Prediction of Cognitive Progression Due to Alzheimer's Disease in Normal Subjects Based on Individual Default Mode Network Metabolic Connectivity Strength

BACKGROUND

Predicting cognitive decline in those already Aβ positive or Tau positive among the aging population poses clinical challenges. In Alzheimer's disease (AD) research, intra-default mode network (DMN) connections play a pivotal role in diagnosis. This paper proposes metabolic connectivity within the DMN as a supplementary biomarker to the AT(N) framework.

METHODS

Extracting data from 1292 subjects in the Alzheimer's Disease Neuroimaging Initiative, we collected paired T1-weighted structural MRI and 18F-labeled-fluorodeoxyglucose positron emission computed tomography (PET) scans. Individual metabolic DMN networks were constructed, and metabolic connectivity (MC) strength in DMN was assessed. In the cognitively unimpaired (CU) group, the Cox model identified CU(MC+), high-risk subjects, with Kaplan-Meier survival analyses and hazard ratio (HR) revealing MC strength's predictive performance. Spearman correlation analyses explored relationships between MC strength, AT(N) biomarkers, and clinical scales. DMN standard uptake value ratio (SUVR) provided comparative insights in the analyses.

RESULTS

Both MC strength and SUVR exhibit gradual declines with cognitive deterioration, displaying significant intergroup differences. Survival analyses indicate enhanced Aβ and Tau prediction with both metrics, with MC strength outperforming SUVR. Combined MC strength and Aβ yield optimal predictive performance (HR = 9.29), followed by MC strength and Tau (HR = 8.92). In CU(MC+), MC strength correlates significantly with CSF Aβ42 and AV45 PET SUVR (r = 0.22, -0.19). Generally, MC strength's correlation with AT(N) biomarkers exceeded SUVR.

CONCLUSIONS

Individuals with normal cognition and disrupted DMN metabolic connectivity face an elevated cognitive decline risk linked to Aβ, preceding metabolic issues.

Intra and peritumoral PET radiomics analysis to predict the pathological response in breast cancer patients receiving neoadjuvant chemotherapy

OBJECTIVE

The aim of our study was to evaluate the contribution of 18Fluorine-Fluorodeoxyglucose Positron Emission Tomography (18F-FDG PET) radiomic data obtained from both the tumoral and peritumoral area in predicting pathological complete response (pCR) in patients with locally advanced breast cancer receiving neoadjuvant chemotherapy (NAC).

METHODS

Female patients with a diagnosis of invasive ductal carcinoma who received NAC were evaluated retrospectively. The volume of interest (VOI) of the primary tumor (VOI-T) was manually segmented, then a voxel-thick VOI was added around VOI-T to define the peritumoral area (VOI-PT). Morphological, intensity-based, histogram and texture parameters were obtained from VOIs. The patients were divided into two groups as pCR and non-complete pathological response (npCR). A "radiomic model" was created with only radiomic features, and a "patho-radiomic model" was created using radiomic features and immunohistochemical data.

RESULTS

Of the 66 patients included in the study, 21 were in the pCR group. The only statistically significant feature from the primary tumor among patients with pCR and npCR was Morphological_Compacity-T (AUC: 0.666). Between response groups, a significant difference was detected in 2 morphological, 1 intensity, 4 texture features from VOI-PT; no correlation was found between Morphological_Compacity-PT and NGTDM_contrast-PT. The obtained radiomic model's sensitivity and accuracy values were calculated as 61.9% and 75.8%, respectively (AUC: 0.786). When HER2 status was added, sensitivity and accuracy values of the patho-radiomic model increased to 85.7% and 81.8%, respectively (AUC: 0.903).

CONCLUSIONS

Evaluation of PET peritumoral radiomic features together with the primary tumor, rather than just the primary tumor, provides a better prediction of the pCR to NAC in patients with breast cancer.

A 3D convolutional neural network to classify subjects as Alzheimer's disease, frontotemporal dementia or healthy controls using brain 18F-FDG PET

With the arrival of disease-modifying drugs, neurodegenerative diseases will require an accurate diagnosis for optimal treatment. Convolutional neural networks are powerful deep learning techniques that can provide great help to physicians in image analysis. The purpose of this study is to introduce and validate a 3D neural network for classification of Alzheimer's disease (AD), frontotemporal dementia (FTD) or cognitively normal (CN) subjects based on brain glucose metabolism. Retrospective [18F]-FDG-PET scans of 199 CE, 192 FTD and 200 CN subjects were collected from our local database, Alzheimer's disease and frontotemporal lobar degeneration neuroimaging initiatives. Training and test sets were created using randomization on a 90 %-10 % basis, and training of a 3D VGG16-like neural network was performed using data augmentation and cross-validation. Performance was compared to clinical interpretation by three specialists in the independent test set. Regions determining classification were identified in an occlusion experiment and Gradient-weighted Class Activation Mapping. Test set subjects were age- and sex-matched across categories. The model achieved an overall 89.8 % accuracy in predicting the class of test scans. Areas under the ROC curves were 93.3 % for AD, 95.3 % for FTD, and 99.9 % for CN. The physicians' consensus showed a 69.5 % accuracy, and there was substantial agreement between them (kappa = 0.61, 95 % CI: 0.49-0.73). To our knowledge, this is the first study to introduce a deep learning model able to discriminate AD and FTD based on [18F]-FDG PET scans, and to isolate CN subjects with excellent accuracy. These initial results are promising and hint at the potential for generalization to data from other centers.

Synthesizing images of tau pathology from cross-modal neuroimaging using deep learning

Given the prevalence of dementia and the development of pathology-specific disease-modifying therapies, high-value biomarker strategies to inform medical decision-making are critical. In vivo tau-PET is an ideal target as a biomarker for Alzheimer's disease diagnosis and treatment outcome measure. However, tau-PET is not currently widely accessible to patients compared to other neuroimaging methods. In this study, we present a convolutional neural network (CNN) model that imputes tau-PET images from more widely available cross-modality imaging inputs. Participants (n = 1192) with brain T1-weighted MRI (T1w), fluorodeoxyglucose (FDG)-PET, amyloid-PET and tau-PET were included. We found that a CNN model can impute tau-PET images with high accuracy, the highest being for the FDG-based model followed by amyloid-PET and T1w. In testing implications of artificial intelligence-imputed tau-PET, only the FDG-based model showed a significant improvement of performance in classifying tau positivity and diagnostic groups compared to the original input data, suggesting that application of the model could enhance the utility of the metabolic images. The interpretability experiment revealed that the FDG- and T1w-based models utilized the non-local input from physically remote regions of interest to estimate the tau-PET, but this was not the case for the Pittsburgh compound B-based model. This implies that the model can learn the distinct biological relationship between FDG-PET, T1w and tau-PET from the relationship between amyloid-PET and tau-PET. Our study suggests that extending neuroimaging's use with artificial intelligence to predict protein specific pathologies has great potential to inform emerging care models.

False positive metastatic disease due to combined thoracic and subcutaneous splenosis

A 56-year-old man presented with dyspnea secondary to pulmonary emboli and dilated cardiomyopathy. His past medical history included a history of emergency laparotomy, splenectomy, and splenic flexure resection following a gunshot injury 30 years ago. CT and MRI imaging demonstrated multiple homogeneously enhancing lobulated lesions at the left-sided pleura and chest wall with an irregular calcified spleen. The aforementioned lesions demonstrated a similar level of tracer uptake to the splenic activity with no evidence of other FDG avid malignancy on the follow-up 18F-FDG PET study. All the above-mentioned pleural and chest wall lesions demonstrated intense tracer accumulation on technetium-99m labeled heat-damaged red cell scintigraphy, consistent with combined thoracic and subcutaneous splenosis.

Characterisation and management of expected and unexpected urgent findings from positron emission tomography with 18F-fluorodeoxyglucose integrated with computed tomography in cardiovascular disease

BACKGROUND

Cardiac 18F-fluorodeoxyglucose (FDG)-PET-CT plays an important role in the assessment of cardiovascular diseases. Effective management of urgent scan findings facilitates optimal patient care.

METHODS

We characterised the management of urgent, expected and unexpected findings in patients referred for cardiac [18F]fluorodeoxyglucose integrated with computed tomography (FDG-PET-CT) at the Royal Brompton Hospital (United Kingdom). Urgent findings are escalated by the reporting physicians/radiologists raising RadAlert notifications to the referring clinician. We characterised the indications and time to management (TTM) between the RadAlert and the resulting management. As controls, we characterised the TTM of 33 urgent findings identified before the RadAlert system was implemented.

RESULTS

Of the 1497 consecutive FDG-PET-CT scans screened (April 2021 to February 2023), 93 RadAlerts were suitable for analysis (TTM 7 days [interquartile range: 2-14]). Expected urgent findings included active cardiac sarcoidosis (56%; TTM 8 days [5-18]), heart transplant rejection (12%; 6 ± 4 days), infective endocarditis (9%; 2 days [1-12]), cardiac device infections (5%; 1 day [0-2]), acute myocarditis (2%; 5 and 14 days) and epicardial mass (1%; 1 day). TTM did not differ significantly between indications (P = 0.06). RadAlert cases had significantly shorter TTM than controls without RadAlert, P = 0.001. After the RadAlerts, 81% of patients had clinical reviews, and 55% had escalation of medical/surgical therapies. Unexpected findings (total N = 45; median TTM 6 days [1-10]) included malignancies (N = 3), infections (N = 2), pneumothorax (N = 1), benign diagnosis (N = 30), unclear diagnosis (N = 5) and 4 findings disappeared on repeat imaging.

CONCLUSIONS

Cardiac FDG-PET-CT identifies expected and unexpected findings in a range of cardiovascular diseases. Serious, unexpected findings are rare and can be effectively escalated by the RadAlert system.

