In the recent years there have been a number of studies that applied deep learning algorithms to neuroimaging data. Pipelines used in those studies mostly require multiple processing steps for feature extraction, although modern advancements in deep learning for image classification can provide a powerful framework for automatic feature generation and more straightforward analysis. In this paper, we show how similar performance can be achieved skipping these feature extraction steps with the residual and plain 3D convolutional neural network architectures. We demonstrate the performance of the proposed approach for classification of Alzheimer's disease versus mild cognitive impairment and normal controls on the Alzheimers Disease National Initiative (ADNI) dataset of 3D structural MRI brain scans.

No abstract

No abstract

This article surveys the current achievements concerning smart thermostats for smart homes and smart buildings and observes the level of smartness of the microclimate control systems for buildings and rooms. The article shows the results of analyzes the advantages and disadvantages of the marketed solutions and sets requirements for smart thermostat as a component of building automation system. All solutions were classified at three "smartness levels", from zero to second, where zero level means simple manually adjusted temperature regulator, first level assumes working according the programmable schedule. The second level smart thermostat automatically generates and dynamically changes its schedule basing on the users' activity and presence, as well as on user-defined priorities. The article sets goals for further research in the field of self-learning algorithms for smart thermostat, because it seems actual to develop a self-learning smart thermostat capable of supporting a large buildings; it should analyze user behavior and effect control both at the level of a whole building and at the level of single rooms.