The blood groups prediction and categorization may be useful in various medical and emergent studies including blood transfusion, organ transplant, and forensic examinations. Normal blood group forecasting and classification procedures are carried out in a lab, In other words it is by serological method. The techniques are time consuming and inaccurate as it is carried out manually by one person or a team of qualified experts in a laboratory. In this regard, it has been proposed to envision and cluster blood groups through the help of AI technology using microscopic pictures of blood samples as one of the technological innovations of medical and science research. These processes entail the analysis and the extraction of primary images in blood samples to a system of neural networks comprising deep-learning through CNNs to identify patterns in red blood cells to categorize them into four major blood groups which are the A, B, and AB and O blood groups according to massive labeled pictures of blood samples.