Predicting Mortality Risk for COVID in Diabetic and Non-Diabetic Patients using Machine Learning: A Systematic Review

In an effort to control and lessen the disaster caused by the COVID-19 pandemic, which has lately created a global emergency, scientists, physicians, and other healthcare professionals from all over the world have been constantly investigating cutting-edge technology. The demonstrated effectiveness of machine learning (ML) and artificial intelligence (AI) in previous outbreaks has led to the presentation of new concepts in the fight against the novel coronavirus. This systematic study examines the predictive potential of machine learning models in estimating the mortality risk among COVID-19 patients, primarily comparing populations with and without diabetes. Furthermore, our investigation delves into the methodology, datasets, and performance Following a thorough review of the literature and databases, we assessed studies that used machine learning techniques for mortality prediction, including logistic regression, random forest, XGBoost, support vector machines, and neural networks. The review emphasizes this issue since diabetes mellitus significantly raises the risk of serious complications, prolonged hospitalization, and fatality in COVID-19 cases. metrics that were utilized in a variety of research, with a particular focus on the function that machine learning plays in improving early risk assessment and clinical decision-making. Based on these findings, it is clear that personalized healthcare methods and machine learning-based tools are required in order to properly identify individuals who are at a high risk of developing diabetes. In the end, incorporating artificial intelligence and machine learning into clinical workflows presents a viable route for enhancing outcomes and resource allocation during ongoing and emergency situations in the healthcare industry.