Relative cerebral blood flow (rCBF), assessed using pulsed arterial spin labeling (pASL) MRI, and the standardized uptake value ratio (SUVr) in early-phase amyloid-PET (ePET) are used as proxies for brain perfusion. These methods have the potential to streamline clinical workflows and reduce the burden on patients by eliminating the need for additional procedures. While both techniques have shown good agreement with the gold standard for glucose metabolism assessment, F-fluorodeoxyglucose-PET, a direct comparison between them has yet to be fully clarified. This retrospective study aimed to compare perfusion-like data from pASL (rCBF) and ePET (SUVr) in a memory clinic cohort. We included 46 subjects (69 ± 8 years; 37 women) from the Geneva Memory Center (cognitively impaired-CI n = 29; cognitively unimpaired-CU n = 17), with available pASL and ePET. We evaluated the association between rCBF and SUVr values across 18 cortical and subcortical regions using linear regression and the within-subject coefficient of variation (wsCV). Regional differences between CU and CI groups were assessed using linear regression model corrected for age. We observed significant association between rCBF and SUVr in precuneus (β = 0.69, wsCV = 16.9), angular gyrus (β = 0.64, wsCV = 19.4), and hippocampus (β = 0.23, wsCV = 16.1). Additionally, significant differences in rCBF between CU and CI were also observed in the posterior cingulate, precuneus, calcarine, hippocampus, and composite (p < 0.05), while SUVr showed significant differences only in the hippocampus. Our findings indicate weak to moderate local correlations between the two techniques. However, both exhibited differing regional perfusion levels in CU and CI groups, with rCBF showing more regional differences between cognitive stages in comparison with SUVr.

Purpose As dual-phase amyloid-PET can evaluate amyloid (A) and neurodegeneration (N) with a single tracer injection, dual-phase tau-PET might be able to provide both tau (T) and N. Our study aims to assess the association of early-phase tau-PET scans and 18 F-fluorodeoxyglucose (FDG) PET and their comparability in discriminating Alzheimer’s disease (AD) patients and differentiating neurodegenerative patterns. Methods 58 subjects evaluated at the Geneva Memory Center underwent dual-phase 18 F-Flortaucipir-PET with early-phase acquisition (eTAU) and 18 F-FDG-PET within 1 year. A subsample of 36 participants also underwent dual-phase amyloid-PET (eAMY). Standardized uptake value ratios (SUVRs) were calculated to assess the correlation of eTAU and their respective 18 F-FDG-PET and eAMY scans. Hypometabolism and hypoperfusion maps and their spatial overlap were also evaluated at the individual level visually and semiquantitatively. Receiver operating characteristic analyses were performed to compare the discriminative power of eTAU, FDG, and eAMY SUVR between A-/T- and A+/T + participants. Results Strong positive correlations were found between eTAU and FDG SUVRs ( r = 0.84, p < 0.001) and eTAU and eAMY SUVRs ( r > 0.87, p < 0.001). Clusters of significant hypoperfusion with good correspondence to hypometabolism topographies were found at the individual level, independently of the underlying neurodegenerative patterns. Both eTAU and FDG SUVRs significantly distinguished A+/T + from A-/T- individuals (AUC eTAU =0.604, AUC FDG =0.748) with FDG performing better than eTAU ( p = 0.04). eAMY and eTAU SUVR showed comparable discriminative power. Conclusion Early-phase 18 F-Flortaucipir-PET can provide perfusion information closely related to brain regional glucose metabolism and perfusion measured by early-phase amyloid-PET, even if less accurate than FDG-PET as a biomarker for neurodegeneration.