Current Role and Future Prospects of Positron Emission Tomography (PET)/Computed Tomography (CT) in the Management of Breast Cancer

Breast cancer has become the most diagnosed cancer in women globally, with 2.3 million new diagnoses each year. Accurate early staging is essential for improving survival rates with metastatic spread from loco regional to distant metastasis, decreasing mortality rates by 50%. Current guidelines do not advice the routine use of positron emission tomography (PET)-computed tomography (CT) in the staging of early breast cancer in the absence of symptoms. However, there is a growing body of evidence to suggest that the use of PET-CT in this early stage can benefit the patient by improving staging and as a result treatment and outcomes, as well as psychological burden, without increasing costs to the health service. Ongoing research in PET radiomics and artificial intelligence is showing promising future prospects in its use in diagnosis, staging, prognostication, and assessment of responses to the treatment of breast cancer. Furthermore, ongoing research to address current limitations of PET-CT by improving techniques and tracers is encouraging. In this narrative review, we aim to evaluate the current evidence of the usefulness of PET-CT in the management of breast cancer in different settings along with its future prospects, including the use of artificial intelligence (AI), radiomics, and novel tracers.

Glucose metabolism in posterior cingulate cortex has supplementary value to predict the progression of cognitively unimpaired to dementia due to Alzheimer's disease: an exploratory study of 18F-FDG-PET

Amyloid-β (Aβ) and tau are important biomarkers to predict the progression of cognitively unimpaired (CU) to dementia due to Alzheimer's disease (AD), according to the diagnosis framework from the US National Institute on Aging and the Alzheimer's Association (NIA-AA). However, it is clinically difficult to predict those subjects who were already with Aβ positive (A +) or tau positive (T +). As a typical characteristic of neurodegeneration in the diagnosis framework, the hypometabolism of the posterior cingulate cortex (PCC) has significant clinical value in the early prediction and prevention of AD. In this paper, we proposed the glucose metabolism in the PCC as a biomarker supplement to Aβ and tau biomarkers. First, we calculated the standard uptake value ratio (SUVR) of PCC based on fluorodeoxyglucose positron emission computed tomography (FDG PET) imaging. Secondly, we performed Kaplan-Meier (KM) survival analyses to explore the predictive performance of PCC SUVR, and the hazard ratio (HR) was calculated. Finally, we performed Pearson correlation analyses to explore the physiological significance of PCC SUVR. As a result, the PCC SUVR showed a consistent downward trend along the AD continuum. KM analyses showed better predictive performance when we combined PCC SUVR with cerebro-spinal fluid (CSF) Aβ42 (from HR = 2.56 to 3.00 within 5 years; from HR = 2.76 to 4.20 within 10 years) and ptau-181 (from 2.83 to 3.91 within 5 years; from HR = 2.32 to 4.17 within 10 years). There was a slight correlation between Aβ42/Aβ40 and PCC SUVR (r = 0.14, p = 0.02). In addition, several cognition scales were also correlated to PCC SUVR (from r = -0.407 to 0.383, p < 0.05). Our results showed that glucose metabolism in PCC may be a potential biomarker supplement to the Aβ and tau biomarkers to predict the progression of CU to AD.

A pilot study to evaluate the effect of CT1812 treatment on synaptic density and other biomarkers in Alzheimer's disease

BACKGROUND

Effective, disease-modifying therapeutics for the treatment of Alzheimer's disease (AD) remain a large unmet need. Extensive evidence suggests that amyloid beta (Aβ) is central to AD pathophysiology, and Aβ oligomers are among the most toxic forms of Aβ. CT1812 is a novel brain penetrant sigma-2 receptor ligand that interferes with the binding of Aβ oligomers to neurons. Preclinical studies of CT1812 have demonstrated its ability to displace Aβ oligomers from neurons, restore synapses in cell cultures, and improve cognitive measures in mouse models of AD. CT1812 was found to be generally safe and well tolerated in a placebo-controlled phase 1 clinical trial in healthy volunteers and phase 1a/2 clinical trials in patients with mild to moderate dementia due to AD. The unique objective of this study was to incorporate synaptic positron emission tomography (PET) imaging as an outcome measure for CT1812 in AD patients.

METHODS

The present phase 1/2 study was a randomized, double-blind, placebo-controlled, parallel-group trial conducted in 23 participants with mild to moderate dementia due to AD to primarily evaluate the safety of CT1812 and secondarily its pharmacodynamic effects. Participants received either placebo or 100 mg or 300 mg per day of oral CT1812 for 24 weeks. Pharmacodynamic effects were assessed using the exploratory efficacy endpoints synaptic vesicle glycoprotein 2A (SV2A) PET, fluorodeoxyglucose (FDG) PET, volumetric MRI, cognitive clinical measures, as well as cerebrospinal fluid (CSF) biomarkers of AD pathology and synaptic degeneration.

RESULTS

No treatment differences relative to placebo were observed in the change from baseline at 24 weeks in either SV2A or FDG PET signal, the cognitive clinical rating scales, or in CSF biomarkers. Composite region volumetric MRI revealed a trend towards tissue preservation in participants treated with either dose of CT1812, and nominally significant differences with both doses of CT1812 compared to placebo were found in the pericentral, prefrontal, and hippocampal cortices. CT1812 was safe and well tolerated.

CONCLUSIONS

The safety findings of this 24-week study and the observed changes on volumetric MRI with CT1812 support its further clinical development.

TRIAL REGISTRATION

The clinical trial described in this manuscript is registered at clinicaltrials.gov (NCT03493282).

Use of [18F]Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography after Curative Treatment of Non-Small-Cell Lung Cancer Patients: A Nationwide Cohort Study

[¹⁸F]Fluorodeoxyglucose positron emission tomography/computed tomography ([¹⁸F]FDG PET/CT) is a valuable imaging tool in the post-treatment management of non-small-cell lung cancer (NSCLC). This study aimed to investigate the trends in utilization and factors associated with the use of [¹⁸F]FDG PET/CT after curative-intent treatment. Data from 13,758 NSCLC patients diagnosed between 2007 and 2020 identified in the Danish Lung Cancer Registry, who underwent curative-intent treatment, were analyzed using multivariable regression. The results showed a significant increase in the use of [¹⁸F]FDG PET/CT scans, from 10.4 per 100 patients per year in 2007 to 39.6 in 2013, followed by a period of stability. Higher utilization rates were observed in patients who received radiotherapy (22% increase compared to surgical resection) and in patients with stage II-III disease (14% and 20% increase compared to stage I, respectively). Additionally, utilization was increased when other diagnostic procedures were performed, such as MRI, ultrasound, endoscopy, and biopsy. These findings highlight an increasing reliance on [¹⁸F]FDG PET/CT in post-treatment NSCLC, especially after radiotherapy and in patients with locally advanced disease, where treatment-induced radiographic changes and an increased risk of recurrence present a significant diagnostic challenge.

PCA and logistic regression in 2-[18F]FDG PET neuroimaging as an interpretable and diagnostic tool for Alzheimer's disease

Objective.to develop an optimization and training pipeline for a classification model based on principal component analysis and logistic regression using neuroimages from PET with 2-[18F]fluoro-2-deoxy-D-glucose (FDG PET) for the diagnosis of Alzheimer's disease (AD).Approach.as training data, 200 FDG PET neuroimages were used, 100 from the group of patients with AD and 100 from the group of cognitively normal subjects (CN), downloaded from the repository of the Alzheimer's Disease Neuroimaging Initiative (ADNI). Regularization methods L1 and L2 were tested and their respective strength varied by the hyperparameter C. Once the best combination of hyperparameters was determined, it was used to train the final classification model, which was then applied to test data, consisting of 192 FDG PET neuroimages, 100 from subjects with no evidence of AD (nAD) and 92 from the AD group, obtained at the Centro de Diagnóstico por Imagem (CDI).Main results.the best combination of hyperparameters was L1 regularization andC≈ 0.316. The final results on test data were accuracy = 88.54%, recall = 90.22%, precision = 86.46% and AUC = 94.75%, indicating that there was a good generalization to neuroimages outside the training set. Adjusting each principal component by its respective weight, an interpretable image was obtained that represents the regions of greater or lesser probability for AD given high voxel intensities. The resulting image matches what is expected by the pathophysiology of AD.Significance.our classification model was trained on publicly available and robust data and tested, with good results, on clinical routine data. Our study shows that it serves as a powerful and interpretable tool capable of assisting in the diagnosis of AD in the possession of FDG PET neuroimages. The relationship between classification model output scores and AD progression can and should be explored in future studies.

Distinct subtypes of spatial brain metabolism patterns in Alzheimer's disease identified by deep learning-based FDG PET clusters

PURPOSE

Alzheimer's disease (AD) is a heterogeneous disease that presents a broad spectrum of clinicopathologic profiles. To date, objective subtyping of AD independent of disease progression using brain imaging has been required. Our study aimed to extract representations of unique brain metabolism patterns different from disease progression to identify objective subtypes of AD.

METHODS

A total of 3620 FDG brain PET images with AD, mild cognitive impairment (MCI), and cognitively normal (CN) were obtained from the ADNI database from 1607 participants at enrollment and follow-up visits. A conditional variational autoencoder model was trained on FDG brain PET images of AD patients with the corresponding condition of AD severity score. The k-means algorithm was applied to generate clusters from the encoded representations. The trained deep learning-based cluster model was also transferred to FDG PET of MCI patients and predicted the prognosis of subtypes for conversion from MCI to AD. Spatial metabolism patterns, clinical and biological characteristics, and conversion rate from MCI to AD were compared across the subtypes.

RESULTS

Four distinct subtypes of spatial metabolism patterns in AD with different brain pathologies and clinical profiles were identified: (i) angular, (ii) occipital, (iii) orbitofrontal, and (iv) minimal hypometabolic patterns. The deep learning model was also successfully transferred for subtyping MCI, and significant differences in frequency (P < 0.001) and risk of conversion (log-rank P < 0.0001) from MCI to AD were observed across the subtypes, highest in S2 (35.7%) followed by S1 (23.4%).

CONCLUSION

We identified distinct subtypes of AD with different clinicopathologic features. The deep learning-based approach to distinguish AD subtypes on FDG PET could have implications for predicting individual outcomes and provide a clue to understanding the heterogeneous pathophysiology of AD.

