Ultra-fast [¹⁸F]florbetapir PET imaging using the uMI Panorama PET/CT system

Abstract

Background

There is a need for faster amyloid PET scans to reduce patients’ discomfort, minimize movement artifacts, and increase throughput. The recently introduced uMI Panorama PET/CT system featuring enhanced spatial resolution and sub-200ps TOF offers the potential for shorter scan duration without sacrificing image quality or efficacy to detect Aβ deposition. The study aims to establish a faster acquisition protocol for [¹⁸F]florbetapir PET imaging using digital PET/CT scanner uMI Panorama, while ensuring adequate image quality and amyloid-β (Aβ) detectability comparable to the standard 10-minute scan.

Methods

Thirty-eight participants (29 Aβ positive and 9 Aβ negative) from a prospective dementia cohort at Peking Union Medical University Hospital underwent routine [¹⁸F]florbetapir PET scans using the uMI Panorama PET/CT scanner and a T1-weighted brain MRI scan. List-mode PET data were reconstructed into durations of 10 min, 2 min, 1 min, 45 s, and 30 s (G10min, G2min, G1min, G45s, G30s). Two trained nuclear medicine physicians independently evaluated the image quality using a 5-point scale and provided binary diagnosis. Standardized uptake value ratios (SUVr) of the composite cortex (frontal, lateral parietal, lateral temporal, and cingulate cortices) were calculated to discriminate Aβ status and coefficient of variation assessed objective image quality. Comparisons of image quality and Aβ detectability between various fast scan groups and G10min group were conducted.

Results

The subjective image quality evaluation and Aβ detectability results from the two physicians showed both good intra-reader and inter-reader agreements (Cohen’s kappa coefficient: 0.759-1.000). The subjective and objective image qualities of the G2min scans were comparable to the G10min scans, whereas adequate image quality was achieved with the G1min and G45s scans (5-point score ≥ 3). Subjective visual diagnosis by two physicians yielded consistent accuracy for G10min, G2min, and G1min groups, but lower specificity for G45s and G30s groups. The objective detection of Aβ status by cortex SUVr across all scan durations maintained perfect discriminatory efficiency and relatively high effect size (Hedge’s G: 2.48–2.54).

Conclusions

A 1-min ultra-fast scan is feasible for [¹⁸F]florbetapir PET imaging using uMI Panorama PET/CT, while maintaining adequate image quality and Aβ diagnostic efficiency.

Clinical trial registration

NCT05023564. Registered September 2022 https://clinicaltrials.gov/search?term=NCT05023564.

Background

β-Amyloid (Aβ) positron emission tomography (PET) enables in-vivo assessment of the early pathogenesis of Aβ plaques in Alzheimer’s Disease (AD), and has advanced its clinical diagnosis, research, and drug development [1–3]. Three Fluorine-18 Aβ PET radiotracers—[¹⁸F]florbetapir, [¹⁸F]flutemetamol, and [¹⁸F]florbetaben— have been approved by the Food and Drug Administration and the European Medicines Agency for clinical use in Aβ PET imaging. The three tracers showed sensitivity of 88–98% and specificity of 80–95% in distinguishing older adults with moderate to frequent Aβ plaques from those without, as confirmed in autopsy studies [1, 4–6]. Currently, visual assessment of Aβ PET images by trained physicians is the standard diagnostic method in clinical practice, classifying Aβ status as either positive or negative. Notably, visual assessment can be influenced by the readers’ experience, and some cases may be equivocal when amyloid deposition is focal or emerging. Incorporating quantitative measures of Aβ deposition, such as standardized uptake value ratios (SUVr), provides benefits for clinical diagnosis and patient management [7–9]. Quantitative amyloid measurements are also critical for subject enrollment for therapeutic intervention and response evaluation in clinical trials [10–13].

Typically, acquisition durations for Aβ PET scans range from 10 to 20 min [14, 15]. Such lengthy durations can be uncomfortable for patients with cognitive impairment or the elderly, lead to head movement artifacts, and slow down clinical workflows [16]. As the use of Aβ PET for dementia patients increases, there is a growing demand to shorten scan times. Furthermore, the exploration of fast scans allows for the indirect assessment of low-dose injections, which is beneficial for decreasing PET examination costs and radiation burden, and makes PET imaging more available for vulnerable population. However, shorter acquisition times result in fewer detected coincidence events and effect PET image quality, quantitative accuracy and diagnostic value. The shortest scan duration possible for Aβ PET while maintaining diagnostic efficiency depends largely on the PET system. Previous studies have shown that the acquisition duration for clinical routine [¹⁸F]florbetapir and [¹⁸F]florbetaben PET scans could be reduced to a minimum of 5 min with acceptable image quality and diagnostic accuracy [17–19]. Additionally, the implementation of deep leaning techniques has further reduced the scan time to a minimum of 2 min [20, 21]. However, to our knowledge, no shorter duration for Aβ PET imaging has been reported.

