Analysis of glutathione supplementation effects on female metabolic syndrome condition using classification techniques / Nur Rasyidah Hasan Basri

• Main Author: Nur Rasyidah , Hasan Basri
• Format: Thesis
• Published: 2020
• Subjects: TA Engineering (General). Civil engineering (General)
• Online Access: http://studentsrepo.um.edu.my/14771/1/Nur_Rasyidah_Hasan_Basri.jpg; http://studentsrepo.um.edu.my/14771/3/KGA160047_ANALYSIS_OF_GLUTATHIONE_SUPPLEMENTATION_EFFECTS_ON_FEMALE_METABOLIC_SYNDROME_CONDITION_USING_CLASSIFICATION_TECHNIQUES.pdf; http://studentsrepo.um.edu.my/14771/

Lesser known to the public that main antioxidants in our body, Glutathione are also said to influence the metabolic syndrome (MS) condition. Oral supplementation consisting of glutathione precursors and vitamin C introduced in the study was to improve glutathione (GSH) status of the consumer and the main objective of this study is to investigate GSH effects on selected MS condition. Several known studies had proven that oral supplementation improved GSH level of consumer. However, there is no definite proof on how it’s affecting MS parameters. Also, there is less study carried out on the relationship between glutathione levels and metabolic syndrome especially among the Malaysian population. Prediction models will be constructed through classification technique to predict the MS parameter level based on GSH level and several other predictors. A randomized study was carried out on a total of 195 female volunteer subjects from Petaling Jaya, Malaysia and blood samples collected for analysis. Subjects were divided into 3 groups: Control group and 2 Intervention groups; Group 1 (consumed 1.6g of glutathione precursor’s supplementation daily) and Group 2 (consumed 3.2g of glutathione precursor’s supplementation daily). All data samples were used for model training purposes using three different classifiers; logistic regression, k-nearest neighbor, and decision tree). After 12 weeks, the supplementation were influencing the GSH level and some of the MS conditions, such as fasting glucose, triglycerides, LDL and total cholesterols showing that GSH alteration might closely related to the metabolic syndrome condition changes. Five predictors used in testing the significance of MS parameters is GSH, weight, body mass index (BMI) and waist hip ratio (WHR) and dosage groups. From the results, multiple variables models were found to be significant on MS conditions compared to single variable. Model 1 to Model 3 with more than three combined predictors show significant results on glucose level and triglyceride cholesterols level at p value less than 0.05. Overall, prediction models through logistic regression classifiers performed the best in classifying the MS condition (glucose and lipid profiles) into normal and abnormal level at accuracy of more than 80%.