for validation. With this system, it will be possible to contribute to the reduction of casualties in blood transfusions associated to human error or blood incompatibilities.