The newly introduced uMI Panorama is an innovative PET/CT system that utilizes multiplexed silicon photomultiplier with improved photon detection efficiency and application-specific integrated circuit technologies. It offers favorable spatial resolution and sub-200-ps time-of-flight (TOF) timing resolution [22]. These advancements provide potential for an ultra-fast Aβ PET acquisition protocol to maximize the device’s capabilities, enhance scan efficiency, and improve patient comfort. In this study, we aim to establish a faster acquisition protocol for [¹⁸F]florbetapir PET imaging with uMI Panorama, while ensuring adequate image quality and Aβ detectability comparable to the standard 10-minute scan. This study was presented in accordance with the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines [23].

Materials and methods

Participants

In this cross-sectional study, 38 participants were enrolled from an ongoing prospective dementia cohort at Peking Union Medical University Hospital (PUMCH). The PUMCH Dementia Cohort is a hospital-based, observational study of subjects with cognitive impairment and cognitive normal healthy volunteers. The aims and eligibility criteria of the original study are described on ClinicalTrials. gov (NCT05023564). All participants underwent detailed clinical examination, neuropsychological battery tests, blood biochemical tests, CSF testing, APOE genotyping, and neuroimaging assessments [24].

The study included 29 patients in the Alzheimer’s continuum according to the National Institute on Aging and Alzheimer’s Association (NIA-AA) 2018 framework with clinical diagnosis of AD or mild cognitive impairment (MCI) due to AD [3]; 5 patients with frontotemporal dementia according to Diagnostic and Statistical Manual of Mental Disorders, 5th Edition [25]; 1 patient with progressive supranuclear palsy based on the 2017 Hoglinger criteria [26]; 1 patient with Creutzfeldt-jakob disease; 1 patient with dementia of undetermined cause; and 1 healthy volunteer. All participants underwent [¹⁸F]florbetapir PET scan on the uMI Panorama PET/CT scanner between January 2023 and December 2023, along with a T1-weighted brain MRI scan within three days. The study received approval from the ethics committee of the PUMCH and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all subjects or caregivers.

Image acquisition

Each participant underwent a routine brain PET scan 50 min after receiving an intravenous injection of [¹⁸F]florbetapir (median [inter-quantile range]: 308.95 [284.90, 341.33 MBq]), using the digital uMI Panorama PET/CT scanner (United Imaging Healthcare). Prior to PET acquisition, a low-dose CT scan was performed for attenuation correction. Corrections for randoms, attenuation, scatter, normalization, decay, and dead time were applied during the reconstruction. The list-mode PET data were reconstructed into varying durations: 10 min, 2 min, 1 min, 45 s, and 30 s (G10min, G2min, G1min, G45s, G30s). All PET images were reconstructed using a 3D-ordered subset expectation maximization algorithm with point spread function and TOF modeling. The reconstruction parameters were set at 4 iterations, 10 subsets, a 192 × 192 matrix, a 300 mm field of view (FOV), a 1.56 mm slice thickness, and a Gaussian post-filter with 5 mm of full-width half maximum. Additionally, high-resolution brain MR T1 images with a resolution of 1 × 1 × 1 mm³ were acquired using a fast spoiled gradient-echo sequence on a 3T MRI scanner (uPMR790, United Imaging Healthcare) for all subjects.

Subjective visual assessment

Two trained nuclear medicine physicians, blinded to all clinical and diagnostic information, independently visually assessed all PET images presented in randomized order. Each [¹⁸F]florbetapir PET image was given a binary diagnosis of Aβ status (Aβ positive or Aβ negative) based on the manufacturer’s reading guideline [27]. The G10min group’s diagnostic results served as the ground truth for each participant. In cases of disagreements in the visual findings of G10min group between the two physicians, a senior nuclear medicine physician was consulted to resolve the discrepancy. The accuracy, sensitivity, specificity of the visual diagnosis by the two physicians were then separately calculated across all scan duration groups. For subjective image quality evaluation, each image was rated on a 5-point scale were given (1 = uninterpretable, 2 = poor, 3 = adequate, 4 = good, 5 = excellent) in 3 perspectives: overall quality, noise, and diagnostic confidence. Additionally, 10% percent of the PET images were randomly selected and re-read one month after the initial reading by the two physicians to assess intra-reader agreement.

