Artificial neural network modelling of uncertainty in gamma-ray spectrometry

Dragović, S.; Onjia, Antonije; Stanković, Srdjan; Aničin, I.; Bačić, Goran

doi:10.1016/j.nima.2004.11.045

Cited by 30 publications

(9 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Due to the possible disequilibrium between 238 U and 226 Ra, the activity of 238 U was evaluated from all the lines in the spectrum which originate from the isotopes in this series. Since in each of our samples the activity of 238 U, as deduced from all the lines, gave consistent values within the uncertainties of their intensities, which justifies the assumption of unperturbed equilibrium in the series, the final results were based on gamma-ray lines of 214 Bi at 609.3, 1,120.3, and 1,764.5 keV, which yield the highest statistical accuracy (Hamby and Tynybekov 2002;Al-Jundi 2002;Tzortzis and Tsertos 2004;Dragović et al 2005). The specific activity of 232 Th was evaluated from gamma-ray lines of 228 Ac at 338.4, 911.1, and 968.9 keV.…”

Section: Radioactivity Measurementsmentioning

confidence: 52%

Distribution of natural radionuclides in surface soils in the vicinity of abandoned uranium mines in Serbia

Momčilović

Kovačević²,

Tanic³

et al. 2012

Environ Monit Assess

View full text Add to dashboard Cite

The activity concentrations of natural radionuclides in soils from the area affected by uranium mining at Stara Planina Mountain in Serbia were studied and compared with the results obtained from an area with no mining activities (background area). In the affected area, the activity concentrations ranged from 1.75 to 19.2 mg kg(-1) for uranium and from 1.57 to 26.9 mg kg(-1) for thorium which is several-fold higher than those in the background area. The Th/U, K/U, and K/Th activity ratios were also determined and compared with data from similar studies worldwide. External gamma dose rate in the air due to uranium, thorium, and potassium at 1 m above ground level in the area affected by uranium mining was found to be 91.3 nGy h(-1), i.e., about two-fold higher than that in background area. The results of this preliminary study indicate the importance of radiological evaluation of the area and implementation of remedial measures in order to prevent further dispersion of radionuclides in the environment.

show abstract

Section: Radioactivity Measurementsmentioning

confidence: 52%

Distribution of natural radionuclides in surface soils in the vicinity of abandoned uranium mines in Serbia

Momčilović

Kovačević²,

Tanic³

et al. 2012

Environ Monit Assess

View full text Add to dashboard Cite

show abstract

“…72 And uncertainty was estimated by an artificial neural network model. 73 In addition, neural networks have also been applied in ground-based gamma-ray astronomy 74 and x-ray mapping methods. 75 In addition, neural networks have been widely applied to pattern recognition as shown in the next section.…”

Section: Neural Network and Their Application To Xrsmentioning

confidence: 99%

Chemometrics and its applications to x‐ray spectrometry

Luo

2006

X-Ray Spectrometry

View full text Add to dashboard Cite

Recent developments in curve fitting, multivariate calibration, and pattern recognition in chemometrics, and their application to x-ray spectrometry, are reviewed. Relatively innovated algorithms, namely genetic algorithms, neural networks and support vector machines, are discussed. Together with the three algorithms, the performances of different algorithms are compared briefly, which mainly includes principal component analysis, partial least-squares regression, factor analysis, cluster analysis, nearest neighbor methods, linear discriminant analysis, linear learning machine, and soft independent modeling of class analogy. In general, the chemometrics methods are superior to the conventional methods, such as Fourier transform and Marquardt-Levenberg algorithms, to a certain extent.

show abstract

“…Recently, artificial neural networks (ANNs) have been used in gamma-ray spectrometry for automatic identification of radionuclides from their spectra [2], prediction of uncertainty [3] and peak-to-background ratio [4], as well as for quantitative spectrometric analysis [5][6][7]. In our work, the ANN model for the prediction of the minimum detectable activity of radionuclides in soil samples as a function of measurement time was optimized using the simplex method.…”

Section: Introductionmentioning

confidence: 99%

“…In our previous work [3,4], the network was optimized by a ''trial and error'' method. This procedure for the optimization of ANN parameters can be very slowly convergent since the variables are usually not fully independent [8].…”

Section: Introductionmentioning

confidence: 99%