Substantial variability in tau accumulation patterns in Alzheimer’s disease (AD) population has now become accepted. Subtype and Stage Inference (SuStaIn) has distinguished four distinct spatiotemporal trajectories of tau pathology: limbic (S1), medial temporal lobe-sparing (S2), posterior (S3), and lateral temporal (S4). A visual method to validate and identify them is a requirement for their clinical translation. Our study aims to provide evidence for multiple tau accumulation patterns in a clinical setting by developing and testing a novel topographic visual method for tau-PET based on SuStaIn subtypes. We included 245 participants from the Geneva Memory Clinic, who underwent 18 F-Flortaucipir-PET. All scans were classified into different subtypes by visual ratings and the automated SuStain algorithm. Cohen's kappa (k) tested the agreement between raters and between visual and automated subtypes. Chi-squared and Kruskal-Wallis tests were used to test differences in clinical features, tau, and amyloid (Aβ) loads between subtypes. Differences in cognitive trajectories were tested using linear mixed-effects models, controlling for age, sex, and tau stages. A substantial agreement between raters was found in visually interpreting tau pattern subtypes (k>0.65, p<0.001) and a fair agreement between visual and automated subtypes (k=0.39, p<0.001), with the automated approach detecting a higher rate of negative scans. According to the visual classification, individuals with S2 subtype were younger than S1 and S3 (p<0.001), had worse MMSE and verbal fluency scores (p<0.05) than S4 and S1, showed higher global tau than other subtypes (p<0.05), and a steeper cognitive decline. Our results show that visual classification can reliably identify four tau patterns, differing in global tau loads, clinical characteristics, and long-term outcomes, suggesting its clinical usefulness for the detection of higher-risk AD variants. A clinically implementable classification in subtypes with faster decline is paramount for personalized diagnosis and therapy.

INTRODUCTION:Multiplexed PET imaging revolutionized clinical decision-making by simultaneously capturing various radiotracer data in a single scan, enhancing diagnostic accuracy and patient comfort. Through a transformer-based deep learning, this study underscores the potential of advanced imaging techniques to streamline diagnosis and improve patient outcomes. PATIENTS AND METHODS:The research cohort consisted of 120 patients spanning from cognitively unimpaired individuals to those with mild cognitive impairment, dementia, and other mental disorders. Patients underwent various imaging assessments, including 3D T1-weighted MRI, amyloid PET scans using either 18 F-florbetapir (FBP) or 18 F-flutemetamol (FMM), and 18 F-FDG PET. Summed images of FMM/FBP and FDG were used as proxy for simultaneous scanning of 2 different tracers. A SwinUNETR model, a convolution-free transformer architecture, was trained for image translation. The model was trained using mean square error loss function and 5-fold cross-validation. Visual evaluation involved assessing image similarity and amyloid status, comparing synthesized images with actual ones. Statistical analysis was conducted to determine the significance of differences. RESULTS:Visual inspection of synthesized images revealed remarkable similarity to reference images across various clinical statuses. The mean centiloid bias for dementia, mild cognitive impairment, and healthy control subjects and for FBP tracers is 15.70 ± 29.78, 0.35 ± 33.68, and 6.52 ± 25.19, respectively, whereas for FMM, it is -6.85 ± 25.02, 4.23 ± 23.78, and 5.71 ± 21.72, respectively. Clinical evaluation by 2 readers further confirmed the model's efficiency, with 97 FBP/FMM and 63 FDG synthesized images (from 120 subjects) found similar to ground truth diagnoses (rank 3), whereas 3 FBP/FMM and 15 FDG synthesized images were considered nonsimilar (rank 1). Promising sensitivity, specificity, and accuracy were achieved in amyloid status assessment based on synthesized images, with an average sensitivity of 95 ± 2.5, specificity of 72.5 ± 12.5, and accuracy of 87.5 ± 2.5. Error distribution analyses provided valuable insights into error levels across brain regions, with most falling between -0.1 and +0.2 SUV ratio. Correlation analyses demonstrated strong associations between actual and synthesized images, particularly for FMM images (FBP: Y = 0.72X + 20.95, R2 = 0.54; FMM: Y = 0.65X + 22.77, R2 = 0.77). CONCLUSIONS:This study demonstrated the potential of a novel convolution-free transformer architecture, SwinUNETR, for synthesizing realistic FDG and FBP/FMM images from summation scans mimicking simultaneous dual-tracer imaging.