Image post-processing and objective quantitative analysis

First, all brain T1-MRI images were processed using FreeSurfer v7.4.1 software to conduct cortical parcellations based on the Desikan-Killiany Atlas in native space [28]. Following this, the [¹⁸F]florbetapir PET images were co-registered with the corresponding individual T1-MRIs. Region of interest (ROI) masks for four grouped cortical regions (frontal, cingulate, lateral parietal, and lateral temporal) and a composite cortex comprising these regions, along with a reference region mask of the whole cerebellum, were created and applied to the PET images [29]. The SUVr for the composite cortex and the four regional cortical areas was calculated as the mean radioactivity concentration of each ROI normalized to that of the whole cerebellum.

The SUVr value of the composite cortex was used to quantitatively discriminate the binary Aβ status. The best cut-off value was determined by the maximal Yueden’s index via receiver operating characteristic (ROC) analyses for each scan duration group. The corresponding area under the ROC curve (AUC), accuracy, sensitivity, and specificity were calculated to evaluate the quantitative discrimination efficiency. Hedge’s G effect size was reported for the difference between the two groups. Objective PET image quality for each image was quantified using the coefficient of variation (CoV = standard deviation/mean) of the composite cortex.

Statistical analysis

The normality of all quantitative variables was assessed using histograms and the Shapiro-Wilk test. Quantitative variables with a normal distribution were described using mean ± standard deviation and compared between the Aβ-positive and Aβ-negative groups using Student’s t-tests. For quantitative variables without a normal distribution and ordinal categorical variable, data were described using the median and inter-quantile range and compared using the Mann-Whitney U-tests. Nominal variables were summarized by counts and percentages and compared using the Fisher’s exact test. To assess both inter-reader and intra-reader agreement of the subjective visual assessments, the Cohen’s kappa (κ) coefficient and 95% confidence interval (95%CI) were calculated. The image quality evaluation scores on the 5-point scale and coefficient of variation (CoV) between G10min group and other fast scan groups were compared using the Wilcoxon signed-rank test, while the SUVr between them were compared using the paired t-test. Multiple comparison corrections were performed using the Bonferroni method. Bland–Altman analyses were used to assess the potential bias between the composite cortical SUVr of the G10min group and other fast scan groups, with a linear regression line fitted to assess the relationship between the bias and the SUVr value. Statistical significance was defined as P < 0.05. All statistical analysis and the creation of statistical graphs were conducted using SPSS 26.0 (IBM Corporation, Armonk, NY, USA), R 4.1.0 (R Foundation for Statistical Computing, Vienna, Austria) and GraphPad Prism version 9 (Graph-Pad Software, La Jolla, CA, USA).

Results

Participant characteristics

For the 38 participants included, 29 were visually diagnosed as Aβ positive based on the G10min image and all of them has a clinical diagnostic of AD or MCI due to AD. Table 1 displayed the demographics, cognitive scores of MMSE, and levels of radiotracer injection for participants of Aβ positive and Aβ negative groups, with no statistical differences in these characteristics between the two groups.

Table 1.

Characteristics of participants in Aβ positive and Aβ negative groups

	Aβ positive (N = 29)	Aβ negative (N = 9)	Statistics
Age (years)	63.62 ± 7.88	62.44 ± 13.28	t = 0.33, P = 0.744
Sex (Female) (N(%))	17 (58.6)	7 (77.8)	Fisher’s exact P = 0.438
Weight (kg)	58 (54, 65)	62 (54, 68)	U = 127.00, P = 0.919
BMI (kg/m2)	22.74 ± 3.99	22.76 ± 3.69	T = -0.02, P = 0.987
MMSE	15.38 ± 7.57	15.00 ± 11.19	T = 0.10, P = 0.926
Injected Dose (MBq)	303.40 (279.35, 342.25)	318.20 (308.95, 358.90)	U = 80.00, P = 0.086
Injected Dose/Weight (MBq/kg)	5.22 ± 1.01	5.64 ± 0.91	T = -1.11, P = 0.273

BMI: Body Mass Index; MMSE: Mini-Mental State Examination score

Subjective and Objective Image Quality Assessment

For the subjective image quality assessment, utilizing a 5-point scale, two physicians demonstrated both good inter-reader agreement (κ(95%CI): 0.851(0.803,0.899)) and intra-reader agreement (κ(95%CI) for physician 1: 0.759(0.549,0.969); κ(95%CI) for physician 2: 0.783 (0.545,1.000)).