The feasibility of quantitative assessment of dynamic 18F-fluorodeoxyglucose PET in Takayasu's arteritis: a pilot study

PURPOSE

PET has been demonstrated to be sensitive for detecting active inflammation in Takayasu's arteritis (TAK) patients, but semi-quantitative-based assessment may be susceptible to various biological and technical factors. Absolute quantification via dynamic PET (dPET) may provide a more reliable and quantitative assessment of TAK-active arteries. The purpose of this study was to investigate the feasibility and efficacy of dPET in quantifying TAK-active arteries compared to static PET.

MATERIALS AND METHODS

This prospective study enrolled 10 TAK-active patients (fulfilled the NIH criteria) and 5 control participants from March to October 2022. One-hour dPET scan (all TAK and control participants) and delayed static PET scan at 2-h (all TAK patients) were acquired. For 1-h static PET, summed images from 50 to 60 min of the dPET were extracted. PET parameters derived from 1- and 2-h static PET including SUV (SUV1H and SUV2H), target-to-background ratio (TBR) (TBR1H and TBR2H), net influx rate (Ki), and TBRKi extracted from dPET were obtained. The detectability of TAK-active arteries was compared among different scanning methods using the generalized estimating equation (GEE) with a logistic regression with repeated measures, and the GEE with gamma distribution and log link function was used to evaluate the different study groups or scanning methods.

RESULTS

Based on the disease states, 5 cases of TAK were classified as untreated and relapsed, respectively. The SUVmax on 2-h PET was higher than that on 1-h PET in the untreated patients (P < 0.05). However, no significant differences were observed in the median SUVmax between 1-h PET and 2-h PET in the relapsed patients (P > 0.05). The TBRKi was significantly higher than both TBR1H and TBR2H (all P < 0.001). Moreover, the detectability of TAK-active arteries by dPET-derived Ki was significantly higher than 1-h and 2-h PET (all P < 0.001). Significant differences were observed in Kimax, SUVmax-1H, TBR1H, and TBRKi among untreated, relapsed, and control groups (all P < 0.05).

CONCLUSIONS

Absolute quantitative assessment by dPET provides an improved sensitivity and detectability in both visualization and quantification of TAK-active arteries. This elucidates the clinical significance of dPET in the early detection of active inflammation and monitoring recurrence.

Prognostic value of baseline metabolic tumor volume (MTV) for forecasting chemotherapy outcome in early-stage unfavorable Hodgkin lymphoma: Data from the phase III HD17 trial

OBJECTIVES

The prognostic relevance of metabolic tumor volume (MTV) having recently been demonstrated in patients with early-stage favorable and advanced-stage Hodgkin lymphoma. The current study aimed to assess the potential prognostic value of 18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET) in early-stage unfavorable Hodgkin lymphoma patients treated within the German Hodgkin Study Group HD17 trial.

METHODS

18 F-FDG PET/CT images were available for MTV analysis in 154 cases. We used three different threshold methods (SUV2.5 , SUV4.0 , and SUV41% ) to calculate MTV. Receiver-operating-characteristic analysis was performed to describe the value of these parameters in predicting an adequate therapy response. Therapy response was evaluated as PET negativity after 2 cycles of eBEACOPP followed by 2 cycles of ABVD.

RESULTS

All three threshold methods analyzed for MTV showed a positive correlation with the PET response after chemotherapy. Areas under the curve (AUC) were 0.70 (95% CI 0.53-0.87) and 0.65 (0.50-0.80) using the fixed thresholds of SUV4.0 and SUV2.5 , respectively, for MTV- calculation. The calculation of MTV using a relative threshold of SUV41% showed an AUC of 0.63 (0.47-0.79).

CONCLUSIONS

MTV does have predictive value after chemotherapy in early-stage unfavorable Hodgkin lymphoma, particularly when the fixed threshold of SUV4.0 is used for MTV calculation.

TRIAL REGISTRATION

ClinicalTrials.gov NCT01356680.

Long-Term Follow-Up of Tamoxifen Treatment and the Use of Imaging in Psammocarcinoma: A Case Report, Review of the Literature and Discussion of Diagnostic and Therapeutic Challenges

Psammocarcinoma (PsC) represents a rare form of low-grade serous tumor of the ovary or peritoneum. Although ovarian cancer generally has a poor prognosis in its late stages, PsC seems to have a more indolent course. We present a patient with a history of unspecific abdominal pain for more than a year, with sudden acute onset of severe inguinal pain. On admission to the hospital, a computed tomography (CT) revealed a pelvic mass of suspected ovarian origin. Radical surgery was attempted but not achieved due to widespread tumor growth. Histopathological evaluation revealed estrogen receptor-positive stage III PsC. Tamoxifen treatment was thus initiated, still maintaining stable disease 10 years later. The patient has undergone extensive radiological work-up, including CT, chest X-ray, 18F-fluoro-deoxy-glucose positron emission tomography (PET)/CT, 99mTc- hydroxymethylene diphosphonate (HDP) bone scintigraphy, 18F-fluoro-thymidine (FLT) PET/CT, Tc-99m depreotide scintigraphy and magnetic resonance imaging. In conclusion, we demonstrate that PsC has characteristic radiological features and different imaging modalities can be suitable in different clinical situations. In contrast to most other ovarian cancers, PsC does not always warrant adjuvant chemotherapy, even in advanced stages. This emphasizes the need for a deeper knowledge of the biological behavior of this rare tumor, to select the optimal treatment strategy.

PSMA PET/CT in Castration-Resistant Prostate Cancer: Myth or Reality?

BACKGROUND

prostate-specific membrane antigen (PSMA) ligand PET has been recently incorporated into international guidelines for several different indications in prostate cancer (PCa) patients. However, there are still some open questions regarding the role of PSMA ligand PET in castration-resistant prostate cancer (CRPC). The aim of this work is to assess the clinical value of PSMA ligand PET/CT in patients with CRPC.

RESULTS

PSMA ligand PET has demonstrated higher detection rates in comparison to conventional imaging and allows for a significant reduction in the number of M0 CRPC patients. However, its real impact on patients' prognosis is still an open question. Moreover, in CRPC patients, PSMA ligand PET presents some sensitivity and specificity limitations. Due to its heterogeneity, CRPC may present a mosaic of neoplastic clones, some of which could be PSMA-/FDG+, or vice versa. Likewise, unspecific bone uptake (UBU) and second primary neoplasms (SNPs) overexpressing PSMA in the neoangiogenic vessels represent potential specificity issues. Integrated multi-tracer imaging (PSMA ligand and [18F]FDG PET) together with a multidisciplinary discussion could allow for reaching the most accurate evaluation of each patient from a precision medicine point of view.

Effects of COVID-19 on cognition and brain health

COVID-19 is associated with a range of neurological, cognitive, and mental health symptoms both acutely and chronically that can persist for many months after infection in people with long-COVID syndrome. Investigations of cognitive function and neuroimaging have begun to elucidate the nature of some of these symptoms. They reveal that, although cognitive deficits may be related to brain imaging abnormalities in some people, symptoms can also occur in the absence of objective cognitive deficits or neuroimaging changes. Furthermore, cognitive impairment may be detected even in asymptomatic individuals. We consider the evidence regarding symptoms, cognitive deficits, and neuroimaging, as well as their possible underlying mechanisms.

Modulation of FDG Uptake by Cell Cycle Synchronization Using a T-Type Calcium Channel Inhibitor

BACKGROUND

We investigated whether cell cycle synchronization induced by the T-type calcium channel inhibitor mibefradil could increase tumoral 2-[18F] fluoro-2-deoxy-d-glucose (FDG) uptake in vitro and in vivo.

METHODS

Human prostate cancer cells (PC-3) were treated with 10 μM mibefradil for 24, 48, and 72 h to induce G1 arrest. Cell cycle distribution was analyzed at 0, 4, 8, 12, 15, 18, and 24 h after mibefradil withdrawal. Cellular uptake was measured after incubating cells with [3H] Deoxy-d-Glucose (DDG) for 1 h at the same time points used in the cell cycle analysis. The correlation between [3H] DDG uptake and each cell cycle phase was evaluated in the early (0-12 h) and late phases (15-24 h) of synchronization. In vivo FDG PET imaging was performed in PC-3-bearing mice at baseline, 24 h, and 48 h after mibefradil treatment.

RESULTS

The G0/G1 fraction of PC-3 cells was significantly increased from 33.1% ± 0.2% to 60.9% ± 0.8% after 24 h mibefradil treatment, whereas the S and G2/M fractions were decreased from 36.3% ± 1.4% to 23.2% ± 1.1% and from 29.7% ± 1.3% to 14.9% ± 0.9%, respectively, which were similar to the results by serum starvation. Mibefradil treatment for 24, 48, and 72 h increased the number of cells in S phase at 18-24 h after withdrawal; however, only the 72 h treatment increased [3H] DDG uptake (145.8 ± 5.8% of control at 24 h after withdrawal). [3H] DDG uptake was positively correlated with the size of the S phase fraction and negatively correlated with the size of the G0/G1 fraction in the late phase of synchronization. DDG uptake was significantly increased by mibefradil-induced cell cycle synchronization and correlated with the sizes of cell cycle fractions. In vivo FDG PET imaging also demonstrated a significant increase in tumor uptake after mibefradil treatment. Quantified tumor FDG uptake (%ID/g) increased from 4.13 ± 2.10 to 4.7 ± 2.16 at 24 h, and 5.95 ± 2.57 at 48 h (p < 0.05).

CONCLUSION

Cell cycle synchronization could be used to increase the diagnostic sensitivity of clinical FDG positron emission tomography.

