Analysing the risk level of diabetic neuropathy using data mining techniques

Abstract

Diabetic neuropathy is one of the severe complications that damage diabetic patient’s nerves throughout their bodies. This complication will cause pain and numbness in the legs and feet and affect their digestive system, urinary tract, blood vessels, and heart. This may affect as many as 50% of people; however, diabetic patients can prevent this complication or slow its progress with consistent blood sugar management and a healthy lifestyle.The main objective of this research is to analyse the risk level of getting diabetic europathy, and using these results; patients can perfectly manage their blood sugar level and get medical treatment before getting any serious medical issue. Mainly, data mining methodologies are used to build this model. In this model, a k-means algorithm was applied to analyse diabetes patient databases by taking various attributes of diabetes to predict diabetes disease. There are five main factors considered to build this model and considered factors are haemoglobin A1c (HbA1c), fasting blood sugar, cholesterol level, gender, and age. The finding from this study suggests that the data set is possibly divided into three main clusters with high accuracy. With ten iterations, the data set could be successfully clustered into three clusters, namely low-risk level (cluster 0), high-risk level (cluster 1), and intermediate-risk level (cluster 2) using the k-means algorithm. urthermore, 6772 patients’ data divide into several groups; cluster 0 with 5377, cluster 1 with 318, and cluster 2 with 1077.