The subjective image quality scores and objective image quality evaluated by the CoV value across five different scan durations were presented in Table 2. Compared to the G10min group, the G2min group’s subjective image quality scores and objective CoV values showed no significant difference. The G1min, G45s, and G30s groups exhibited statistically significant decreases in image quality compared to the prior two groups (all P < 0.05). Despite this, images of G1min and G45s groups all received subjective quality scores of 3 or higher, indicating adequate diagnostic quality (Fig. 1). Meanwhile, 5.26% (2/38) of the G30s images were rated as poor quality, with a score of 2 by physician 1.

Table 2.

Subjective and objective image quality assessment across five scan duration groups

Participant	Image quantity	G10min	G2min	G1min	G45s	G30s
Total	5-point score − 1	5 (5,5)	5 (5,5)	5 (4,5)*	4 (4,5)*	4 (3,4)*
	5-point score − 2	5 (5,5)	5 (5,5)	5 (4,5)*	4 (4,5)*	4 (3,4)*
	CoV values	0.19 (0.18,0.21)	0.20 (0.18,0.22)	0.21 (0.19,0.23)*	0.22 (0.20,0.23)*	0.23 (0.21,0.25)*
Aβ positive	5-point score − 1	5 (5,5)	5 (5,5)	5 (4,5)*	4 (4,5)*	4 (3,4)*
	5-point score − 2	5 (5,5)	5 (5,5)	5 (4,5)*	4 (4,5)*	4 (4,4)*
	CoV values	0.18 (0.17,0.20)	0.19 (0.18,0.21)	0.20 (0.19,0.21)*	0.21 (0.20,0.22)*	0.22 (0.21,0.23)*
Aβ negative	5-point score − 1	5 (5,5)	5 (5,5)	4 (4,5)	4 (3,4)*	4 (3,4)*
	5-point score − 2	5 (5,5)	5 (5,5)	4 (4,5)	4 (4,4)*	3 (3,4)*
	CoV values	0.23 (0.20,0.26)	0.25 (0.22,0.27)	0.26 (0.23,0.28)*	0.27 (0.23,0.29)*	0.28 (0.25,0.30)*

The subjective image quality evaluation, assessed by a 5-point score from two physicians, and the objective image quality evaluation using the coefficient of variation (CoV), were compared between the G15min and other fast scan groups using the Wilcoxon signed-rank test. *: P < 0.05

Fig. 1.

The distribution of the subjective image quality scores by a five-point scale (1 = uninterpretable, 2 = poor, 3 = adequate, 4 = good, 5 = excellent) by Physician 1 (a) and Physician 2 (b)

Fig. 2 displayed representative images from one Aβ positive and one Aβ negative participant across five different scan durations to visually illustrate the impact of reducing scan duration on image quality.

Fig. 2.

Representative transverse [¹⁸F]florbetapir PET images of one Aβ positive and one Aβ negative participant across 5 scan durations. The SUVr values of the composite cortex for each image are also shown below each image

Aβ detectability by visual and quantitative discrimination

For the subjective visual diagnostic of Aβ status, two physicians demonstrated both good inter-reader agreement (κ(95%CI): 0.888(0.808, 0.969)) and intra-reader agreement (κ(95%CI) for physician 1: 0.875(0.638, 1.000); κ(95%CI) for physician 2: 1.000 (1.000,1.000)).

The subjective visual diagnostic efficiency of Aβ by the two physicians showed no difference between the G10min, G2min, and G1min groups, while there was a slight decrease in accuracy and specificity for both two physicians in the G45s and G30s groups (Table 3).

Table 3.