Head and Neck Cancer Segmentation in FDG PET Images: Performance Comparison of Convolutional Neural Networks and Vision Transformers

Convolutional neural networks (CNNs) have a proven track record in medical image segmentation. Recently, Vision Transformers were introduced and are gaining popularity for many computer vision applications, including object detection, classification, and segmentation. Machine learning algorithms such as CNNs or Transformers are subject to an inductive bias, which can have a significant impact on the performance of machine learning models. This is especially relevant for medical image segmentation applications where limited training data are available, and a model's inductive bias should help it to generalize well. In this work, we quantitatively assess the performance of two CNN-based networks (U-Net and U-Net-CBAM) and three popular Transformer-based segmentation network architectures (UNETR, TransBTS, and VT-UNet) in the context of HNC lesion segmentation in volumetric [F-18] fluorodeoxyglucose (FDG) PET scans. For performance assessment, 272 FDG PET-CT scans of a clinical trial (ACRIN 6685) were utilized, which includes a total of 650 lesions (primary: 272 and secondary: 378). The image data used are highly diverse and representative for clinical use. For performance analysis, several error metrics were utilized. The achieved Dice coefficient ranged from 0.833 to 0.809 with the best performance being achieved by CNN-based approaches. U-Net-CBAM, which utilizes spatial and channel attention, showed several advantages for smaller lesions compared to the standard U-Net. Furthermore, our results provide some insight regarding the image features relevant for this specific segmentation application. In addition, results highlight the need to utilize primary as well as secondary lesions to derive clinically relevant segmentation performance estimates avoiding biases.

Functional Brain Networks to Evaluate Treatment Responses in Parkinson's Disease

Network analysis of functional brain scans acquired with [18F]-fluorodeoxyglucose positron emission tomography (FDG PET, to map cerebral glucose metabolism), or resting-state functional magnetic resonance imaging (rs-fMRI, to map blood oxygen level-dependent brain activity) has increasingly been used to identify and validate reproducible circuit abnormalities associated with neurodegenerative disorders such as Parkinson's disease (PD). In addition to serving as imaging markers of the underlying disease process, these networks can be used singly or in combination as an adjunct to clinical diagnosis and as a screening tool for therapeutics trials. Disease networks can also be used to measure rates of progression in natural history studies and to assess treatment responses in individual subjects. Recent imaging studies in PD subjects scanned before and after treatment have revealed therapeutic effects beyond the modulation of established disease networks. Rather, other mechanisms of action may be at play, such as the induction of novel functional brain networks directly by treatment. To date, specific treatment-induced networks have been described in association with novel interventions for PD such as subthalamic adeno-associated virus glutamic acid decarboxylase (AAV2-GAD) gene therapy, as well as sham surgery or oral placebo under blinded conditions. Indeed, changes in the expression of these networks with treatment have been found to correlate consistently with clinical outcome. In aggregate, these attributes suggest a role for functional brain networks as biomarkers in future clinical trials.

Longitudinal changes in metabolic network activity in early Alzheimer's disease

INTRODUCTION

The progression of Alzheimer's disease (AD) has been linked to two metabolic networks, the AD-related pattern (ADRP) and the default mode network (DMN).

METHODS

Converting and clinically stable cognitively normal subjects (n = 47) and individuals with mild cognitive impairment (n = 96) underwent 2-[18 F]fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET) three or more times over 6 years (nscans  = 705). Expression levels for ADRP and DMN were measured in each subject and time point, and the resulting changes were correlated with cognitive performance. The role of network expression in predicting conversion to dementia was also evaluated.

RESULTS

Longitudinal increases in ADRP expression were observed in converters, while age-related DMN loss was seen in converters and nonconverters. Cognitive decline correlated with increases in ADRP and declines in DMN, but conversion to dementia was predicted only by baseline ADRP levels.

DISCUSSION

The results point to the potential utility of ADRP as an imaging biomarker of AD progression.

Metabolic Brain Changes Can Predict the Underlying Pathology in Neurodegenerative Brain Disorders: A Case Report of Sporadic Creutzfeldt-Jakob Disease with Concomitant Parkinson's Disease

The co-occurrence of multiple proteinopathies is being increasingly recognized in neurodegenerative disorders and poses a challenge in differential diagnosis and patient selection for clinical trials. Changes in brain metabolism captured by positron emission tomography (PET) with 18 F-fluorodeoxyglucose (FDG) allow us to differentiate between different neurodegenerative disorders either by visual exploration or by studying disease-specific metabolic networks in individual patients. However, the impact of multiple proteinopathies on brain metabolism and metabolic networks remains unknown due to the absence of pathological studies. In this case study, we present a 67-year-old patient with rapidly progressing dementia clinically diagnosed with probable sporadic Creutzfeldt-Jakob disease (sCJD). However, in addition to the expected pronounced cortical and subcortical hypometabolism characteristic of sCJD, the brain FDG PET revealed an intriguing finding of unexpected relative hypermetabolism in the bilateral putamina, raising suspicions of coexisting Parkinson's disease (PD). Additional investigation of disease-specific metabolic brain networks revealed elevated expression of both CJD-related pattern (CJDRP) and PD-related pattern (PDRP) networks. The patient eventually developed akinetic mutism and passed away seven weeks after symptom onset. Neuropathological examination confirmed neuropathological changes consistent with sCJD and the presence of Lewy bodies confirming PD pathology. Additionally, hyperphosphorylated tau and TDP-43 pathology were observed, a combination of four proteinopathies that had not been previously reported. Overall, this case provides valuable insights into the complex interplay of neurodegenerative pathologies and their impact on metabolic brain changes, emphasizing the role of metabolic brain imaging in evaluating potential presence of multiple proteinopathies.

Altered brain metabolism in frontotemporal dementia and psychiatric disorders: involvement of the anterior cingulate cortex

BACKGROUND

Behavioural symptoms and frontotemporal hypometabolism overlap between behavioural variant of frontotemporal dementia (bvFTD) and primary psychiatric disorders (PPD), hampering diagnostic distinction. Voxel-wise comparisons of brain metabolism might identify specific frontotemporal-(hypo)metabolic regions between bvFTD and PPD. We investigated brain metabolism in bvFTD and PPD and its relationship with behavioural symptoms, social cognition, severity of depressive symptoms and cognitive functioning.

RESULTS

Compared to controls, bvFTD showed decreased metabolism in the dorsal anterior cingulate cortex (dACC) (p < 0.001), orbitofrontal cortex (OFC), temporal pole, dorsolateral prefrontal cortex (dlPFC) and caudate, whereas PPD showed no hypometabolism. Compared to PPD, bvFTD showed decreased metabolism in the dACC (p < 0.001, p < 0.05_FWE), insula, Broca's area, caudate, thalamus, OFC and temporal cortex (p < 0.001), whereas PPD showed decreased metabolism in the motor cortex (p < 0.001). Across bvFTD and PPD, decreased metabolism in the temporal cortex (p < 0.001, p < 0.05_FWE), dACC and frontal cortex was associated with worse social cognition. Decreased metabolism in the dlPFC was associated with compulsiveness (p < 0.001). Across bvFTD, PPD and controls, decreased metabolism in the PFC and motor cortex was associated with executive dysfunctioning (p < 0.001).

CONCLUSIONS

Our findings indicate subtle but distinct metabolic patterns in bvFTD and PPD, most strongly in the dACC. The degree of frontotemporal and cingulate hypometabolism was related to impaired social cognition, compulsiveness and executive dysfunctioning. Our findings suggest that the dACC might be an important region to differentiate between bvFTD and PPD but needs further validation.

Prognostic Utility of Neck Lymph Node-to-Primary Tumor Standardized Uptake Value Ratio in Oral Cavity Cancer

We investigated the prognostic utility of preoperative neck lymph node-to-primary tumor maximum standardized uptake value ratios (NTRs) in oral cavity squamous cell carcinoma (OSCC). We retrospectively reviewed the medical records of 141 consecutive patients who were diagnosed as having OSCC and had received fluorodeoxyglucose-positron emission tomography within 2 weeks prior to radical surgery between 2009 and 2018. To determine the optimal NTR cutoff, receiver operating characteristic analysis for overall survival (OS) was executed. The NTR's prognostic value for disease-free survival (DFS) and OS were determined through Cox proportional hazards analysis and the Kaplan-Meier method. We determined the median (range) follow-up duration to be 35.2 (2.1-122.4) months. The optimal NTR cutoff was 0.273, and patients with a higher NTR (≥0.273) exhibited significantly worse DFS and OS (p = 0.010 and 0.003, respectively). A higher NTR (≥0.273) predicted poorer DFS (hazard ratio: 2.696, p = 0.008) and OS (hazard ratio: 4.865, p = 0.003) in multivariable analysis. We created a nomogram on the basis of the NTR, and it could accurately predict OS (concordance index: 0.774). Preoperative NTRs may be a useful prognostic biomarker for DFS and OS in patients with OSCC who have undergone surgery. NTR-based nomograms may also be helpful prognostic tools in clinical trials.

A convolutional neural network with self-attention for fully automated metabolic tumor volume delineation of head and neck cancer in [Formula: see text]F]FDG PET/CT

PURPOSE

PET-derived metabolic tumor volume (MTV) and total lesion glycolysis of the primary tumor are known to be prognostic of clinical outcome in head and neck cancer (HNC). Including evaluation of lymph node metastases can further increase the prognostic value of PET but accurate manual delineation and classification of all lesions is time-consuming and prone to interobserver variability. Our goal, therefore, was development and evaluation of an automated tool for MTV delineation/classification of primary tumor and lymph node metastases in PET/CT investigations of HNC patients.

METHODS

Automated lesion delineation was performed with a residual 3D U-Net convolutional neural network (CNN) incorporating a multi-head self-attention block. 698 [Formula: see text]F]FDG PET/CT scans from 3 different sites and 5 public databases were used for network training and testing. An external dataset of 181 [Formula: see text]F]FDG PET/CT scans from 2 additional sites was employed to assess the generalizability of the network. In these data, primary tumor and lymph node (LN) metastases were interactively delineated and labeled by two experienced physicians. Performance of the trained network models was assessed by 5-fold cross-validation in the main dataset and by pooling results from the 5 developed models in the external dataset. The Dice similarity coefficient (DSC) for individual delineation tasks and the primary tumor/metastasis classification accuracy were used as evaluation metrics. Additionally, a survival analysis using univariate Cox regression was performed comparing achieved group separation for manual and automated delineation, respectively.

