Background Machine learning studies require a large number of images often obtained on different PET scanners. When merging these images, the use of harmonized images following EARL-standards is essential. However, when including retrospective images, EARL accreditation might not have been in place. The aim of this study was to develop a convolutional neural network (CNN) that can identify retrospectively if an image is EARL compliant and if it is meeting older or newer EARL-standards. Materials and methods 96 PET images acquired on three PET/CT systems were included in the study. All images were reconstructed with the locally clinically preferred, EARL1, and EARL2 compliant reconstruction protocols. After image pre-processing, one CNN was trained to separate clinical and EARL compliant reconstructions. A second CNN was optimized to identify EARL1 and EARL2 compliant images. The accuracy of both CNNs was assessed using fivefold cross-validation. The CNNs were validated on 24 images acquired on a PET scanner not included in the training data. To assess the impact of image noise on the CNN decision, the 24 images were reconstructed with different scan durations. Results In the cross-validation, the first CNN classified all images correctly. When identifying EARL1 and EARL2 compliant images, the second CNN identified 100% EARL1 compliant and 85% EARL2 compliant images correctly. The accuracy in the independent dataset was comparable to the cross-validation accuracy. The scan duration had almost no impact on the results. Conclusion The two CNNs trained in this study can be used to retrospectively include images in a multi-center setting by, e.g., adding additional smoothing. This method is especially important for machine learning studies where the harmonization of images from different PET systems is essential.
Aim Recently, dose reference levels (DRLs) have been defined in Germany for auxiliary low-dose CT scans in hybrid SPECT/CT and PET/CT examinations, based on data from 2016/17. Here, another survey from 2020 was evaluated and compared with the new DRLs as well as with similar surveys from foreign countries. Methods The survey, which had already been conducted in the Nordic countries, queried for various examinations including the following values: patient weight and height, volume CT dose index (CTDIvol), dose length product (DLP). For each examination, statistical parameters such as the third quartile (Q3) were determined from all submitted CTDIvol and DLP values. Additionally, for examinations comprising datasets from at least 10 systems, the third quartile (Q3-Med) of the respective median values of each system was calculated. Q3 and Q3-Med were compared with the newly published DRLs from Germany and values from similar studies from other countries. Results Data from 15 SPECT/CT and 13 PET/CT systems from 15 nuclear medicine departments were collected. For the following examinations datasets from more than 10 systems were submitted: SPECT lung VQ, SPECT bone, SPECT&PET cardiac, PET brain, PET oncology. Especially for examinations of the thorax and heart, the new DRLs are very strict compared to this study. The CTDIvol values for examinations of the head were lower in this study than the DRLs prescribe now. Conclusions For certain examination types, there is a need for dose optimization at some clinics and devices in order to take into account the new DRLs in Germany in the future.
Background: Machine learning studies require a large number of images often obtained on different PET scanners. When merging these images, the use of harmonized images following EARL-standards is essential. However, when including retrospective images, EARL accreditation might not have been in place. The aim of this study was to develop a convolutional neural network (CNN) that can identify retrospectively if an image is EARL compliant and if it is meeting older or newer EARL-standards. Materials and Methods: 96 PET images acquired on three PET/CT systems were included in the study. All images were reconstructed with the locally clinically preferred, EARL1, and EARL2 compliant reconstruction protocols. After image pre-processing, one CNN was trained to separate clinical and EARL compliant reconstructions. A second CNN was optimized to identify EARL1 and EARL2 compliant images. The accuracy of both CNNs was assessed using 5-fold cross validation. The CNNs were validated on 24 images acquired on a PET scanner not included in the training data. To assess the impact of image noise on the CNN decision, the 24 images were reconstructed with different scan durations.Results: In the cross-validation, the first CNN classified all images correctly. When identifying EARL1 and EARL2 compliant images, the second CNN identified 100% EARL1 compliant and 85% EARL2 compliant images correctly. The accuracy in the independent dataset was comparable to the cross-validation accuracy. The scan duration had almost no impact on the results. Conclusion: The two CNNs trained in this study can be used to retrospectively include images in a multi-center setting by e.g. adding additional smoothing. This method is especially important for machine learning studies where the harmonization of images from different PET systems is essential.
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.