Visual diagnostic efficiency of Aβ status across five scan duration groups

	Physician 1			Physician 2
	Accuracy	Sensitivity	Specificity	Accuracy	Sensitivity	Specificity
G10min	100.00 (38/38)	100.00 (29/29)	100.00 (9/9)	97.37 (37/38)	100.00 (29/29)	88.89 (8/9)
G2min	100.00 (38/38)	100.00 (29/29)	100.00 (9/9)	97.37 (37/38)	100.00 (29/29)	88.89 (8/9)
G1min	100.00 (38/38)	100.00 (29/29)	100.00 (9/9)	97.37 (37/38)	100.00 (29/29)	88.89 (8/9)
G45s	97.37 (37/38)	100.00 (29/29)	88.89 (8/9)	92.11 (35/38)	100.00 (29/29)	66.67 (6/9)
G30s	97.37 (37/38)	100.00 (29/29)	88.89 (8/9)	92.11 (35/38)	100.00 (29/29)	66.67 (6/9)

Accuracy, Sensitivity, and Specificity are expressed as percentages (%), each followed by the corresponding sample number

For objective quantitative discrimination of Aβ status using SUVr of the composite cortex, statistically significant differences were observed between the Aβ positive and Aβ negative groups across all scan duration groups (Fig. 3). As shown in Table 4, the effect size decreased slightly starting with 2-minute scan. However, quantitative discrimination between Aβ positive and Aβ negative participants using composite cortex SUVr was achieved perfectly across all scan duration groups with similar cut-off value, all with an AUC of 1.00.

Fig. 3.

Box and scatter plots of the SUVr of the composite cortex for Aβ positive (red) and Aβ negative (blue) groups across five scan duration groups. The mean SUVr value between the Aβ positive and negative groups was compared using a two-sample t test. *: P < 0.05

Table 4.

The objective discrimination efficiency of Aβ status across five scan duration groups

	Effect size (Hedge’s G)	AUC	Cut-off value	Accuracy	Sensitivity	Specificity
G10min	2.54	1.00	1.036	100.00 (38/38)	100.00 (29/29)	100.00 (9/9)
G2min	2.50	1.00	1.033	100.00 (38/38)	100.00 (29/29)	100.00 (9/9)
G1min	2.48	1.00	1.034	100.00 (38/38)	100.00 (29/29)	100.00 (9/9)
G45s	2.49	1.00	1.035	100.00 (38/38)	100.00 (29/29)	100.00 (9/9)
G30s	2.49	1.00	1.035	100.00 (38/38)	100.00 (29/29)	100.00 (9/9)

Accuracy, Sensitivity, and Specificity are expressed as percentages (%), each followed by the corresponding sample number. AUC: Area under the receiver operating characteristic curve

The SUVr value of the composite cortex in the four fast scan groups all exhibited good agreement with the value in the G10min group (Fig. 4). The mean composite cortex SUVr value of the G2min, G1min, G45s, G30s groups decreased with a maximum bias of 0.010 and 0.78% compared to the G10min group (Fig. 4; Table 5). However, this difference was significant in the Aβ positive group with a maximum bias of 0.013 and 0.94% decrease, but not in the Aβ negative group. The linear trend observed in Fig. 4 suggests increased difference with higher SUVr values.

Fig. 4.

Bland-Altman agreement analysis results of the SUVr value of the composite cortex between the G10min group and G2min (a), G1min (b), G45s (c), G30s (d) groups for Aβ positive (red) and Aβ negative (blue) groups. A linear regression line illustrating the relationship between the average SUVr of two compared groups and the absolute difference between them, and the regression coefficient (slope) and the statistical test results of the slope were showed. The absolute bias of the SUVr value of the composite cortex between the G10min group and the fast scan duration groups are presented, accompanied by a 95% confidence interval (95%CI) enclosed in parentheses

Table 5.

The SUVr values of the different cortical regions across five scan duration groups

