Purpose: Evaluation of count rate performance (CRP) is an integral component of gamma camera quality assurance and system deadtime (τ ) may be utilized for image correction in quantitative studies. This work characterizes the CRP of three modern gamma cameras and estimates τ using two established methods (decay and dual source) under a variety of experimental conditions. Methods: For the decay method, uncollimated detectors were exposed to a Tc-99m source of relatively high activity and count rates were sampled regularly over 48 h. Input count rate at each time point was based on the lowest observed count rate data point. The input count rate was plotted against the observed count rate and fit via least-squares to the paralyzable detector model (PDM) to estimate τ (rates method). A novel expression for observed counts as a function of measurement time interval was derived, taking into account the PDM and the presence of background but making no assumption regarding input count rate. The observed counts were fit via least-squares to this novel expression to estimate τ (counts method). Correlation and Bland-Altman analyses were performed to assess agreement in estimates of τ between the rates and counts methods. The dependence of τ on energy window definition and incident energy spectrum were characterized. The dual source method was also used to estimate τ and its agreement with estimates from the decay method under identical conditions was also investigated. The dependences of τ on the total activity and the ratio of source activities were characterized. Results: The observed CRP curves for each gamma camera agreed with the PDM at low count rates but deviated substantially from it at high count rates. The estimates of τ determined from the paralyzable portion of the CPR curve using the rates method and the counts method were found to be highly correlated (r = 0.999) but with a small (∼6%) difference. No statistically significant difference was observed between the estimates of τ using the decay or dual source methods under identical experimental conditions (p = 0.13). Estimates of τ increased as a power-law function with decreasing ratio of counts in the photopeak to the total counts. Also, estimates of τ increased linearly as spectral effective energy decreased. No significant difference was observed between the dependences of τ on energy window definition or incident spectrum between the decay and dual source methods. Estimates of τ using the dual source method varied as a quadratic on the ratio of the single source to combined source activities and linearly with total activity. Conclusions: The CRP curves for three modern gamma camera models have been characterized, demonstrating unexpected behavior that necessitates the determination of both τ and maximum count rate to fully characterize the CRP curve. τ was estimated under a variety of experimental conditions, based on which guidelines for the performance of CRP testing in a clinical setting have been proposed.
Background Gastroenteropancreatic neuroendocrine tumors most commonly metastasize to the liver; however, high normal background 68Ga-DOTATATE activity and high image noise make metastatic lesions difficult to detect. The purpose of this study is to develop a rapid, automated and highly specific method to identify 68Ga-DOTATATE PET/CT hepatic lesions using a 2D U-Net convolutional neural network. Methods A retrospective study of 68Ga-DOTATATE PET/CT patient studies (n = 125; 57 with 68Ga-DOTATATE hepatic lesions and 68 without) was evaluated. The dataset was randomly divided into 75 studies for the training set (36 abnormal, 39 normal), 25 for the validation set (11 abnormal, 14 normal) and 25 for the testing set (11 abnormal, 14 normal). Hepatic lesions were physician annotated using a modified PERCIST threshold, and boundary definition by gradient edge detection. The 2D U-Net was trained independently five times for 100,000 iterations using a linear combination of binary cross-entropy and dice losses with a stochastic gradient descent algorithm. Performance metrics included: positive predictive value (PPV), sensitivity, F1 score and area under the precision–recall curve (PR-AUC). Five different pixel area thresholds were used to filter noisy predictions. Results A total of 233 lesions were annotated with each abnormal study containing a mean of 4 ± 2.75 lesions. A pixel filter of 20 produced the highest mean PPV 0.94 ± 0.01. A pixel filter of 5 produced the highest mean sensitivity 0.74 ± 0.02. The highest mean F1 score 0.79 ± 0.01 was produced with a 20 pixel filter. The highest mean PR-AUC 0.73 ± 0.03 was produced with a 15 pixel filter. Conclusion Deep neural networks can automatically detect hepatic lesions in 68Ga-DOTATATE PET. Ongoing improvements in data annotation methods, increasing sample sizes and training methods are anticipated to further improve detection performance.
Proper patient positioning plays a large role in the function of TCM, and hence CTDI and SSDE. In addition, body mass distribution may affect how patients ought to be positioned within the scanner. Understanding these effects is critical in optimizing CT scanning practices.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.