Direct comparison of brain [18F]FDG images acquired by SiPM-based and PMT-based PET/CT: phantom and clinical studies

Abstract

Background Silicon photomultiplier-positron emission tomography (SiPM-PET) has better sensitivity, spatial resolution, and timing resolution than photomultiplier tube (PMT)-PET. The present study aimed to clarify the advantages of SiPM-PET in 18F-fluoro-2-deoxy-D-glucose ([18F]FDG) brain imaging in a head-to-head comparison with PMT-PET in phantom and clinical studies. Methods Contrast was calculated from images acquired from a Hoffman 3D brain phantom, and image noise and uniformity were calculated from images acquired from a pool phantom using SiPM- and PMT-PET. Sequential PMT-PET and SiPM-PET [18F]FDG images were acquired over a period of 10 min from 22 controls and 10 patients. All images were separately normalized to a standard [18F]FDG PET template, then the mean standardized uptake values (SUVmean) and Z-score were calculated using MIMneuro and CortexID Suite, respectively. Results Image contrast, image noise, and uniformity in SiPM-PET changed 19.2, 3.5, and − 40.0% from PMT-PET, respectively. These physical indices of both PET scanners satisfied the criteria for acceptable image quality published by the Japanese Society of Nuclear Medicine of contrast > 55%, CV ≤ 15%, and SD ≤ 0.0249, respectively. Contrast was 70.0% for SiPM-PET without TOF and 59.5% for PMT-PET without TOF. The TOF improved contrast by 3.5% in SiPM-PET. The SUVmean using SiPM-PET was significantly higher than PMT-PET and did not correlate with a time delay. Z-scores were also significantly higher in images acquired from SiPM-PET (except for the bilateral posterior cingulate) than PMT-PET because the peak signal that was extracted by the calculation of Z-score in CortexID Suite was increased. The hypometabolic area in statistical maps was reduced and localized using SiPM-PET. The trend was independent of whether the images were derived from controls or patients. Conclusions The improved spatial resolution and sensitivity of SiPM-PET contributed to better image contrast and uniformity in brain [18F]FDG images. The SiPM-PET offers better quality and more accurate quantitation of brain PET images. The SUVmean and Z-scores were higher in SiPM-PET than PMT-PET due to improved PVE. [18F]FDG images acquired using SiPM-PET will help to improve diagnostic outcomes based on statistical image analysis because SiPM-PET would localize the distribution of glucose metabolism on Z-score maps.

EANM procedure guidelines for brain PET imaging using [18F]FDG, version 3

“…When automated assessment of [ 18 F]FDG-PET images is performed, several aspects should be taken into consideration: Semiquantitative/voxel-based approaches to [ 18 F]FDG-PET analysis should always be used in conjunction with visual reading (considering visual reading as the first step for images evaluation and mandatory for quality control). Freeware and commercial software are available allowing for semiquantification or voxel-based analysis based on different methods [ 18 , 145 – 148 ]. Available freeware software for voxel-based analyses often are non-CE licensed.…”

Detailed Assessment of 18F-THK5351 Distribution Pattern in the Midbrain: Comparison With Progressive Supranuclear Palsy and Corticobasal Syndrome

“…The Discovery MI system is expected to have superior spatial resolution and sensitivity, compared with the Discovery PET/CT 710 system; however, upon visual inspection, the difference between the PET images obtained by each system is very small. 26 In addition, considering the FWHM values (less than 5 mm), the Discovery PET/CT 710 seems to have sufficient capacity for imaging the trimodal pattern and its collapse in the midbrain, as shown in Figure 2 (ie, both images came from the Discovery PET/CT 710). Meanwhile, we designated the AD group as a hallmark of nonparkinsonian disorders and found no significant difference in the MUR values between the AD and HC groups (Fig.…”

“…One of the limitations of the present study is that we analyzed the PET data together, which came from the Discovery PET/CT 710 and the Discovery MI. The Discovery MI system is expected to have superior spatial resolution and sensitivity, compared with the Discovery PET/CT 710 system; however, upon visual inspection, the difference between the PET images obtained by each system is very small 26 . In addition, considering the FWHM values (less than 5 mm), the Discovery PET/CT 710 seems to have sufficient capacity for imaging the trimodal pattern and its collapse in the midbrain, as shown in Figure 2 (ie, both images came from the Discovery PET/CT 710).…”

Differential Diagnosis of Alzheimer Disease vs. Mild Cognitive Impairment Based on Left Temporal Lateral Lobe Hypomethabolism on 18F-FDG PET/CT and Automated Classifiers

“…Finally, we wish to underline the usefulness of semi-quantitative analysis to assist visual reading, as it has been widely evidenced by international literature [ 15 , 37 ]. We used Cortex ID suite, a fully automated post-processing software for quantifying 18 F-FDG PET/CT and beta-amyloid brain scans that used three-dimensional stereotactic surface projections (3D-SSP) for statistical image analysis [ 45 , 46 ]. Future works may focus on the combination of semi-quantitative analysis with direct extraction of traditional and/or deep learning imaging features from the brain scans, as discussed in [ 47 , 48 , 49 ].…”