Methods: This was a post-hoc case-control exploratory sub-analysis of a cross-sectional study among GDM women to determine which candidate single nucleotide polymorphisms (SNPs) related to neuroendocrine disorders may be associated with obesity. Factors were adjusted for socio-demographic characteristics and concurrent medical problems in this particular population. Pre-pregnancy BMI and concurrent medical profiles were obtained from maternal health records. Obesity is defined as BMI of ≥27.5 kg/m2 for Asian criteria-based BMI and >30 kg/m2 for International criteria-based BMI. Thirteen candidate genes were genotyped using Agena® MassARRAY and examined for association with pre-pregnancy obesity using multiple logistic regression analysis. The significant difference threshold was set at P value <0.05.

Results: Three hundred and twelve GDM women were included in this study; 60.9% and 44.2% of GDM patients were obese using Asian and International criteria-based BMI, respectively. GDM patients with AA or AG genotypes in specific SNP of brain-derived neurotrophic factor (BDNF) (G > A in rs6265) are more likely to be obese (adjusted odd ratio =2.209, 95% CI, 1.305, 3.739, P=0.003) compared to those who carry the GG genotype in the SNP adjusted for parity, underlying with asthma, heart disease, anaemia, education background in the International criteria-based BMI stratification group. On the other hand, there were no associations between other candidate genes (NRG1, FKBP5, RORA, OXTR, PLEKHG1, HTR2C, LHPP, SDK2, TEX51, EPHX2, NPY5R and ANO2) and maternal obesity.

METHOD: In this work, resting-state EEG-derived features were utilized as input data to the proposed feature selection and classification method. The aim was to perform automatic classification of AUD patients and healthy controls. The validation of the proposed method involved real-EEG data acquired from 30 AUD patients and 30 age-matched healthy controls. The resting-state EEG-derived features such as synchronization likelihood (SL) were computed involving 19 scalp locations resulted into 513 features. Furthermore, the features were rank-ordered to select the most discriminant features involving a rank-based feature selection method according to a criterion, i.e., receiver operating characteristics (ROC). Consequently, a reduced set of most discriminant features was identified and utilized further during classification of AUD patients and healthy controls. In this study, three different classification models such as Support Vector Machine (SVM), Naïve Bayesian (NB), and Logistic Regression (LR) were used.

RESULTS: The study resulted into SVM classification accuracy=98%, sensitivity=99.9%, specificity=95%, and f-measure=0.97; LR classification accuracy=91.7%, sensitivity=86.66%, specificity=96.6%, and f-measure=0.90; NB classification accuracy=93.6%, sensitivity=100%, specificity=87.9%, and f-measure=0.95.