Machine learning algorithms for Diabetes prediction and neural network method for blood glucose measurement

Objectives: To facilitate painless and easy method for prediction of diabetes with high accuracy to measure blood glucose by noninvasive using Photoplethysmography (PPG). Method: In this study, is done different machine learning algorithms on a dataset created samples from PIMA Indian Diabetes in vivo dataset. Machine used are Support Vector (SVM), Decision Tree, Naïve Bayes Classifier K Nearest Neighbor (KNN). A PPG data 182 individuals recorded over 1 minute duration each. Various frequency time domain features single extracted pulse wave analysis. Neural network trained measurement performed. Findings: With decision tree algorithm we got highest 89.97% it proves be good consideration treatment.Clarke Error Grid analysis clinically accepted, so performed similar Using features, 94.27 % points accepted regions (Region Region B). Novelty: Based the collected or analyzed, our results, encouraging see that further research may lead affordable detection at early stage. Keywords: Noninvasive; glucose; NIR; algorithms; photoplethysmography