Search

Purpose: Quantitative images of metabolic activity can be derived through dynamic PET. However, the conventional approach necessitates invasive blood sampling to acquire the input function, thus limiting its noninvasive nature. The aim of this study was to devise a system based on convolutional neural network (CNN) capable of estimating the time-radioactivity curve of arterial plasma and accurately quantify the cerebral metabolic rate of glucose (CMRGlc) directly from PET data, thereby eliminating the requirement for invasive sampling., Methods: This retrospective investigation analyzed 29 patients with neurological disorders who underwent comprehensive whole-body 18 F-FDG-PET/CT examinations. Each patient received an intravenous infusion of 185 MBq of 18 F-FDG, followed by dynamic PET data acquisition and arterial blood sampling. A CNN architecture was developed to accurately estimate the time-radioactivity curve of arterial plasma., Results: The CNN estimated the time-radioactivity curve using the leave-one-out technique. In all cases, there was at least one frame with a prediction error within 10% in at least one frame. Furthermore, the correlation coefficient between CMRGlc obtained from the sampled blood and CNN yielded a highly significant value of 0.99., Conclusion: The time-radioactivity curve of arterial plasma and CMRGlc was determined from 18 F-FDG dynamic brain PET data using a CNN. The utilization of CNN has facilitated noninvasive measurements of input functions from dynamic PET data. This method can be applied to various forms of quantitative analysis of dynamic medical image data., (Copyright © 2023 The Author(s). Published by Wolters Kluwer Health, Inc.)

Introduction: Three-dimensional (3D) data acquisition is now standard on PET/computed tomography scanners. The aim of this study was to evaluate the repeatability of myocardial blood flow (MBF) estimation with rubidium-82 (Rb) 3D PET and to validate regional MBF measurements by comparison with two-dimensional (2D) PET., Patients and Methods: Fifteen healthy individuals (31.6 ± 11.4 years old) were enrolled for the evaluation of the short-term repeatability of rest 3D MBF quantification. Another 19 healthy individuals (35.3 ± 12.6 years old) underwent rest and pharmacological stress PET using 2D and 3D data acquisition within a 1-month interval. The injected dose was 1500 MBq for 2D and 555 MBq for 3D PET acquisition., Results: MBF at rest showed good repeatability [whole left ventricular MBF; 0.54 ± 0.13 vs. 0.52 ± 0.13 mL/min/g, P = 0.98]. Rest MBF, stress MBF, and myocardial flow reserve (MFR) were not significantly different between 3D and 2D data acquisition. 3D MBF correlated well with 2D MBF over a wide flow range for both whole left ventricular (r = 0.97, P < 0.0001) and regional values (r = 0.61, P < 0.0001)., Conclusion: MBF measured with 3D PET showed very good test-retest repeatability. Whole left ventricular and regional MBF measurements obtained using lower Rb-dose 3D PET were highly correlated over a wide range with those from 2D PET. Therefore, MBF with 3D PET can be applied using a lower Rb dosage in clinical settings with reduced radiation exposure.

Objective: Cerebral blood flow (CBF) estimation with C15O2 PET usually assumes a single tissue compartment model and a fixed brain-blood partition coefficient of water. However, the partition coefficient may change in pathological conditions. The purpose of this study was to investigate the changes in the partition coefficient of water in pathological regions and its effect on regional CBF assessment., Methods: The study protocol included 22 patients with occlusive cerebrovascular disease to compare the partition coefficients among three regions (infarction area, noninfarct hypoperfusion area, and contralateral area) in the pathological brain (analysis A), and to compare the CBF estimated by using a fixed partition coefficient and CBF estimated using floating partition coefficients (analysis B)., Results: The partition coefficient in the infarction area (0.55±0.07 ml/g) was lower than that in the contralateral normal cortex (0.68±0.05 ml/g), whereas noninfarct hypoperfusion area did not show a significant change (0.67±0.06 ml/g). As a result, the use of a fixed partition coefficient of normal volunteers (0.70 ml/g) resulted in an underestimation in regional CBF by 12% in infarction area (P<0.05), whereas the estimation errors were smaller and induced no significant difference in the noninfarct hypoperfusion area or in contralateral areas., Conclusion: The partition coefficient is stable except for the infarction area, and CBF estimation using a fixed partition coefficient of normal volunteers provides clinically appreciable information in patients with cerebrovascular disease.