BACKGROUND:This case report presents a patient with progressive memory loss and choreiform movements.CASE PRESENTATION:Neuropsychological tests indicated multi-domain amnestic mild cognitive impairment (aMCI), and neurological examination revealed asymmetrical involuntary hyperkinetic movements. Imaging studies showed severe left-sided atrophy and hypometabolism in the left frontal and temporoparietal cortex. [18F]Flortaucipir PET exhibited moderately increased tracer uptake in hypometabolic areas. The diagnosis initially considered Alzheimer's disease (AD), frontotemporal degeneration (FTD), and corticobasal degeneration (CBD), cerebral hemiatrophy syndrome, but imaging and cerebrospinal fluid analysis excluded AD and suggested fused-in-sarcoma-associated FTD (FTLD-FUS), a subtype of the behavioural variant of FTD.CONCLUSIONS:Our case highlights that despite the lack of specific FUS biomarkers the combination of clinical features and neuroimaging biomarkers can guide choosing the most likely differential diagnosis in a complex neurological case. Imaging in particular allowed an accurate measure of the topography and severity of neurodegeneration and the exclusion of AD-related pathology.

Introduction Multiplexed PET imaging revolutionized clinical decision-making by simultaneously capturing various radiotracer data in a single scan, enhancing diagnostic accuracy and patient comfort. Through a transformer-based deep learning, this study underscores the potential of advanced imaging techniques to streamline diagnosis and improve patient outcomes. Patients and Methods The research cohort consisted of 120 patients spanning from cognitively unimpaired individuals to those with mild cognitive impairment, dementia, and other mental disorders. Patients underwent various imaging assessments, including 3D T1-weighted MRI, amyloid PET scans using either 18F-florbetapir (FBP) or 18F-flutemetamol (FMM), and 18F-FDG PET. Summed images of FMM/FBP and FDG were used as proxy for simultaneous scanning of 2 different tracers. A SwinUNETR model, a convolution-free transformer architecture, was trained for image translation. The model was trained using mean square error loss function and 5-fold cross-validation. Visual evaluation involved assessing image similarity and amyloid status, comparing synthesized images with actual ones. Statistical analysis was conducted to determine the significance of differences. Results Visual inspection of synthesized images revealed remarkable similarity to reference images across various clinical statuses. The mean centiloid bias for dementia, mild cognitive impairment, and healthy control subjects and for FBP tracers is 15.70 ± 29.78, 0.35 ± 33.68, and 6.52 ± 25.19, respectively, whereas for FMM, it is −6.85 ± 25.02, 4.23 ± 23.78, and 5.71 ± 21.72, respectively. Clinical evaluation by 2 readers further confirmed the model's efficiency, with 97 FBP/FMM and 63 FDG synthesized images (from 120 subjects) found similar to ground truth diagnoses (rank 3), whereas 3 FBP/FMM and 15 FDG synthesized images were considered nonsimilar (rank 1). Promising sensitivity, specificity, and accuracy were achieved in amyloid status assessment based on synthesized images, with an average sensitivity of 95 ± 2.5, specificity of 72.5 ± 12.5, and accuracy of 87.5 ± 2.5. Error distribution analyses provided valuable insights into error levels across brain regions, with most falling between −0.1 and +0.2 SUV ratio. Correlation analyses demonstrated strong associations between actual and synthesized images, particularly for FMM images (FBP: Y = 0.72X + 20.95, R 2 = 0.54; FMM: Y = 0.65X + 22.77, R 2 = 0.77). Conclusions This study demonstrated the potential of a novel convolution-free transformer architecture, SwinUNETR, for synthesizing realistic FDG and FBP/FMM images from summation scans mimicking simultaneous dual-tracer imaging.

