A data-driven approach to predict the risk of readmission among patients with Diabetes Mellitus

Diabetes is infamous for clutching individuals into the disarray of health degradation. The count of affected patients is rising with each passing day and an increasing number of them are afflicted with that variety of this disease which is of the incurable kind. The high cost of treatment is another major shortcoming associated with this despicable matter. These cases require immediate attention and sitting on the fence cannot be an option with respect to treatment procedures as wrong treatments can lead to early readmission. This can be very expensive for the patients and it begs the need to look for solutions that can help avoid such situations. Thus, predicting the readmission of patients is a leading matter of concern with respect to both treatment and cost effectiveness. To this end, we review the literature and develop a novel data-driven approach that draws from the previous works to make better predictions. It helps to find hidden dependencies in the data to outperform basic methods.