In the present study, machine learning (ML) methods for the identification of abnormal glow curves (GC) of CaSO4:Dy-based thermoluminescence dosimeters in individual monitoring are presented. The classifier algorithms, random forest (RF), artificial neural network (ANN) and support vector machine (SVM) are employed for identifying not only the abnormal glow curve but also the type of abnormality. For the first time, the simplest and computationally efficient algorithm based on RF is presented for GC classifications. About 4000 GCs are used for the training and validation of ML algorithms. The performance of all algorithms is compared by using various parameters. Results show a fairly good accuracy of 99.05% for the classification of GCs by RF algorithm. Whereas 96.7% and 96.1% accuracy is achieved using ANN and SVM, respectively. The RF-based classifier is recommended for GC classification as well as in assisting the fault determination of the TLD reader system.

The objective of this paper is to study the effect of consecutive heating of TL elements of a thermoluminescence dosemeter (TLD) card in hot N2 gas-based TLD badge reader. The effect is studied by theoretical simulations of clamped heating profiles of the discs and resulting TL glow curves. The simulated temperature profile accounts for heat transfer to disc from hot gas as well as radiative and convective heat exchanges between the disc and the surrounding. The glow curves are simulated using 10 component glow peak model for CaSO4:Dy using the simulated temperature profile. The shape of the simulated glow curves and trend in total TL signal of the three discs were observed to match closely with the experimental observations when elevated surrounding temperature was considered for simulation. It is concluded that the readout (heating) of adjacent TLD disc affects the surrounding temperature leading to the changes in temperature profile of the next disc.

The study presents a novel approach to analyzing the thermoluminescence (TL) glow curves of CaSO4:Dy-based personnel monitoring dosimeters using machine learning. This study demonstrates the qualitative and quantitative impact of different types of anomalies on the TL signal and trains machine learning algorithms to estimate correction factors to account for these anomalies. The results show a good degree of agreement between the predicted and actual correction factors, with a coefficient of determination greater than 0.95, a root mean squared error less than 0.025, and a mean absolute error less than 0.015. The utilization of machine learning algorithms leads to a significant two-fold reduction in the coefficient of variation of TL counts from anomalous glow curves. This study proposes a promising approach to address anomalies caused by dosimeter, reader, and handling-related factors. Furthermore, it accounts for non-radiation-induced thermoluminescence at low dose levels towards improving the dosimetric accuracy in personnel monitoring.

The objective of this paper is to estimate the combined uncertainty in the measurement of dose equivalent at laboratory level using CaSO4:Dy-based thermoluminescent dosemeter badge system by including variations in the components of the system. The variability of performance of the system is analysed using random effects one way analysis of variance model. The model enables estimation of the overall variance of the performance of the sampled population. The population in the study comprises all possible indicated dose equivalents on irradiation of dosemeters to a specific dose equivalent and radiation quality. Coefficient of variation and combined uncertainty at 95% level of confidence in the measurement of Hp(10) due to S-Cs radiation quality are found to be 6.6 and 14.3%, respectively, at the dose level of 5.31 mSv. The above parameters in the measurement of in-use quantity, i.e. whole body dose or photon dose equivalent are found to be 7.4 and 16.4%, respectively. The performance of the monitoring system on relative response has been observed to be satisfactory. Various factors affecting the variability of performance of the system are identified for further improvement in coefficient of variation.

In this study, the Bayesian probabilistic approach is applied for the estimation of the actual dose using personnel monitoring dose records of occupational workers. To implement the Bayesian approach, the probability distribution of the uncertainty in the reported dose as a function of the actual dose is derived. Using the uncertainty distribution function of reported dose and prior knowledge of dose levels generally observed in a monitoring period, the posterior probability distribution of the actual dose is estimated. The posterior distributions of each monitoring period in a year are convoluted to arrive at actual annual dose distribution. The estimated actual doses distributions show a significant deviation from reported annual doses particularly for low annual doses.