Participant	Region	G10min	G2min	G1min	G45s	G30s
Total	composite cortex	1.277 ± 0.245	1.268 ± 0.239*	1.267 ± 0.240*	1.268 ± 0.241*	1.268 ± 0.240*
	frontal	1.268 ± 0.246	1.260 ± 0.242*	1.260 ± 0.243*	1.261 ± 0.244	1.260 ± 0.243
	lateral parental	1.295 ± 0.240	1.281 ± 0.236*	1.280 ± 0.237*	1.280 ± 0.237*	1.279 ± 0.235*
	lateral temporal	1.254 ± 0.241	1.244 ± 0.235*	1.244 ± 0.236*	1.245 ± 0.237*	1.245 ± 0.237*
	cingulate	1.281 ± 0.245	1.268 ± 0.237*	1.268 ± 0.236*	1.268 ± 0.239*	1.269 ± 0.239*
Aβ positive	composite cortex	1.378 ± 0.184	1.366 ± 0.181*	1.365 ± 0.183*	1.367 ± 0.183*	1.366 ± 0.183*
	frontal	1.370 ± 0.184	1.360 ± 0.182*	1.359 ± 0.184*	1.361 ± 0.184*	1.359 ± 0.185*
	lateral parental	1.395 ± 0.176	1.379 ± 0.176*	1.378 ± 0.177*	1.378 ± 0.177*	1.377 ± 0.175*
	lateral temporal	1.349 ± 0.190	1.338 ± 0.185*	1.337 ± 0.188*	1.339 ± 0.188*	1.339 ± 0.187*
	cingulate	1.384 ± 0.179	1.368 ± 0.174*	1.366 ± 0.173*	1.368 ± 0.174*	1.369 ± 0.174*
Aβ negative	composite cortex	0.953 ± 0.054	0.952 ± 0.052	0.952 ± 0.055	0.950 ± 0.056	0.951 ± 0.053
	frontal	0.939 ± 0.059	0.940 ± 0.058	0.941 ± 0.061	0.940 ± 0.061	0.942 ± 0.059
	lateral parental	0.971 ± 0.055	0.967 ± 0.051	0.965 ± 0.053	0.964 ± 0.054	0.964 ± 0.051
	lateral temporal	0.946 ± 0.046	0.944 ± 0.046	0.944 ± 0.049	0.943 ± 0.049	0.942 ± 0.047
	cingulate	0.950 ± 0.058	0.949 ± 0.059	0.949 ± 0.063	0.945 ± 0.065	0.947 ± 0.065

The SUVr values were compared between the G10min and other four fast scan groups using the paired t test. *: P < 0.05

Discussion

The study investigated an ultra-fast [¹⁸F]florbetapir PET imaging protocol using the uMI Panorama PET/CT scanner. We found that reducing the [¹⁸F]florbetapir PET scan duration to 1 min yielded equivalent Aβ visual diagnostic efficiency to the standard 10-minute image while maintaining adequate image quality. Furthermore, 30-second scan achieved equivalent Aβ discrimination efficiency with SUVr for quantitative evaluation. These findings proved the feasibility of ultra-fast scan of amyloid PET using a sub-200s TOF PET/CT scanner and underscore the robustness of quantitative measures in enhancing diagnostic accuracy.

The ultra-fast PET protocol owed largely to the technical enhancements of the uMI Panorama PET/CT device and favorable brain image quality. The uMI Panorama had spatial resolution of 2.88 mm, TOF resolution of 189ps at 5.3 kBq/mL, robust count rate performance, and high quantification accuracy, contributing to reduced noise levels and enhanced gray-white matter contrast in brain imaging [22]. This study validated the clinical benefits and found that a 2-minute [¹⁸F]florbetapir PET scan maintained the high image quality of a 10-minutes scan, while a 1-minute scan maintained equivalent visual diagnostic accuracy with adequate image quality. These findings set a new benchmark for ultra-fast scan times in amyloid PET imaging.

We observed a decrease in visual diagnostic accuracy in the 45-second and 30-second images, stemmed from misdiagnosing Aβ negative images as Aβ positive, resulting in lower specificity. This was attributed to the visually unclear gray-white matter differentiation caused by increased noise levels. Nonetheless, the quantitative measures of composite cortex SUVr in the 30-second images retained diagnostic performance equivalent to the 10-minute scans, demonstrating the robustness of quantitative measures with images of less favorable quality. Previous studies have also shown that the ability to quantitatively differentiate Aβ negative from Aβ positive using cortical SUVr of [¹⁸F]flutemetamol and [¹⁸F]florbetaben PET images remained stable across a wide range of doses, while visual assessment robustness decreased with lower doses [30]. The robustness of quantitative measures benefited diagnostic capabilities with amyloid PET imaging and potential role in clinical management. While the visual assessment remains the cornerstone of clinical evaluation of amyloid deposition, a minimal scan duration of 1-minute is recommended for now.

An optimal SUVr cut-off value of 1.034 for a 1-minute [¹⁸F]florbetapir PET scan with the uMI Panorama was derived from the study sample. Although the image processing procedure in this study aligns with the Alzheimer’s Disease Neuroimaging Initiative (ADNI) recommendation, a lower SUVr cutoff was found (1.034 vs. 1.11) [31]. Establishing a robust SUVr cut-off value remains an ongoing topic, as this threshold is susceptible to variations in scanners, acquisition protocols, image processing procedures, and populations [7]. The applicability of the current cut-off value is to be tested in larger multi-center cohorts.