Amyloid-β (Aβ) plaques is a significant hallmark of Alzheimer's disease (AD), detectable via amyloid-PET imaging. The Fluorine-18-Fluorodeoxyglucose ([18F]FDG) PET scan tracks cerebral glucose metabolism, correlated with synaptic dysfunction and disease progression and is complementary for AD diagnosis. Dual-scan acquisitions of amyloid PET allows the possibility to use early-phase amyloid-PET as a biomarker for neurodegeneration, proven to have a good correlation to [18F]FDG PET. The aim of this study was to evaluate the added value of synthesizing the later from the former through deep learning (DL), aiming at reducing the number of PET scans, radiation dose, and discomfort to patients. A total of 166 subjects including cognitively unimpaired individuals (N = 72), subjects with mild cognitive impairment (N = 73) and dementia (N = 21) were included in this study. All underwent T1-weighted MRI, dual-phase amyloid PET scans using either Fluorine-18 Florbetapir ([18F]FBP) or Fluorine-18 Flutemetamol ([18F]FMM), and an [18F]FDG PET scan. Two transformer-based DL models called SwinUNETR were trained separately to synthesize the [18F]FDG from early phase [18F]FBP and [18F]FMM (eFBP/eFMM). A clinical similarity score (1: no similarity to 3: similar) was assessed to compare the imaging information obtained by synthesized [18F]FDG as well as eFBP/eFMM to actual [18F]FDG. Quantitative evaluations include region wise correlation and single-subject voxel-wise analyses in comparison with a reference [18F]FDG PET healthy control database. Dice coefficients were calculated to quantify the whole-brain spatial overlap between hypometabolic ([18F]FDG PET) and hypoperfused (eFBP/eFMM) binary maps at the single-subject level as well as between [18F]FDG PET and synthetic [18F]FDG PET hypometabolic binary maps. The clinical evaluation showed that, in comparison to eFBP/eFMM (average of clinical similarity score (CSS) = 1.53), the synthetic [18F]FDG images are quite similar to the actual [18F]FDG images (average of CSS = 2.7) in terms of preserving clinically relevant uptake patterns. The single-subject voxel-wise analyses showed that at the group level, the Dice scores improved by around 13

Purpose [ 18 F]Flortaucipir PET is a powerful diagnostic and prognostic tool for Alzheimer’s disease (AD). Tau status definition is mainly based in the literature on semi-quantitative measures while in clinical settings visual assessment is usually preferred. We compared visual assessment with established semi-quantitative measures to classify subjects and predict the risk of cognitive decline in a memory clinic population. Methods We included 245 individuals from the Geneva Memory Clinic who underwent [ 18 F]flortaucipir PET. Amyloid status was available for 207 individuals and clinical follow-up for 135. All scans were blindly evaluated by three independent raters who visually classified the scans according to Braak stages. Standardized uptake value ratio (SUVR) values were obtained from a global meta-ROI to define tau positivity, and the Simplified Temporo-Occipital Classification (STOC) was applied to obtain semi-quantitatively tau stages. The agreement between measures was tested using Cohen’s kappa ( k ). ROC analysis and linear mixed-effects models were applied to test the diagnostic and prognostic values of tau status and stages obtained with the visual and semi-quantitative approaches. Results We found good inter-rater reliability in the visual interpretation of tau Braak stages, independently from the rater’s expertise ( k >0.68, p <0.01). A good agreement was equally found between visual and SUVR-based classifications for tau status ( k =0.67, p <0.01). All tau-assessment modalities significantly discriminated amyloid-positive MCI and demented subjects from others (AUC>0.80) and amyloid-positive from negative subjects (AUC>0.85). Linear mixed-effect models showed that tau-positive individuals presented a significantly faster cognitive decline than the tau-negative group ( p <0.01), independently from the classification method. Conclusion Our results show that visual assessment is reliable for defining tau status and stages in a memory clinic population. The high inter-rater reliability, the substantial agreement, and the similar diagnostic and prognostic performance of visual rating and semi-quantitative methods demonstrate that [ 18 F]flortaucipir PET can be robustly assessed visually in clinical practice.

Purpose: to investigate the preoperative role of ML-based classification using conventional 18F-FDG PET parameters and clinical data in predicting features of EC aggressiveness. Methods: retrospective study, including 123 EC patients who underwent 18F-FDG PET (2009–2021) for preoperative staging. Maximum standardized uptake value (SUVmax), SUVmean, metabolic tumour volume (MTV), and total lesion glycolysis (TLG) were computed on the primary tumour. Age and BMI were collected. Histotype, myometrial invasion (MI), risk group, lymph-nodal involvement (LN), and p53 expression were retrieved from histology. The population was split into a train and a validation set (80–20%). The train set was used to select relevant parameters (Mann-Whitney U test; ROC analysis) and implement ML models, while the validation set was used to test prediction abilities. Results: on the validation set, the best accuracies obtained with individual parameters and ML were: 61% (TLG) and 87% (ML) for MI; 71% (SUVmax) and 79% (ML) for risk groups; 72% (TLG) and 83% (ML) for LN; 45% (SUVmax; SUVmean) and 73% (ML) for p53 expression. Conclusions: ML-based classification using conventional 18F-FDG PET parameters and clinical data demonstrated ability to characterize the investigated features of EC aggressiveness, providing a non-invasive way to support preoperative stratification of EC patients.