RESULTS

In the cross-validation experiment, delineation of all malignant lesions with the trained U-Net models achieves DSC of 0.885, 0.805, and 0.870 for primary tumor, LN metastases, and the union of both, respectively. In external testing, the DSC reaches 0.850, 0.724, and 0.823 for primary tumor, LN metastases, and the union of both, respectively. The voxel classification accuracy was 98.0% and 97.9% in cross-validation and external data, respectively. Univariate Cox analysis in the cross-validation and the external testing reveals that manually and automatically derived total MTVs are both highly prognostic with respect to overall survival, yielding essentially identical hazard ratios (HR) ([Formula: see text]; [Formula: see text] vs. [Formula: see text]; [Formula: see text] in cross-validation and [Formula: see text]; [Formula: see text] vs. [Formula: see text]; [Formula: see text] in external testing).

CONCLUSION

To the best of our knowledge, this work presents the first CNN model for successful MTV delineation and lesion classification in HNC. In the vast majority of patients, the network performs satisfactory delineation and classification of primary tumor and lymph node metastases and only rarely requires more than minimal manual correction. It is thus able to massively facilitate study data evaluation in large patient groups and also does have clear potential for supervised clinical application.

Imaging Cancer in Neuroradiology

Neuroimaging plays a pivotal role in the diagnosis, management, and prognostication of brain tumors. Recently, the World Health Organization published the fifth edition of the WHO Classification of Tumors of the Central Nervous System (CNS5), which places greater emphasis on tumor genetics and molecular markers to complement the existing histological and immunohistochemical approaches. Recent advances in computational power allowed modern neuro-oncological imaging to move from a strictly morphology-based discipline to advanced neuroimaging techniques with quantifiable tissue characteristics such as tumor cellularity, microstructural organization, hemodynamic, functional, and metabolic features, providing more precise tumor diagnosis and management. The aim of this review is to highlight the key imaging features of the recently published CNS5, outlining the current imaging standards and summarizing the latest advances in neuro-oncological imaging techniques and their role in complementing traditional brain tumor imaging and management.

Diagnostic Value of FDG PET/CT in Surveillance after Curative Resection of Breast Cancer

With increasing incidence of breast cancer and improvement in treatment, the concern about surveillance management also has increased. This retrospective study was designed to evaluate the diagnostic value of routine surveillance FDG PET/CT in patients with breast cancer. The diagnostic performance of surveillance PET/CT was analyzed regarding sensitivity, specificity, positive predictive value, negative predictive value and accuracy. The diagnostic accuracy was defined as the ability to differentiate recurrence and no-disease correctly and the proportion of true results, either true positive or true negative, in the population. Findings from pathologic examination; other imaging modalities such as CT, MRI and bone scan; or clinical follow-up were used as the reference standard. In this study of 1681 consecutive patients with breast cancer who underwent curative surgery, surveillance fluorodeoxyglucose PET/CT showed good diagnostic performance in the detection of clinically unexpected recurrent breast cancer or other malignancy, with a sensitivity of 100%, specificity of 98.5%, positive predictive value of 70.5%, negative predictive value of 100% and accuracy of 98.5%. In conclusion, surveillance fluorodeoxyglucose PET/CT showed good diagnostic performance in the detection of clinically unexpected recurrent breast cancer after curative surgery.

Generation of Whole-Body FDG Parametric Ki Images from Static PET Images Using Deep Learning

FDG parametric Ki images show great advantage over static SUV images, due to the higher contrast and better accuracy in tracer uptake rate estimation. In this study, we explored the feasibility of generating synthetic Ki images from static SUV ratio (SUVR) images using three configurations of U-Nets with different sets of input and output image patches, which were the U-Nets with single input and single output (SISO), multiple inputs and single output (MISO), and single input and multiple outputs (SIMO). SUVR images were generated by averaging three 5-min dynamic SUV frames starting at 60 minutes post-injection, and then normalized by the mean SUV values in the blood pool. The corresponding ground truth Ki images were derived using Patlak graphical analysis with input functions from measurement of arterial blood samples. Even though the synthetic Ki values were not quantitatively accurate compared with ground truth, the linear regression analysis of joint histograms in the voxels of body regions showed that the mean R2 values were higher between U-Net prediction and ground truth (0.596, 0.580, 0.576 in SISO, MISO and SIMO), than that between SUVR and ground truth Ki (0.571). In terms of similarity metrics, the synthetic Ki images were closer to the ground truth Ki images (mean SSIM = 0.729, 0.704, 0.704 in SISO, MISO and MISO) than the input SUVR images (mean SSIM = 0.691). Therefore, it is feasible to use deep learning networks to estimate surrogate map of parametric Ki images from static SUVR images.

Optimized 18F-FDG PET-CT Method to Improve Accuracy of Diagnosis of Metastatic Cancer

The diagnosis of cancer by FDG PET-CT is often inaccurate owing to subjectivity of interpretation. We compared the accuracy of a novel normalized (standardized) method of interpretation with conventional non-normalized SUV. Patients (n = 393) with various malignancies were studied with FDG PET/CT to determine the presence or absence of cancer. Target lesions were assessed by two methods: (1) conventional SUV_max (conSUV_max) and (2) a novel method that combined multiple factors to optimize SUV (optSUVmax), including the patient's normal liver SUV_max, a liver constant (k) derived from a review of the literature, and use of site-specific thresholds for malignancy. The two methods were compared to pathology findings in 154 patients being evaluated for mediastinal and/or hilar lymph node (MHLNs) metastases, 143 evaluated for extra-thoracic lymph node (ETLNs) metastases, and 96 evaluated for liver metastases. OptSUV_max was superior to conSUV_max for all patient groups. For MHLNs, sensitivity was 83.8% vs. 80.7% and specificity 88.7% vs. 9.6%, respectively; for ETLNs, sensitivity was 92.1% vs. 77.8% and specificity 80.1% vs. 27.6%, respectively; and for lesions in the liver parenchyma, sensitivity was 96.1% vs. 82.3% and specificity 88.8% vs. 23.0%, respectively. Optimized SUV_max increased diagnostic accuracy of FDG PET-CT for cancer when compared with conventional SUV_max interpretation.

Quantitative MRI to Characterize Hypoxic Tumors in Comparison to FMISO PET/CT for Radiotherapy in Oropharynx Cancers

Intratumoral hypoxia is associated with a poor prognosis and poor response to treatment in head and neck cancers. Its identification would allow for increasing the radiation dose to hypoxic tumor subvolumes. 18F-FMISO PET imaging is the gold standard; however, quantitative multiparametric MRI could show the presence of intratumoral hypoxia. Thus, 16 patients were prospectively included and underwent 18F-FDG PET/CT, 18F-FMISO PET/CT, and multiparametric quantitative MRI (DCE, diffusion and relaxometry T1 and T2 techniques) in the same position before treatment. PET and MRI sub-volumes were segmented and classified as hypoxic or non-hypoxic volumes to compare quantitative MRI parameters between normoxic and hypoxic volumes. In total, 13 patients had hypoxic lesions. The Dice, Jaccard, and overlap fraction similarity indices were 0.43, 0.28, and 0.71, respectively, between the FDG PET and MRI-measured lesion volumes, showing that the FDG PET tumor volume is partially contained within the MRI tumor volume. The results showed significant differences in the parameters of SUV in FDG and FMISO PET between patients with and without measurable hypoxic lesions. The quantitative MRI parameters of ADC, T1 max mapping and T2 max mapping were different between hypoxic and normoxic subvolumes. Quantitative MRI, based on free water diffusion and T1 and T2 mapping, seems to be able to identify intra-tumoral hypoxic sub-volumes for additional radiotherapy doses.

PET imaging characteristics of neuromuscular choristoma and associated desmoid-type fibromatosis

BACKGROUND

Neuromuscular choristoma (NMC) is a rare peripheral nerve lesion characterized by abnormal presence of muscle within nerve. Associated desmoid-type fibromatosis (NMC-DTF) often develops. We report ¹⁸F-fluorodeoxyglucose positron emission tomography (FDG PET) characteristics of NMC and NMC-DTF and propose that increased FDG activity within NMCs may be associated with subclinical NMC-DTF or NMC-DTF "precursor" tissue.

METHODS

Our institutional database was searched for all NMC cases. Inclusion criteria were 1) confirmed diagnosis of NMC with or without biopsy, and 2) available PET and MRI studies. PET data included SUVmax and SUVmean of NMCs, contralateral limb normal skeletal muscle and unaffected nerves, and SUVmax of NMC-DTF if present. SUV values were compared using paired t-test. A p value of < 0.05 was considered statistically significant.

RESULTS

Our cohort consisted of 9 patients with NMC, 8 cases involving sciatic nerve and 1 of brachial plexus. On PET imaging, all NMC-affected nerve segments showed significantly higher FDG uptake (SUVmax/mean) compared to both contralateral normal nerve and normal skeletal muscle (all P < 0.05). Similar to sporadic DTF, NMC-DTF was highly FDG-avid (average SUVmax of 4.2). SUVmax in NMC with or without concurrent NMC-DTF did not differ (p = 0.76). Within NMC-affected nerve segment, FDG activity was relatively higher in areas with low T1/T2 MR signal.

CONCLUSION

All NMCs were more FDG avid compared to both normal skeletal muscle and contralateral unaffected nerve, arguing against the presence of heterotopic muscle in NMC as the source of FDG avidity. FDG avidity within NMC may reflect subclinical NMC-DTF or a precursor lesion, as NMC-DTF are highly FDG-avid, and the highest regions of FDG avidity in NMC occurred in regions with MR characteristics associated with NMC-DTF (i.e., lower T1/T2 signal). We believe that the integration of FDG PET with serial MR imaging in patient follow up will clarify its utility in both detection and surveillance of NMC-DTF.

Consensus statement on the management of incidentally discovered FDG avid thyroid nodules in patients being investigated for other cancers

With the widespread use of 18F-fluorodeoxyglucose positron emission tomography (FDG PET/CT) in the investigation and staging of cancers, incidental discovery of FDG-avid thyroid nodules is becoming increasingly common, with a reported incidence in the range 1%-4% of FDG PET/CT scans. The risk of malignancy in an incidentally discovered FDG avid thyroid nodule is not clear due to selection bias in reported retrospective series but is likely to be less than 15%. Even in cases where the nodule is found to be malignant, the majority will be differentiated thyroid cancers with an excellent prognosis even without treatment. If, due to index cancer diagnosis, age and co-morbidities, it is unlikely that the patient will survive 5 years, further investigation of an incidental FDG avid thyroid nodule is unlikely to be warranted. We provide a consensus statement on the circumstances in which further investigation of FDG avid thyroid nodules with ultrasound and fine needle aspiration might be appropriate.

