Incremental Ensemble Gaussian Processes.

Belonging to the family of Bayesian nonparametrics, Gaussian process (GP) based approaches have well-documented merits not only in learning over a rich class of nonlinear functions, but also in quantifying the associated uncertainty. However, most GP methods rely on a single preselected kernel function, which may fall short in characterizing data samples that arrive sequentially in time-critical applications. To enable online kernel adaptation, the present work advocates an incremental ensemble (IE-) GP framework, where an EGP assembler employs an ensemble of GP learners, each having a unique kernel belonging to a prescribed kernel dictionary.

FULLY AUTOMATIC SEGMENTATION OF THE RIGHT VENTRICLE VIA MULTI-TASK DEEP NEURAL NETWORKS.

Segmentation of ventricles from cardiac magnetic resonance (MR) images is a key step to obtaining clinical parameters useful for prognosis of cardiac pathologies. To improve upon the performance of existing fully convolutional network (FCN) based automatic right ventricle (RV) segmentation approaches, a multi-task deep neural network (DNN) architecture is proposed. The multi-task model can employ any FCN as a building block, allows for leveraging shared features between different tasks, and can be efficiently trained end-to-end.

Type 2 diabetes mellitus is associated with obesity, smoking and low socioeconomic status in large and representative samples of rural, urban, and suburban adult Greek populations.

Objective: In this cross-sectional epidemiologic study we examined the association between type 2 diabetes mellitus and demographic, clinical, and socioeconomic parameters in large rural, urban and suburban populations of adult Greeks.

Design: Of the total target adult population (≥19 years, n=14233) in nine selected geographical regions covering rural, suburban, and urban areas of Greece, 10,647 subjects were included in the study. Data were collected by physicians who interviewed subjects at their homes between 1996 and 1999.