RESUMEN
Good feature engineering is a prerequisite for accurate classification, especially in challenging scenarios such as detecting the breathing of living persons trapped under building rubble using bioradar. Unlike monitoring patients' breathing through the air, the measuring conditions of a rescue bioradar are very complex. The ultimate goal of search and rescue is to determine the presence of a living person, which requires extracting representative features that can distinguish measurements with the presence of a person and without. To address this challenge, we conducted a bioradar test scenario under laboratory conditions and decomposed the radar signal into different range intervals to derive multiple virtual scenes from the real one. We then extracted physical and statistical quantitative features that represent a measurement, aiming to find those features that are robust to the complexity of rescue-radar measuring conditions, including different rubble sites, breathing rates, signal strengths, and short-duration disturbances. To this end, we utilized two methods, Analysis of Variance (ANOVA), and Minimum Redundancy Maximum Relevance (MRMR), to analyze the significance of the extracted features. We then trained the classification model using a linear kernel support vector machine (SVM). As the main result of this work, we identified an optimal feature set of four features based on the feature ranking and the improvement in the classification accuracy of the SVM model. These four features are related to four different physical quantities and independent from different rubble sites.
Asunto(s)
Radar , Frecuencia Respiratoria , Humanos , Máquina de Vectores de SoporteRESUMEN
By using the statistical techniques of the ANOVA means test and regression, it was found that theNKRcalibration factor of Standard Imaging (SI) model HDR 1000 plus chambers presents a quadratic dependence with the Reference air kerma rateKR(from 6.9 mGy h-1to 43.9 mGy h-1). In order to understand and correct this dependency one model is presented for total recombination:ks=I300I150=1+kini+kd+kvol·I300+kscreen·I3002,wherekiniis the initial recombination,kvolthe thermal diffusion recombination,kvolthe volumetric recombination andkscreenthe screening for the currents/charges collected at the potential differences of 300 and 150 V. In conclusion, the total recombinationksis composed by onekiniwith a constant contribution of 0.019%, onekdcontribution of 0.017%, onekvol·I300contribution from 0.022% to 0.138%, and thekscreen·I3002effects from 0.002% to 0.09% in the range ofKÌRrate above. However, when this model forksis applied to try to correct the quadratic dependence of theNKRversusKR,explicitly there is no improvement in the variation range of 0.5% of theNKRversusKR.Nonetheless, it allows to obtainNKRvalues consistent with auc ≤ 0.7%, which is less than 1.25% reported in the literature by ADCLs or SSDLs.
RESUMEN
17α-ethynylestradiol (EE2), a ubiquitous synthetic endocrine disrupting chemical, was the principal component of contraceptive drugs and one of common hormone medications. The detrimental impact of EE2 on the reproduction of organisms was widely recognized. However, the underlying mechanisms of physiological and metabolome effects of EE2 on freshwater fish are still unclear. This study investigated the toxic effects and related mechanisms of EE2 on freshwater fish crucian carp (Carassius auratus) based on metabolomics. Crucian carp were exposed to EE2 at environmentally relevant concentration for 9 days, 18 days, and 27 days, and the biological responses were explored through analysis of the physiological endpoints, steroid hormones, and metabolome. The physiological endpoints of crucian carp had no distinct change after EE2 exposure. However, metabolomics analysis probed significant deviation based on chemometrics, indicating that the metabolomics approach was more sensitive to the effects of EE2 at environmentally relevant concentration to freshwater fish than the traditional endpoints. The alterations of 24 metabolites in gonad and 16 metabolites in kidney were induced by treatment with EE2, respectively, which suggesting the perturbations in amino acid metabolism, lipid metabolism, energy metabolism, and oxidative stress. Moreover, EE2 exposure could induce the disruption of lipid metabolism and then broke the homeostasis of endogenous steroid hormones. Metabolomics provided a new strategy for the studies on contaminant exposure at a low dose in the short term and gave important information for the toxicology and mechanism of EE2.