I-PET score: Combining whole body iodine and 18 F-FDG PET/CT imaging to predict progression in structurally or biochemically incomplete thyroid cancer

OBJECTIVE

We propose a new scoring system (I-PET) combining whole body scan (WBS) and FDG findings to identify patients who have or are likely to become refractory to radioactive iodine.

DESIGN

Retrospective analysis of 142 patients age >18 with differentiated thyroid cancer who had a F-18 labelled fluoro-2-deoxyglucose (¹⁸ F-FDG) positron emission tomography (PET) and WBS within a 6-month period between 2010 and 2020. Pairs of ¹⁸ F-FDG PET and WBS were reviewed by three independent nuclear medicine physicians and an I-PET score was assigned: I-PET [0]: Iodine -ve/FDG -ve, I-PET [1]: Iodine +ve/FDG -ve, I-PET [2]: Iodine +ve/FDG +ve and I-PET [3]: Iodine -ve/FDG +ve. Patients with FDG +ve lesions (I-PET [2] and I-PET [3]) were further classified into groups A and B if SUVmax was ≤5 or >5, respectively. Follow-up data were obtained by chart review. Progression was defined as structural progression as per RECIST 1.1 or further surgical intervention; or biochemical progression as unstimulated thyroglobulin increasing >20% from baseline.

RESULTS

Of 142 patients included in the study 121 patients had follow-up data available for review. At baseline, 49 patients were classified as I-PET [0], 10 as I-PET [1], 16 as I-PET [2] and 46 as I-PET [3]. Progression was seen in 11/49 (22%) of I-PET [0], 4/10 (40%) of I-PET [1], 10/16 (63%) of I-PET [2] and 34/46 (74%) of I-PET [3] (p < 0.001). I-PET [2B] and I-PET [3B] had a progression rate of 88% (7/8) and 78% (25/32), respectively. I-PET [3B] were 9.6 times more likely to commence multikinase inhibitor therapy (p = 0.001) and had 8 times greater mortality (p = 0.003) than patients in other I-PET groups combined.

CONCLUSION

I-PET is a simple readily acquired imaging biomarker that potentially enhances the dynamic risk stratification and guide treatment in thyroid cancer.

Adaptive and pathological connectivity responses in Parkinson's disease brain networks

Functional imaging has been used extensively to identify and validate disease-specific networks as biomarkers in neurodegenerative disorders. It is not known, however, whether the connectivity patterns in these networks differ with disease progression compared to the beneficial adaptations that may also occur over time. To distinguish the 2 responses, we focused on assortativity, the tendency for network connections to link nodes with similar properties. High assortativity is associated with unstable, inefficient flow through the network. Low assortativity, by contrast, involves more diverse connections that are also more robust and efficient. We found that in Parkinson's disease (PD), network assortativity increased over time. Assoratitivty was high in clinically aggressive genetic variants but was low for genes associated with slow progression. Dopaminergic treatment increased assortativity despite improving motor symptoms, but subthalamic gene therapy, which remodels PD networks, reduced this measure compared to sham surgery. Stereotyped changes in connectivity patterns underlie disease progression and treatment responses in PD networks.

Primary pulmonary meningioma with associated multiple micronodules: a case report and literature review

Primary pulmonary meningioma (PPM) is a rare and benign slow growing tumor with good prognosis. It often presents as an asymptomatic, well-circumscribed, solitary pulmonary nodule. Wedge resection is the management of choice for both diagnosis and treatment. Here, we report one case of PPM with increased fluorodeoxyglucose (FDG) uptake and associated micronodules, which was clinically suspicious for malignancy. The patient was a 60-year-old female who presented with persistent shortness of breath for 1 year. Chest computed tomography showed a 1.5-cm well-circumscribed homogenous nodule in the left upper lobe with increased FDG uptake and multiple smaller well-circumscribed micronodules scattered in both lungs. Left upper lobe wedge resection confirmed the diagnosis of PPM. PPM can deceptively mimic malignancy, so recognizing this rare entity and including it in the differential diagnoses of pulmonary nodules, especially with avid uptake of FDG, is crucial to avoid misdiagnosis and overtreatment.

Deep learning model integrating positron emission tomography and clinical data for prognosis prediction in non-small cell lung cancer patients

BACKGROUND

Lung cancer is the leading cause of cancer-related deaths worldwide. The majority of lung cancers are non-small cell lung cancer (NSCLC), accounting for approximately 85% of all lung cancer types. The Cox proportional hazards model (CPH), which is the standard method for survival analysis, has several limitations. The purpose of our study was to improve survival prediction in patients with NSCLC by incorporating prognostic information from F-18 fluorodeoxyglucose positron emission tomography (FDG PET) images into a traditional survival prediction model using clinical data.

RESULTS

The multimodal deep learning model showed the best performance, with a C-index and mean absolute error of 0.756 and 399 days under a five-fold cross-validation, respectively, followed by ResNet3D for PET (0.749 and 405 days) and CPH for clinical data (0.747 and 583 days).

CONCLUSION

The proposed deep learning-based integrative model combining the two modalities improved the survival prediction in patients with NSCLC.

Intrinsic and secondary epileptogenicity in focal cortical dysplasia type II

OBJECTIVE

Favorable seizure outcome is reported following resection of bottom-of-sulcus dysplasia (BOSD). We assessed the distribution of epileptogenicity and dysplasia in and around BOSD to better understand this clinical outcome and the optimal surgical approach.

METHODS

We studied 27 children and adolescents with magnetic resonance imaging (MRI)-positive BOSD who underwent epilepsy surgery; 85% became seizure-free postresection (median = 5.0 years follow-up). All patients had resection of the dysplastic sulcus, and 11 had additional resection of the gyral crown (GC) or adjacent gyri (AG). Markers of epileptogenicity were relative cortical hypometabolism on preoperative 18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET), and spiking, ripples, fast ripples, spike-high-frequency oscillation cross-rate, and phase amplitude coupling (PAC) on preresection and postresection electrocorticography (ECoG), all analyzed at the bottom-of-sulcus (BOS), top-of-sulcus (TOS), GC, and AG. Markers of dysplasia were increased cortical thickness on preoperative MRI, and dysmorphic neuron density and variant allele frequency of somatic MTOR mutations in resected tissue, analyzed at similar locations.

RESULTS

Relative cortical metabolism was significantly reduced and ECoG markers were significantly increased at the BOS compared to other regions. Apart from spiking and PAC, which were greater at the TOS compared to the GC, there were no significant differences in PET and other ECoG markers between the TOS, GC, and AG, suggesting a cutoff of epileptogenicity at the TOS rather than a tapering gradient on the cortical surface. MRI and tissue markers of dysplasia were all maximal in the BOS, reduced in the TOS, and mostly absent in the GC. Spiking and PAC reduced significantly over the GC after resection of the dysplastic sulcus.

SIGNIFICANCE

These findings support the concept that dysplasia and intrinsic epileptogenicity are mostly limited to the dysplastic sulcus in BOSD and support resection or ablation confined to the MRI-visible lesion as a first-line surgical approach. 18 F-FDG PET and ECoG abnormalities in surrounding cortex seem to be secondary phenomena.

More extensive hypometabolism and higher mortality risk in patients with right- than left-predominant neurodegeneration of the anterior temporal lobe

BACKGROUND

Left-predominant neurodegeneration of the anterior temporal lobe (ATL) and the associated syndrome termed semantic variant primary progressive aphasia (svPPA) are well characterized. Less is known about right-predominant neurodegeneration of the ATL, which has been associated with the clinical syndrome named right temporal variant of frontotemporal dementia (rtvFTD). Here, we assessed glucose metabolism across the brain, cognitive performance, and mortality in patients with right-predominant neurodegeneration of the ATL.

METHODS

Patients with predominant hypometabolism of the ATL on FDG PET (as a measure of neurodegeneration) were retrospectively identified and categorized into those with asymmetrical right, left, or symmetric bilateral involvement (N = 10, 17, and 8). We compared whole-brain, normalized regional glucose metabolism using SPM12, cognitive performance on the CERAD Neuropsychological Assessment Battery, and mortality risk (age- and sex-adjusted Cox proportional hazard model) between groups.

RESULTS

Hypometabolism was most pronounced and extensive in patients with right-predominant neurodegeneration of the ATL. Beyond the right temporal lobe, right frontal and left temporal lobes were affected in these patients. Cognitive performance was similarly impaired in all three groups, with predominant naming and hippocampal-dependent memory deficits. Mortality risk was 6.1 times higher in patients with right- than left-predominant ATL neurodegeneration (p < 0.05). Median survival duration after PET was shortest in patients with right- and longest in patients with left-predominant ATL neurodegeneration (5.7 vs 8.3 years after examination).

DISCUSSION

More extensive neurodegeneration and shorter survival duration in patients with right- than left-predominant neurodegeneration of the ATL might indicate that the former consult memory clinics at a later disease stage, when symptoms like naming and episodic memory deficits have already emerged. At the time of diagnosis, the shorter survival duration of patients with right- than left-predominant ATL neurodegeneration should be kept in mind when counseling patients and caregivers.

Regional differences in the reduction in cerebral FDG uptake induced by the ketogenic diet

BACKGROUND

The postulated benefits of the ketogenic diet in the management of multiple medical conditions have seen more patients who are in therapeutic ketosis attending 18F-FDG PET scans. This study aimed to investigate the effect of ketosis on cerebral glucose metabolism in a clinical PET scanning environment using 18F-FDG uptake as a surrogate marker.

METHODS

A retrospective audit was conducted of the brain 18F-FDG uptake in 52 patients who underwent PET scans for possible cardiac sarcoidosis or suspected intracardiac infection, following a ketogenic diet and prolonged fasting. SUVbw for whole brain and separate brain regions was compared with serum glucose and serum ketone body (beta-hydroxybutyrate) levels.