18 F-Flortaucipir-PET allows visualization of tau deposits (T) representing a powerful diagnostic and prognostic tool for Alzheimer’s disease (AD). Dual-phase 18 F-Flortaucipir-PET can also evaluate neurodegeneration (N) through the early-phase images. We aim to assess the diagnostic and prognostic power of visual and semi-quantitative T assessment and the potential added value of early-phase images. We included 245 subjects from the Geneva Memory Clinic who underwent 18 F-Flortaucipir-PET (157 with a dual-phase protocol). Amyloid (Aβ) status was available for 207 subjects and clinical follow-up for 135. T positivity and a Braak staging estimate were evaluated visually and semi-quantitatively by standardized uptake value ratio (SUVr) in AD-related regions. N status was semi-quantitatively defined using early-phase images as a summary metric in the AD meta-ROI region. Receiver Operating Characteristic (ROC) analyses were applied to compare T and N discriminative power. Linear mixed-effects models tested the prognostic values of T alone and T combined with N (T/N profiles). The mean age of the participants was 70.49±9.13 years, 52.2% were females and 55.7% were Aβ+. 53.4% of subjects had Mild Cognitive Impairment (MCI), 19.8% had dementia and 26.7% were cognitively unimpaired. All T-assessment modalities significantly discriminated Aβ+ MCI and demented subjects from others (AUC>0.80), Aβ+ from Aβ- subjects (AUC>0.85), and cognitive decliners from stable individuals at follow-up (AUC>0.79). The discriminative power of N, assessed by early-phase images, was lower compared to T, with AUC ranging from 0.63 in discriminating Aβ+ MCI and demented subjects from others to 0.71 for discriminating Aβ+ T+ from Aβ- T- subjects. Linear mixed-effect models showed that T+ individuals presented a significantly faster cognitive decline than the T- group (p<0.001) as well as T+/N+ and T+/N- compared to T-/N- group (p<0.001), independently from T classification methods. Our results show a similar diagnostic and prognostic power of visual and semi-quantitative T evaluation of 18 F-Flortaucipir-PET, proving its high potential for clinical applicability. The N status assessment with early-phase images did not improve the prediction of cognitive decline. Its usefulness for differential diagnosis by the topographical information provided should be investigated.

The present pilot study investigates the putative role of radiomics from [18F]FDG PET/CT scans to predict PD-L1 expression status in non-small cell lung cancer (NSCLC) patients. In a retrospective cohort of 265 patients with biopsy-proven NSCLC, 86 with available PD-L1 immunohistochemical (IHC) assessment and [18F]FDG PET/CT scans have been selected to find putative metabolic markers that predict PD-L1 status (< 1%, 1–49%, and ≥ 50% as per tumor proportion score, clone 22C3). Metabolic parameters have been extracted from three different PET/CT scanners (Discovery 600, Discovery IQ, and Discovery MI) and radiomics features were computed with IBSI compliant algorithms on the original image and on images filtered with LLL and HHH coif1 wavelet, obtaining 527 features per tumor. Univariate and multivariate analysis have been performed to compare PD-L1 expression status and selected radiomic features. Of the 86 analyzed cases, 46 (53%) were negative for PD-L1 IHC, 13 (15%) showed low PD-L1 expression (1–49%), and 27 (31%) were strong expressors (≥ 50%). Maximum standardized uptake value (SUVmax) demonstrated a significant ability to discriminate strong expressor cases at univariate analysis (p = 0.032), but failed to discriminate PD-L1 positive patients (PD-L1 ≥ 1%). Three radiomics features appeared the ablest to discriminate strong expressors: (1) a feature representing the average high frequency lesion content in a spherical VOI (p = 0.009); (2) a feature assessing the correlation between adjacent voxels on the high frequency lesion content (p = 0.004); (3) a feature that emphasizes the presence of small zones with similar grey levels inside the lesion (p = 0.003). The tri-variate linear discriminant model combining the three features achieved a sensitivity of 81% and a specificity of 82% in the test. The ability of radiomics to predict PD-L1 positive patients was instead scarce. Our data indicate a possible role of the [18F]FDG PET radiomics in predicting strong PD-L1 expression; these preliminary data need to be confirmed on larger or single-scanner series.

