This study aimed to estimate scattered radiation and its spatial distribution around three Cone Beam Computed Tomography (CBCT) devices, in order to determine relatively safe positions for an operator to stand if needed to be inside the CBCT room. The following devices were tested: Morita Accuitomo (CBCT1), Newtom Giano HR (CBCT2), Newtom VGi (CBCT3). Scattered radiation measurements were performed using different kVp, mA and Field of View (FOV) options. An anthropomorphic phantom (NATHANIA) was placed inside the x-ray gantry to simulate clinical conditions. Scattered measurements were taken with the ionization chamber Inovision model 451P Victoreen once placed at fixed distances from each irradiation isocenter, away from the primary beam. A statistically significant (p<0.001) difference was found in the mean value of the scattered radiation estimations between the CBCT devices. Scattered radiation was reduced with a different rate for each CBCT device as distance was increased. For CBCT1 the reduction was 0.047μGy, for the CBCT2 0.036 μGy and 0.079 μGy for CBCT3, for every one meter away from the x-ray gantry. Therefore, at certain distances from the central x-ray, the scattered radiation was below the critical 1mGy. Consequently, an operator could stay inside the room accompanying the patient being scanned, if necessary.

As Artificial Intelligence (AI) technology, particularly Machine Learning algorithms, becomes increasingly ubiquitous, our abilities to understand and interpret AI and machine learning algorithms are becoming increasingly desirable. Visualization has been used as a common tool to view and understand complex machine learning processes. In this paper, we will focus on developing a general framework for the visualization of machine learning models as a scalar valued multi-dimensional function to help users understand how the models behave over different viewing spaces. We first give a formal definition of the visualization problem. Interpolation-based morphing and subspace sampling techniques are applied to generate various renderings through projections and cross-sections of the model space as 3D surfaces or heatmap images. This method will be applied to two real-world datasets and applications: the diagnosis of Alzheimer's Disease (AD) using a human brain networks dataset and a real-world benchmark dataset for predicting home credit default risks. The visualizations show that different machine learning algorithms can behave quite differently under different conditions.

This paper contains a concise overview of the deployment of scattered solar power plants in Poland, mainly from the perspective of their communication networks, and how the recent development of the Polish 4G networks has a very positive impact for the performance of the whole monitoring system (production control and video-surveillance), with a special emphasis on video-analytics, due to its higher bandwidth demand. All the information will be shown from the point of view of the solar photovoltaics developer I+D Energías, and therefore it constitutes a real user’s experience.