Microvascular Complications in Type-2 Diabetes: A Review of Statistical Techniques and Machine Learning Models

Type 2 Diabetes (T2D) has become a serious concern as it contributes to 90% of the total diabetic population. The high glucose levels in diabetic patients result in damage to eyes, kidneys and nerves collectively known as microvascular complications. The paper presents a critical review of existing statistical and machine learning models with respect to such complications namely retinopathy, neuropathy and nephropathy. The statistical tests like chi-square, odds ratio, t-test, ANOVA enables inferential and descriptive analysis. On the other side, machine learning models like support vector machines, k-nearest neighbour, deep neural network, naive bayes also assures predictive analytics. It is worth mentioning that both analytics can reverse impaired glucose regulation early in its course resulting in the prevention of long-term complications associated with T2D. Despite variation in both analytic techniques, their integration in medical health practices can assist patients to effectively adapt a healthy lifestyle and reduce significant healthcare overheads for an individual's family as well as society.