Simulation the distribution of thermodynamic temperatures and microwave radiation of the human head.

BACKGROUND AND OBJECTIVE: The influence of biophysical parameters on the formation of microwave radiation of the human head is poorly studied. Existing approaches to modeling microwave radiation of the human head have limitations associated with simplifying the geometry of human anatomy. The article proposes methodological solutions for numerical modeling of microwave radiation of the brain biological tissues using the geometry obtained from MRI data. METHODS: The geometrical characteristics of biological tissues in model are determined using an MRI image of the head. The methodology proposed in the article allows simulation of a human body voxel models performed the Pennes bio-heat transfer equation using the Fenix software package. RESULTS: Modeling evaluations have shown that anatomical tissues heterogeneities on the surface of the head form temperature gradient of up to 2.0 K, and changes of the microwave radiation up to 0.3 K. CONCLUSIONS: Verification data made by IR thermograph practically coincide with the results of numerical modeling. The fluctuations of the brain microwave radiation are not only the result of thermal processes in its tissues, but are determined by the dynamics of its thermoregulation processes and are an indicator of changes in the physiological processes occurring in it.

Comparison of Machine Learning Methods for the Arterial Hypertension Diagnostics.

The paper presents results of machine learning approach accuracy applied analysis of cardiac activity. The study evaluates the diagnostics possibilities of the arterial hypertension by means of the short-term heart rate variability signals. Two groups were studied: 30 relatively healthy volunteers and 40 patients suffering from the arterial hypertension of II-III degree. The following machine learning approaches were studied: linear and quadratic discriminant analysis, k-nearest neighbors, support vector machine with radial basis, decision trees, and naive Bayes classifier. Moreover, in the study, different methods of feature extraction are analyzed: statistical, spectral, wavelet, and multifractal. All in all, 53 features were investigated. Investigation results show that discriminant analysis achieves the highest classification accuracy. The suggested approach of noncorrelated feature set search achieved higher results than data set based on the principal components.