46 research outputs found
Bayesian Optimization based Hyperparameter Tuning of Ensemble Regression Models in Smart City Air Quality Monitoring Data Analytics
No full textThis is the author accepted manuscript. The final version is available from IEEE via the DOI in this recordThis paper uses the Bayesian optimization for fitting Ensemble regression models for tuning the machine learning model hyperparameters with reduced computation. We use the Pune Smart City air quality monitoring dataset with temporal variation of hazardous chemical pollutants in the air. The aim here is to reliably predict the suspended particulates as the air quality metrics using other environmental variables, considering linear models and nonlinear ensemble of tree models. To achieve good predictive accuracy a computationally expensive optimization method is required which has been achieved using the Gaussian Process surrogate assisted Bayesian optimization. We also show the diagnostics plots of the residuals from the nonlinear models to explain model quality.European Regional Development Fund (ERDF
Estimation of the residential radon levels and the population annual effective dose in dwellings of Al-kharj, Saudi Arabia
No full textAbstractIndoor radon levels and the annual effective dose are measured in Al-kharj city, Saudi Arabia dwellings using CR-39 detector. The dwellings are classified according their types (schools, homes and working area). The influence of some factors like number of floors and ventilation conditions on indoor radon levels, equilibrium factor and radon effective doses were studied. Can and bare method is used for determine the equilibrium factor between radon and its daughters. Based on the dosemetric approach and epidemiological determinations conversions convention for radon exposures, the annual effective doses are calculated and compared. The average radon concentration varies from 76 ± 38 Bq m−3 in work places to 114 ± 41 Bq m−3 in homes. About 77% of the studied dwellings give radon concentration in the range from 50 to 150 Bq m−3. The overall weighted mean of radon level is equal to 94 ± 41 Bq m−3 which about 2.5 times the global average. The equilibrium factor has a wide range from 0.1 to 0.6 with overall weighted average equal to 0.308 ± 0.13. The variety of living style, constructed materials and ventilation rates are responsible for this wide range and subsequently the obtained high uncertainty (42%). Homes showed larger annual effective dose (3.186 ± 0.75 mSv) than other dwellings which locate in the range of the recommended action level but about three times the global average. The result shows that the ventilation condition is the major but not the only factor affects the results. Poor ventilated dwellings showed the maximum annual effective dose on the other hand the number of floor has insignificant difference
Radon levels and the expected population mortality in dwellings of Al-Kharj, Saudi Arabia
No full textAbstractInternal exposure from inhalation of radon and its progeny is one of the most significant sources of natural radiation exposure of the population. Radon levels and radon equilibrium factor were measured in the dwellings of Al-Kharj, Saudi Arabia using passive technique. Calibrated CR-39 diffusion type radon detectors were used for radon measurements and the method of can and bare is adapted for the measurement of radon equilibrium factor. Passive measurements enable the accumulation of the result over a long period and cover a wide area. The probability of cancer induction and then the expected mortality was calculated based on different approaches. The results show that the overall weighted mean of annual effective dose for Al-Kharj resident is equal to 1.51 ± 0.8 mSv and The average expected mortality for residents in dwellings of Al-Kharj city is ranged from 0.596 ± 0.25 to 0.369 ± 0.15 death per 10,000 persons of ages from 40 to 70 years respectively. Also, the lifetime excess absolute risk (LEAR) of the residents of the Al-Kharj city is equal to (2.06 ± 0.8) × 10−4. The effect of dwelling types, ventilation and construction materials on the expected mortality is discussed
Preliminary Monte Carlo Study of 18F-FDG SPECT Imaging with LaBr3:Ce Crystal-based Gamma Cameras
No full textThe utility of 18F-deoxyglucose (18F-FDG) in oncology, cardiology, and neurology has generated great interest in a more economical ways of imaging 18F-FDG than conventional PET scanners. The main thrust of this work is to investigate the potential use of LaBr3:Ce materials in a low-cost FDG-SPECT system compared to NaI(Tl) using GATE Monte Carlo simulation. System performance at 140 keV and 511 keV was assessed using energy spectra, system sensitivity and count rate performance. Comparison of the LaBr3:Ce and NaI(Tl) crystal-based systems showed 4.5% and 8.9% higher system sensitivity for the LaBr3:Ce at 140 keV and 511 keV, respectively. The LaBr3:Ce scintillator significantly improves intrinsic count rate performance due to its fast decay time with respect to NaI(Tl). In conclusion, because LaBr3:Ce crystal combines excellent intrinsic count rate performance with slightly increased system sensitivity, it has the potential to be used for 18F-FDG -SPECT systems
Characterization of LaBr(3): Ce and LaCl(3): Ce scintillators for gamma-ray spectroscopy
No full textMeasurement of coincidence timing resolution of scintillation detectors compared to semiconductor detectors to image three-photon positron annihilation
No full textCharacterization of LaBr(3): Ce and LaCl(3): Ce scintillators for gamma-ray spectroscopy
No full textMeasurement of coincidence timing resolution of scintillation detectors compared to semiconductor detectors to image three-photon positron annihilation
No full textInvestigation of LaBr<inf>3</inf>:Ce and LaCl<inf>3</inf>:Ce scintillators for spect imaging
No full text
