Aiming for [18F]FDG-PET acquisition time reduction in clinical practice for neurological patients.

PURPOSE: Positron emission tomography (PET) imaging with [18F]FDG provides valuable information regarding the underlying pathological processes in neurodegenerative disorders. PET imaging for these populations should be as short as possible to limit head movements and improve comfort. This study aimed to validate an optimized [18F]FDG-PET image reconstruction protocol aiming to reduce acquisition time while maintaining adequate quantification accuracy and image quality. METHODS: A time-reduced reconstruction protocol (5 min) was evaluated in [18F]FDG-PET retrospective data from healthy individuals and Alzheimer's disease (AD) patients. Standard (8 min) and time-reduced protocols were compared by means of image quality and quantification accuracy metrics, as well as standardized uptake value ratio (SUVR) and Z-scores (pons was used as reference). Images were randomly and blindly presented to experienced physicians and scored in terms of image quality. RESULTS: No differences between protocols were identified during the visual assessment. Small differences (p < 0.01) in the pons SUVR were observed between the standard and time-reduced protocols for healthy individuals (-0.002 ±â€¯0.011) and AD patients (-0.007 ±â€¯0.013). Likewise, incorporating the PSF correction in the reconstruction algorithm resulted in small differences (p < 0.01) in SUVR between protocols (healthy individuals: -0.003 ±â€¯0.011; AD patients: -0.007 ±â€¯0.014). CONCLUSION: Quality metrics were similar between time-reduced and standard protocols. In the visual assessment of the images, the physicians did not consider the use of PSF adequate, as it degraded the quality image. Shortening the acquisition time is possible by optimizing the image reconstruction parameters while maintaining adequate quantification accuracy and image quality.

A new model for classification of medical CT images using CNN: a COVID-19 case study.

SARS-CoV-2 is the causative agent of COVID-19 and leaves characteristic impressions on chest Computed Tomography (CT) images in infected patients and this analysis is performed by radiologists through visual reading of lung images, and failures may occur. In this article, we propose a classification model, called Wavelet Convolutional Neural Network (WCNN) that aims to improve the differentiation of images of patients with COVID-19 from images of patients with other lung infections. The WCNN model was based on a Convolutional Neural Network (CNN) and wavelet transform. The model proposes a new input layer added to the neural network, which was called Wave layer. The hyperparameters values were defined by ablation tests. WCNN was applied to chest CT images to images from two internal and one external repositories. For all repositories, the average results of Accuracy (ACC), Sensitivity (Sen) and Specificity (Sp) were calculated. Subsequently, the average results of the repositories were consolidated, and the final values were ACC = 0.9819, Sen = 0.9783 and Sp = 0.98. The WCNN model uses a new Wave input layer, which standardizes the network input, without using data augmentation, resizing and segmentation techniques, maintaining the integrity of the tomographic image analysis. Thus, applications developed based on WCNN have the potential to assist radiologists with a second opinion in the analysis.1.

Women in Medical Physics and Biomedical Engineering: past, present and future.

Women in Medical Physics and Biomedical Engineering (WiMPBME) is a Task Group established in 2014 under the International Union of Physical and Engineering Scientists in Medicine (IUPESM). The group's main role is to identify, develop, implement, and coordinate various tasks and projects related to women's needs and roles in medical physics and biomedical engineering around the world. The current paper summarizes the past, present and future goals and activities undertaken or planned by the Task group in order to motivate, nurture and support women in medical physics and biomedical engineering throughout their professional careers. In addition, the article includes the historical pathway followed by various women's groups and subcommittees from 2004 up to the present day and depicts future aims to further these professions in a gender-balanced manner.

Parametric Imaging of [11C]Flumazenil Binding in the Rat Brain.

PURPOSE: This study evaluates the performance of several parametric methods for assessing [11C]flumazenil binding distribution in the rat brain. PROCEDURES: Dynamic (60 min) positron emission tomography data with metabolite-corrected plasma input function were retrospectively analyzed (male Wistar rats, n = 10). Distribution volume (V T) images were generated from basis function method (BFM), Logan graphical analysis (Logan), and spectral analysis (SA). Using the pons as pseudo-reference tissue, binding potential (BP ND and DVR-1) images were obtained from receptor parametric imaging algorithms (RPM and SRTM2) and reference Logan (RLogan). Standardized uptake value images (SUV and SUVR) were also computed for different intervals post-injection. Next, regional averages were extracted from the parametric images, using pre-defined volumes of interest, which were also applied to the regional time-activity curves from the dynamic data. Parametric data were compared to their regional counterparts and to two-tissue compartment model (2TCM)-based values (previously defined as the model of choice for rats). Parameter agreement was assessed by linear regression analysis and Bland-Altman plots. RESULTS: All parametric methods strongly correlated to their regional counterparts (R 2 > 0.97) and to the 2TCM values (R 2 ≥ 0.95). SA and RLogan underestimated V T and BP ND (slope of 0.93 and 0.86, respectively), while SUVR-1 overestimated BP ND (slope higher than 1.07 for all intervals). While BFM and SRTM2 had the smallest bias to 2TCM values (0.05 for both), ratio Bland-Altman plots showed Logan and RLogan displayed relative errors which were comparable between different regions, in contrast with the other methods. Although SUV consistently underestimated V T, the bias in this method was also constant across regions. CONCLUSIONS: All parametric methods performed well for the analysis of [11C]flumazenil distribution and binding in the rat brain. However, Logan and RLogan slightly outperformed the other methods in terms of precision, providing robust parameter estimation and constant bias. Yet, other methods can be of interest, because they can provide tissue perfusion (i.e., K 1 with BFM and SA), relative flow (i.e., R 1 with RPM and SRTM2), and model order (SA) images.