RESULTS

The expected negative association between serum glucose levels and whole brain 18F-FDG uptake was confirmed. A reduction in SUVbw due to increasing serum ketones levels was also observed that was independent of and in addition to the effects of glucose. The magnitude of the reduction in SUVbw related to serum glucose level and serum ketone level was found to be greater in the precuneus than in the cerebellum or whole brain.

CONCLUSION

In a real-world clinical PET setting, cerebral 18F-FDG uptake appears to be affected by glycaemia and ketonaemia. This means when assessing the brain, both serum glucose and ketone levels need to be considered when SUVs are used to distinguish between pathologic and physiologic states. The magnitude of this effect appears to vary between different brain regions. This regional difference should be taken into consideration when selecting the appropriate brain region for SUV normalisation, particularly when undertaking database comparison in the assessment of dementia.

18-Fluorodeoxyglucose positron emission tomography in the diagnosis of prosthetic aortic graft infection: the difference between open and endovascular repair

OBJECTIVES

18-Fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/computed tomography (CT) has been reported as useful for diagnosing aortic graft infection. However, 18F-FDG uptake may depend upon various factors including open versus endovascular repair and time from surgery. We aimed to elucidate the factors influencing its uptake and the diagnostic value of 18F-FDG PET/CT after open and endovascular repair.

METHODS

Hospital database of PET/CT (N = 14 490) and our departmental database were cross-checked to identify those who underwent 18F-FDG PET/CT after aortic repair. Patient's data were retrieved from the chart. Images were reviewed by 2 nuclear medicine specialists in consensus, and the presence of increased 18F-FDG uptake was recorded. The maximum standardized uptake value (SUV max) was measured.

RESULTS

Among the 1112 patients who underwent aortic repair between 2011 and 2022, 71 patients were identified. Eighteen patients underwent 18F-FDG PET/CT for suspected graft infection and the remaining 53 patients for other purposes (malignancy, etc.). Fourteen patients were treated as aortic graft infection. They had significantly higher SUV max than those without graft infection [mean 8.64 (standard deviation 2.78) vs 3.40 (standard deviation 0.84); P < 0.01]. In the non-infected grafts, SUV max was higher early after open surgical repair, while it remained low after endovascular repair.

CONCLUSIONS

After endovascular aortic repair, a constant cut-off value of 'SUV max = 4.5' seems appropriate for diagnosing graft infection, since it remains low and stable from the early postoperative period. After open surgical repair, it seems acceptable to have 'stepwise cut-off value' depending on the time from surgery.

Gradient boosting decision-tree-based algorithm with neuroimaging for personalized treatment in depression

Introduction

Pretreatment positron emission tomography (PET) with 2-deoxy-2-[18F]fluoro-D-glucose (FDG) and magnetic resonance spectroscopy (MRS) may identify biomarkers for predicting remission (absence of depression). Yet, no such image-based biomarkers have achieved clinical validity. The purpose of this study was to identify biomarkers of remission using machine learning (ML) with pretreatment FDG-PET/MRS neuroimaging, to reduce patient suffering and economic burden from ineffective trials.

Methods

This study used simultaneous PET/MRS neuroimaging from a double-blind, placebo-controlled, randomized antidepressant trial on 60 participants with major depressive disorder (MDD) before initiating treatment. After eight weeks of treatment, those with ≤ 7 on 17-item Hamilton Depression Rating Scale were designated a priori as remitters (free of depression, 37%). Metabolic rate of glucose uptake (metabolism) from 22 brain regions were acquired from PET. Concentrations (mM) of glutamine and glutamate and gamma-aminobutyric acid (GABA) in anterior cingulate cortex were quantified from MRS. The data were randomly split into 67% train and cross-validation (n = 40), and 33% test (n = 20) sets. The imaging features, along with age, sex, handedness, and treatment assignment (selective serotonin reuptake inhibitor or SSRI vs. placebo) were entered into the eXtreme Gradient Boosting (XGBoost) classifier for training.

Results

In test data, the model showed 62% sensitivity, 92% specificity, and 77% weighted accuracy. Pretreatment metabolism of left hippocampus from PET was the most predictive of remission.

Conclusions

The pretreatment neuroimaging takes around 60 minutes but has potential to prevent weeks of failed treatment trials. This study effectively addresses common issues for neuroimaging analysis, such as small sample size, high dimensionality, and class imbalance.

Automated differential diagnosis of dementia syndromes using FDG PET and machine learning

BACKGROUND

Metabolic brain imaging with 2-[¹⁸F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG PET) is a supportive diagnostic and differential diagnostic tool for neurodegenerative dementias. In the clinic, scans are usually visually interpreted. However, computer-aided approaches can improve diagnostic accuracy. We aimed to build two machine learning classifiers, based on two sets of FDG PET-derived features, for differential diagnosis of common dementia syndromes.

METHODS

We analyzed FDG PET scans from three dementia cohorts [63 dementia due to Alzheimer's disease (AD), 79 dementia with Lewy bodies (DLB) and 23 frontotemporal dementia (FTD)], and 41 normal controls (NCs). Patients' clinical diagnosis at follow-up (25 ± 20 months after scanning) or cerebrospinal fluid biomarkers for Alzheimer's disease was considered a gold standard. FDG PET scans were first visually evaluated. Scans were pre-processed, and two sets of features extracted: (1) the expressions of previously identified metabolic brain patterns, and (2) the mean uptake value in 95 regions of interest (ROIs). Two multi-class support vector machine (SVM) classifiers were tested and their diagnostic performance assessed and compared to visual reading. Class-specific regional feature importance was assessed with Shapley Additive Explanations.

RESULTS

Pattern- and ROI-based classifier achieved higher overall accuracy than expert readers (78% and 80% respectively, vs. 71%). Both SVM classifiers performed similarly to one another and to expert readers in AD (F1 = 0.74, 0.78, and 0.78) and DLB (F1 = 0.81, 0.81, and 0.78). SVM classifiers outperformed expert readers in FTD (F1 = 0.87, 0.83, and 0.63), but not in NC (F1 = 0.71, 0.75, and 0.92). Visualization of the SVM model showed bilateral temporal cortices and cerebellum to be the most important features for AD; occipital cortices, hippocampi and parahippocampi, amygdala, and middle temporal lobes for DLB; bilateral frontal cortices, middle and anterior cingulum for FTD; and bilateral angular gyri, pons, and vermis for NC.

CONCLUSION

Multi-class SVM classifiers based on the expression of characteristic metabolic brain patterns or ROI glucose uptake, performed better than experts in the differential diagnosis of common dementias using FDG PET scans. Experts performed better in the recognition of normal scans and a combined approach may yield optimal results in the clinical setting.

Brain (18)F-FDG PET imaging in outpatients with post-COVID-19 conditions: findings and associations with clinical characteristics

BACKGROUND

Brain ¹⁸F-FDG PET imaging has the potential to provide an objective assessment of brain involvement in post-COVID-19 conditions but previous studies of heterogeneous patient series yield inconsistent results. The current study aimed to investigate brain ¹⁸F-FDG PET findings in a homogeneous series of outpatients with post-COVID-19 conditions and to identify associations with clinical patient characteristics.

METHODS

We retrospectively included 28 consecutive outpatients who presented with post-COVID-19 conditions between September 2020 and May 2022 and who satisfied the WHO definition, and had a brain ¹⁸F-FDG PET for suspected brain involvement but had not been hospitalized for COVID-19. A voxel-based group comparison with 28 age- and sex-matched healthy controls was performed (p-voxel at 0.005 uncorrected, p-cluster at 0.05 FWE corrected) and identified clusters were correlated with clinical characteristics.

RESULTS

Outpatients with post-COVID-19 conditions exhibited diffuse hypometabolism predominantly involving right frontal and temporal lobes including the orbito-frontal cortex and internal temporal areas. Metabolism in these clusters was inversely correlated with the number of symptoms during the initial infection (r =  - 0.44, p = 0.02) and with the duration of symptoms (r =  - 0.39, p = 0.04). Asthenia and cardiovascular, digestive, and neurological disorders during the acute phase and asthenia and language disorders during the chronic phase (p ≤ 0.04) were associated with these hypometabolic clusters.

CONCLUSION

Outpatients with post-COVID-19 conditions exhibited extensive hypometabolic right fronto-temporal clusters. Patients with more numerous symptoms during the initial phase and with a longer duration of symptoms were at higher risk of persistent brain involvement.

Prospective Evaluation of Different Methods for Volumetric Analysis on [18F]FDG PET/CT in Pediatric Hodgkin Lymphoma

Rationale: Therapy response evaluation by 18F-fluorodeoxyglucose PET/CT (FDG PET) has become a powerful tool for the discrimination of responders from non-responders in pediatric Hodgkin lymphoma (HL). Recently, volumetric analyses have been regarded as a valuable tool for disease prognostication and biological characterization in cancer. Given the multitude of methods available for volumetric analysis in HL, the AIEOP Hodgkin Lymphoma Study Group has designed a prospective analysis of the Italian cohort enrolled in the EuroNet-PHL-C2 trial. Methods: Primarily, the study aimed to compare the different segmentation techniques used for volumetric assessment in HL patients at baseline (PET1) and during therapy: early (PET2) and late assessment (PET3). Overall, 50 patients and 150 scans were investigated for the current analysis. A dedicated software was used to semi-automatically delineate contours of the lesions by using different threshold methods. More specifically, four methods were applied: (1) fixed 41% threshold of the maximum standardized uptake value (SUVmax) within the respective lymphoma site (V41%), (2) fixed absolute SUV threshold of 2.5 (V2.5); (3) SUVmax(lesion)/SUVmean liver >1.5 (Vliver); (4) adaptive method (AM). All parameters obtained from the different methods were analyzed with respect to response. Results: Among the different methods investigated, the strongest correlation was observed between AM and Vliver (rho > 0.9; p < 0.001 for SUVmean, MTV and TLG at all scan timing), along with V2.5 and AM or Vliver (rho 0.98, p < 0.001 for TLG at baseline; rho > 0.9; p < 0.001 for SUVmean, MTV and TLG at PET2 and PET3, respectively). To determine the best segmentation method, we applied logistic regression and correlated different results with Deauville scores at late evaluation. Logistic regression demonstrated that MTV (metabolic tumor volume) and TLG (total lesion glycolysis) computation according to V2.5 and Vliver significantly correlated to response to treatment (p = 0.01 and 0.04 for MTV and 0.03 and 0.04 for TLG, respectively). SUVmean also resulted in significant correlation as absolute value or variation. Conclusions: The best correlation for volumetric analysis was documented for AM and Vliver, followed by V2.5. The volumetric analyses obtained from V2.5 and Vliver significantly correlated to response to therapy, proving to be preferred thresholds in our pediatric HL cohort.