The maximal bias and percent change of the cortical SUVr of the fast scan group compared to the 10-minute group is 0.013 and 0.94% in Aβ positive group. This percent change is smaller than the reported annual rates of change (1-2%) in the ADNI population and variability in the test-retest study (2.4% ± 1.41% for AD subjects and 1.5% ± 0.84% for controls) [32, 33], indicating that fast scans using the uMI Panorama would not hinder the detection of pathological change or difference. In addition to physical characteristics of the given PET/CT system, the precise post-processing method used in the study, particularly the cortical segmentation by FreeSurfer software, contributes to more accurate and precise quantification [34], though it requires a significant amount of time (about 1.5 h for an image using a computer with a CPU of 2.1 GHz and RAM of 64GB). Both of them contributed to the robust quantification of SUVr for [¹⁸F]florbetapir PET in the gray zone, as well as the nearly consistent variability of SUVr across the five scan duration groups (Fig. 3).

The study has several limitations. First, this single site nature of the study and its relatively small sample size, particularly the limited number of equivocal cases, may restrict the generalizability of the findings. Large-scale, multi-site validation studies are necessary to confirm these results. Second, the post-processing procedure used to achieve precise quantification is time-consuming. Future research could validate more time-efficient and streamlined quantification procedures to facilitate broader clinical adoption of quantitative amyloid measures. Third, while this study focused on [¹⁸F]florbetapir PET due to its widespread use, the minimal acquisition times for other amyloid radiopharmaceuticals using the uMI Panorama should be investigated to extend the applicability of the results presented here. Fourth, this study concentrated on the clinical performance of the uMI Panorama PET/CT system for its promising physical features. PET systems with similar or more advanced physical characteristics are anticipated to support ultra-fast imaging, although the generalizability of these results need be validated for each specific scanner. Additionally, the implementation of advanced AI techniques, such as image quality enhancement and head motion correction algorithms, may further reduce scan times to under one minute, benefiting the diagnosis of less cooperative patients. These advancements warrant further exploration in future studies.

Conclusions

This study demonstrates that the scanning duration of [¹⁸F]florbetapir PET imaging using the uMI Panorama PET/CT system can be effectively reduced to one minute while still maintaining Aβ detectability and adequate image quality. Ultra-fast [¹⁸F]florbetapir PET scans favor workflow efficiency, patient comfort, and potentiates for ultra-low dose scans.

Author contributions

All authors contributed to the study conceptualization. Xueqian Yang: methodology, validation, formal analysis, investigation, writing – original draft; Meiqi Wu: methodology, validation, formal analysis, investigation, data curation, writing – original draft, funding acquisition; Menglin Liang: methodology, formal analysis, investigation, data curation; Haiqiong Zhang: methodology, investigation; Bo Li, Chenhui Mao and Liling Dong: methodology, validation, investigation, resources; Yuan Wang, Haiqun Xing, Chao Ren, Zhenghai Huang and Qingxiang Wen: methodology, investigation, resources; Qi Ge and Zhengqing Yu: software, validation, data curation, visualization; Feng Feng: methodology, validation, resources, supervision; Jing Gao: methodology, resources, writing – review & editing, supervision, project administration, funding acquisition; Li Huo: methodology, resources, writing – review & editing, supervision, project administration, funding acquisition. All authors read and approved the final manuscript.

Funding

This work was supported by National Key Research and Development Program of China (nos. 2020YFA0804500, 2020YFA0804501, 2016YFC0901500), National Natural Science Foundation of China (82402335, 82071967, 22022601), CAMS innovation fund for medical science (CIFMS-2021-I2M-1-025, CIFMS-2021-I2M-1-002, CIFMS-2022-I2M-C&T-A-003), National High Level Hospital Clinical Research Funding (2022-PUMCH-B-070, 2022-PUMCH-B-071), and Shanghai Medical Innovation & Development Foundation.

Data availability

The data presented in this study can be accessed upon reasonable request. A formal project outline and ethics committee approval are required as part of the request process.

Declarations

Research involving human participants

The study received approval from the ethics committee of the PUMCH and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all subjects or caregivers.

Conflicts of interest

The authors have no relevant financial or non-financial interests to disclose.