Ensemble approach on predicting blood platelets using supervised approach via computational bio informatics

Abstract

Platelets are little leukocytes that are essential for such blood coagulation. When an individual does not generate sufficient platelets they won't function effectively, then they will likely to have frequent haemorrhage and bruising since normal deterioration towards the exterior of blood vessels can go inoperable. This may be prescribed to accommodate patients using platelet transfusions via blood donors to halt such haemorrhage. Unfortunately, such medication is not optimal since patients develop antigens against given blood platelets, preventing them from being managed with platelets of similar blood groups in the future. As a result, blood centres are looking towards finding alternatives to donor platelets. Platelet transplants are commonly utilised to reduce or eliminate haemorrhage in a range of patient groups, including those that are currently haemorrhage and those who have malfunctioning or inadequate platelets. Because of the unpredictability of demands, commodity highly perishable, and cost, all institutions and health centres must prioritise estimating hospital-wide platelet consumption. Also, with fast advancement of ML technology, now it is extensively applied in the treatment of a variety of disorders. The benefit of machine learning methods is how they can interpret high-order range of predictor correlations and produce more solid predictions.