Metabolic connectivity is associated with seizure outcome in surgically treated temporal lobe epilepsies: A 18F-FDG PET seed correlation analysis

¹⁸F-FDG PET provides high sensitivity for the pre-surgical assessment of drug-resistant temporal lobe epilepsy (TLE). However, little is known about the metabolic connectivity of epileptogenic networks involved. This study therefore aimed to evaluate the association between metabolic connectivity and seizure outcome in surgically treated TLE.

METHODS

The study included 107 right-handed patients that had undergone a presurgical interictal ¹⁸F-FDG PET assessment followed by an anterior temporal lobectomy and were classified according to seizure outcome 2 years after surgery. Metabolic connectivity was evaluated by seed correlation analysis in left and right epilepsy patients with a Class Engel IA or > IA outcome and compared to age-, sex- and handedness-matched healthy controls.

RESULTS

Increased metabolic connectivity was observed in the >IA compared to the IA group within the operated temporal lobe (respective clusters of 7.5 vs 3.3 cm³ and 2.6 cm³ vs 2.2 cm³ in left and right TLE), and to a lower extent with the contralateral temporal lobe (1.2 vs 0.7 cm³ and 1.7 cm³ vs 0.7 cm³ in left and right TLE). Seed correlations provided added value for the estimated individual performance of seizure outcome over the group comparisons in left TLE (AUC of 0.74 vs 0.67).

CONCLUSION

Metabolic connectivity is associated with outcome in surgically treated TLE with a strengthened epileptogenic connectome in patients with non-free-seizure outcomes. The added value of seed correlation analysis in left TLE underlines the importance of evaluating metabolic connectivity in network related diseases.

Glucose Metabolism Modification Induced by Radioligand Therapy with [(177)Lu]Lu/[(90)Y]Y-DOTATOC in Advanced Neuroendocrine Neoplasms: A Prospective Pilot Study within FENET-2016 Trial

[18F]F-FDG (FDG) PET is emerging as a relevant diagnostic and prognostic tool in neuroendocrine neoplasms (NENs), as a simultaneous decrease in [68Ga]Ga-DOTA peptides and increase in FDG uptake (the “flip-flop” phenomenon) occurs during the natural history of these tumors. The aim of this study was to evaluate the variations on FDG PET in NEN patients treated with two different schemes of radioligand therapy (RLT) and to correlate them with clinical−pathologic variables. A prospective evaluation of 108 lesions in 56 patients (33 males and 23 females; median age, 64.5 years) affected by NENs of various primary origins (28 pancreatic, 13 gastrointestinal, 9 bronchial, 6 unknown primary (CUP-NENs) and 1 pheochromocytoma) and grades (median Ki-67 = 9%) was performed. The patients were treated with RLT within the phase II clinical trial FENET-2016 (CTID: NCT04790708). RLT was offered for 32 patients with the MONO scheme (five cycles of [177Lu]Lu-DOTATOC) and for 24 with the DUO scheme (three cycles of [177Lu]Lu-DOTATOC alternated with two cycles of [90Y]Y-DOTATOC). Variations in terms of the ΔSUVmax of a maximum of three target lesions per patient (58 for MONO and 50 for DUO RLT) were assessed between baseline and 3 months post-RLT FDG PET. In patients with negative baseline FDG PET, the three most relevant lesions on [68Ga]Ga-DOTA-peptide PET were assessed and matched on post-RLT FDG PET, to check for any possible changes in FDG avidity. Thirty-five patients (62.5%) had at least one pathological FDG uptake at the baseline scans, but the number was reduced to 29 (52%) after RLT. In the patients treated with DUO-scheme RLT, 20 out of 50 lesions were FDG positive before therapy, whereas only 14 were confirmed after RLT (p = 0.03). Moreover, none of the 30 FDG-negative lesions showed an increased FDG uptake after RLT. The lesions of patients with pancreatic and CUP-NENs treated with the DUO scheme demonstrated a significant reduction in ΔSUVmax in comparison to those treated with MONO RLT (p = 0.03 and p = 0.04, respectively). Moreover, we found a mild positive correlation between the grading and ΔSUVmax in patients treated with the MONO scheme (r = 0.39, p < 0.02), while no evidence was detected for patients treated with the DUO scheme. Our results suggest that RLT, mostly with the DUO scheme, could be effective in changing NEN lesions’ glycometabolism, in particular, in patients affected by pancreatic and CUP-NENs, regardless of their Ki-67 index. Probably, associating [90Y]Y-labelled peptides, which have high energy emission and a crossfire effect, and [177Lu]Lu ones, characterized by a longer half-life and a safer profile for organs at risk, might represent a valid option in FDG-positive NENs addressed to RLT. Further studies are needed to validate our preliminary findings. In our opinion, FDG PET/CT should represent a potent tool for fully assessing a patient’s disease characteristics, both before and after RLT.

Monitoring Alzheimer's Disease Progression in Mild Cognitive Impairment Stage Using Machine Learning-Based FDG-PET Classification Methods

BACKGROUND

We previously introduced a machine learning-based Alzheimer's Disease Designation (MAD) framework for identifying AD-related metabolic patterns among neurodegenerative subjects.

OBJECTIVE

We sought to assess the efficiency of our MAD framework for tracing the longitudinal brain metabolic changes in the prodromal stage of AD.

METHODS

MAD produces subject scores using five different machine-learning algorithms, which include a general linear model (GLM), two different approaches of scaled subprofile modeling, and two different approaches of a support vector machine. We used our pre-trained MAD framework, which was trained based on metabolic brain features of 94 patients with AD and 111 age-matched cognitively healthy (CH) individuals. The MAD framework was applied on longitudinal independent test sets including 54 CHs, 51 stable mild cognitive impairment (sMCI), and 39 prodromal AD (pAD) patients at the time of the clinical diagnosis of AD, and two years prior.

RESULTS

The GLM showed excellent performance with area under curve (AUC) of 0.96 in distinguishing sMCI from pAD patients at two years prior to the time of the clinical diagnosis of AD while other methods showed moderate performance (AUC: 0.7-0.8). Significant annual increment of MAD scores were identified using all five algorithms in pAD especially when it got closer to the time of diagnosis (p <  0.001), but not in sMCI. The increased MAD scores were also significantly associated with cognitive decline measured by Mini-Mental State Examination in pAD (q <  0.01).

CONCLUSION

These results suggest that MAD may be a relevant tool for monitoring disease progression in the prodromal stage of AD.

Statistical asymmetry analysis of volumetric MRI and FDG PET in temporal lobe epilepsy

PURPOSE

To analyze statistically derived threshold values of volumetric MRI and 18F fluorodeoxyglucose (FDG) PET asymmetry, independent of normative data, for non-invasive detection/exclusion of temporal lobe epilepsy (TLE).

METHODS

We retrospectively analyzed amygdalohippocampal volumetry and temporal FDG PET metabolism in 33 patients (age: 29.27 ± 8.56 years) who underwent surgery following Stereo-EEG implantation and had postsurgical seizure freedom lasting >12 months. The temporal lobe epilepsy group and the extratemporal lobe epilepsy (ETLE) group were compared. Percentage volume loss (PVL) was calculated from manually traced amygdalohippocampal volumetry whereas percentage metabolic loss (PML) was calculated from PET using amygdalohippocampal trace and temporal neocortical Brodmann areas (BA) template.

RESULTS

Receiver operating characteristic (ROC) curve analysis identified a cutoff hippocampal PVL of 4.21% as the minimum indicating probable hippocampal involvement in seizure onset, with sensitivity of 88.89% and the specificity of 100% (p < 0.001). Region of interest (ROI)-based PML values in PET imaging showed a significant correlation with the presence of TLE in the TLE group of patients and its absence in the ETLE group of patients. Region of interest curve analysis yielded PML cutoffs of 5.77% and 8.36%, respectively, for the hippocampus and BA 38 (temporopolar neocortex) to detect TLE with the sensitivity of 72.7% and specificity of 77.8%.

CONCLUSION

We describe statistical thresholds for asymmetry analysis of hippocampal volumetry and FDG PET to improve detection of TLE. These threshold parameters warrant further validation in prospective studies.

PET Image Denoising using a Deep-Learning Method for Extremely Obese Patients

The image quality in clinical PET scan can be severely degraded due to high noise levels in extremely obese patients. Our work aimed to reduce the noise in clinical PET images of extremely obese subjects to the noise level of lean subject images, to ensure consistent imaging quality. The noise level was measured by normalized standard deviation (NSTD) derived from a liver region of interest. A deep learning-based noise reduction method with a fully 3D patch-based U-Net was used. Two U-Nets, U-Nets A and B, were trained on datasets with 40% and 10% count levels derived from 100 lean subjects, respectively. The clinical PET images of 10 extremely obese subjects were denoised using the two U-Nets. The results showed the noise levels of the images with 40% counts of lean subjects were consistent with those of the extremely obese subjects. U-Net A effectively reduced the noise in the images of the extremely obese patients while preserving the fine structures. The liver NSTD improved from 0.13±0.04 to 0.08±0.03 after noise reduction (p = 0.01). After denoising, the image noise level of extremely obese subjects was similar to that of lean subjects, in terms of liver NSTD (0.08±0.03 vs. 0.08±0.02, p = 0.74). In contrast, U-Net B over-smoothed the images of extremely obese patients, resulting in blurred fine structures. In a pilot reader study comparing extremely obese patients without and with U-Net A, the difference was not significant. In conclusion, the U-Net trained by datasets from lean subjects with matched count level can provide promising denoising performance for extremely obese subjects while maintaining image resolution, though further clinical evaluation is needed.