Evaluating machine learning-powered classification algorithms which utilize variants in the GCKR gene to predict metabolic syndrome: Tehran Cardio-metabolic Genetics Study.

BACKGROUND: Metabolic syndrome (MetS) is a prevalent multifactorial disorder that can increase the risk of developing diabetes, cardiovascular diseases, and cancer. We aimed to compare different machine learning classification methods in predicting metabolic syndrome status as well as identifying influential genetic or environmental risk factors. METHODS: This candidate gene study was conducted on 4756 eligible participants from the Tehran Cardio-metabolic Genetic study (TCGS). We compared predictive models using logistic regression (LR), Random Forest (RF), decision tree (DT), support vector machines (SVM), and discriminant analyses. Demographic and clinical features, as well as variables regarding common GCKR gene polymorphisms, were included in the models. We used a 10-repeated tenfold cross-validation to evaluate model performance. RESULTS: 50.6% of participants had MetS. MetS was significantly associated with age, gender, schooling years, BMI, physical activity, rs780094, and rs780093 (P < 0.05) as indicated by LR. RF showed the best performance overall (AUC-ROC = 0.804, AUC-PR = 0.776, and Accuracy = 0.743) and indicated BMI, physical activity, and age to be the most influential model features. According to the DT, a person with BMI < 24 and physical activity < 8.8 possesses a 4% chance for MetS. In contrast, a person with BMI ≥ 25, physical activity < 2.7, and age ≥ 33, has 77% probability of suffering from MetS. CONCLUSION: Our findings indicated that, on average, machine learning models outperformed conventional statistical approaches for patient classification. These well-performing models may be used to develop future support systems that use a variety of data sources to identify persons at high risk of getting MetS.

Survival prediction of gastric cancer patients by Artificial Neural Network model.

AIM: This study aims to predict survival rate of gastric cancer patients and identify the effective factors related to it, using artificial neural network model. BACKGROUND: Gastric cancer is the most deadly disease in north and northeast provinces of Iran. A total of 430 patients with gastric cancer who referred to Baghban clinic in Sari, from early November 2006 to late October 2013 were followed. METHODS: A historical cohort of patients who referred to Baghban Clinic, the cancer research center of Mazandaran University of Medical Sciences in Sari, from early November 2006 to late October 2013 was studied. Three groups of variables (demographic, biological and socio-economic) were studied. Survival rate and effective factors on survival time were calculated using Kaplan-Meier methods and artificial neural networks and the best network structure were chosen using the mean square error and ROC curve. All analyses were performed using SPSS v.18.0 and the level of significance was selected α=0.05. RESULTS: In this research, the median survival time was 19±2.04 months. The 1 to 5-year survival rates for patients were 0.64, 0.44, 0.34, 0.24 and 0.19, respectively. The percentage of right predictions of the selected network and the area under the ROC curve were 92% and 94%, respectively. According to the results, the type of treatment, metastasis, stage of disease, histology grade, histology type and the age of diagnosis were effective factors on survival period. CONCLUSION: the 5 years survival rate of gastric cancer patients in Mazandaran is lower than other provinces which could be due to the delay in diagnosis or patient's referral. Therefore, the use of screening methods and early diagnosis could be influential for improving survival rate of these patients.