Purpose The aim of this study was to investigate the role of 18F-FDG PET/CT in predicting pathological prognostic factors, including tumor type and International Federation of Gynecology and Obstetrics (FIGO) score, in gestational trophoblastic disease (GTD). Methods Retrospective monocentric study including 24 consecutive patients who underwent to 18F-FDG PET/CT from May 2005 to March 2021 for GTD staging purpose. The following semiquantitative PET parameters were measured from the primary tumor and used for the analysis: maximum standardized uptake value (SUVmax), SUVmean, metabolic tumor volume (MTV) and total lesion glycolisis (TLG). Statistical analysis included Spearman correlation coefficient to evaluate the correlations between imaging parameters and tumor type (nonmolar trophoblastic vs postmolar trophoblastic tumors) and risk groups (high vs low, defined according to the FIGO score), whereas area under the curve (AUC) of the receiver operating characteristic (ROC) curve was used to assess the predictive value of the PET parameters. Mann-Whitney U test was used to further describe the parameter’s potential in differentiating the populations. Results SUVmax and SUVmean resulted fair (AUC, 0.783; 95% confidence interval [CI], 0.56–0.95) and good (AUC, 0.811; 95% CI, 0.59–0.97) predictors of tumor type, respectively, showing a low (ρ = 0.489, adjusted P = 0.030) and moderate (ρ = 0.538, adjusted P = 0.027) correlation. According to FIGO score, TLG was instead a fair predictor (AUC, 0.770; 95% CI, 0.50–0.99) for patient risk stratification. Conclusions 18F-FDG PET parameters have a role in predicting GTD pathological prognostic factors, with SUVmax and SUVmean being predictive for tumor type and TLG for risk stratification.

1521 Objectives: Cervical cancer is one of the most frequent forms of neoplasm in women and one of the leading causes of cancer death in this group. An appropriate preoperative staging, particularly focused on the eventual lymph nodes involvement, is crucial to ensure the most appropriate treatment and to obtain the best outcome. In locally advanced disease, 18F-FDG PET/CT already demonstrated a prognostic role. In the present study is evaluated its putative role in the preoperative setting in the prediction of lymph node metastasis through the analysis of radiomic features of the primary cervical lesion. Methods: A subset of 74 patients with early stage cervical cancer that underwent surgery with pelvic lymphadenectomy and available preoperative 18F-FDG PET/CT have been retrospectively (March 2006 - April 2019) evaluated. Histology was used as a standard reference. PET cervical lesions were contoured with PETVCAR (GE Healthcare) using an iterative threshold. SUVmax, SUVmean, MTV, TLG and radiomics features were computed inside tumor contours using standard Image Biomarker Standardization Initiative (IBSI) methods. Radiomics features associated with lymph-node metastases were identified by Mann-Whitney test. Receiver operating characteristic (ROC) curves and area under the curve (AUC) values were computed and optimal cut-of (Youden index) was assessed. Results: Fourteen out of 74 (18.9%) patients had histologically proved nodal metastases, 5 of which identified by PET/CT (9 false negative cases). PET/CT for pelvic nodal metastases demonstrated a sensitivity, specificity, accuracy, positive predictive value and negative predictive of 36%, 93%, 82%, 55%, 86%, respectively. The presence of nodal metastasis significantly correlated with MTV (p value= 0,013; AUC= 0,72 ; cut-of= 17ml). The most significant radiomics feature was Inverse Difference-Grey Level Co-occurrence Matrix (an homogeneity index), that demonstrated a correlation with the presence of nodal metastases (p value=0,03, AUC=0,76, cut-of =0,17). No significant correlation was found for SUV parameters. Conclusions: PET/CT demonstrated low sensitivity, high specificity and high NPV in detecting pelvic nodal metastases in preoperative staging of cervical cancer. These preliminary data suggest a promising application of radiomics PET for predicting the presence of nodal metastases in cervical cancer; further studies on larger populations are needed.