Classification of Aß State From Brain Amyloid PET Images Using Machine Learning Algorithm.

Background and Purpose: Analyzing brain amyloid positron emission tomography (PET) images to access the occurrence of ß-amyloid (Aß) deposition in Alzheimer's patients requires much time and effort from physicians, while the variation of each interpreter may differ. For these reasons, a machine learning model was developed using a convolutional neural network (CNN) as an objective decision to classify the Aß positive and Aß negative status from brain amyloid PET images. Methods: A total of 7,344 PET images of 144 subjects were used in this study. The 18F-florbetaben PET was administered to all participants, and the criteria for differentiating Aß positive and Aß negative state was based on brain amyloid plaque load score (BAPL) that depended on the visual assessment of PET images by the physicians. We applied the CNN algorithm trained in batches of 51 PET images per subject directory from 2 classes: Aß positive and Aß negative states, based on the BAPL scores. Results: The binary classification of the model average performance matrices was evaluated after 40 epochs of three trials based on test datasets. The model accuracy for classifying Aß positivity and Aß negativity was (95.00±0.02) in the test dataset. The sensitivity and specificity were (96.00±0.02) and (94.00±0.02), respectively, with an area under the curve of (87.00±0.03). Conclusions: Based on this study, the designed CNN model has the potential to be used clinically to screen amyloid PET images.

Impact of Metabolic Activity of Vertebra and Amygdala on Stroke Recurrence: A Prospective Cohort Study.

BACKGROUND: Elevated metabolic activity of amygdala is known to be related to atherosclerotic cardiovascular event by increasing inflammatory cell production from bone marrow. We tried to identify the factors of metabolic activity in the amygdala, vertebrae, liver, spleen, and internal carotid artery related to the future vascular events after stroke. METHODS: A total of 110 patients with acute stroke were included (72±10 years of age, 39% women) and underwent whole-body 18F-fluorodeoxyglucose (FDG) positron emission tomography between August 1, 2015 and February 28, 2020. We compared the FDG uptake in the amygdala, vertebrae, liver, spleen, and internal carotid artery between patients with and without recurrent vascular event. Cox proportional hazards model was used to identify factors related to recurrent stroke and vascular event. RESULTS: During the median follow-up period of 18 months, 22 patients experienced vascular events, including 15 stroke recurrence. Patients with recurred vascular event had a significantly higher FDG uptake in the amygdala and vertebrae than those without. The Cox proportional hazard model including diabetes, renal function, and carotid stenosis showed that a higher FDG uptake in the amygdala was independently associated with total vascular events (hazard ratio, 3.11 [95% CI, 1.11-8.70]) and higher FDG uptake in the vertebrae with stroke recurrence (hazard ratio, 4.94 [95% CI, 1.29-18.9]). CONCLUSIONS: The increased metabolic activities of the vertebrae and amygdala are related to future vascular event among stroke survivors.

The uptake pattern of 18F-sodium fluoride radioligand in brain tissue after cerebral infarction.

Positron emission tomography with 18F-sodium fluoride (NaF) radioligand has been actively investigated in atherosclerosis research because it is known to detect microcalcification activity within atheroma. We studied whether NaF shows any uptake in the brain tissue of patients with acute ischemic stroke. This is a post-hoc analysis of previously reported cerebral atherosclerosis research with positron emission tomography which applied the two radioligands, 18F-fluorodeoxyglucose and NaF for the detection of culprit atheroma among 20 acute cerebral infarction patients (mean age = 75.1 ± 9.0 years; 10 women). In this study, we measured the maximum and mean standardized uptake value (SUVmax and SUVmean) of NaF uptake level in the cerebral infarct region between lesions with and without diffusion weighted image (DWI) positivity, indicating acute ischemic cell death. Correlation analysis was performed between NaF uptake levels and imaging and clinical variables, including neurological severity. The NaF uptake levels were significantly higher in DWI positive lesions than in negative lesions (SUVmax: 2.0 [0.60-4.2] versus 0.20 [0.10-0.40], p = 0.021 by Mann-Whitney U test). The intensity of NaF uptake (SUVmax) was significantly correlated with the initial neurological severity (Spearman's ρ = 0.579, p= 0.007) and white blood cell count (Spearman's ρ = 0.626, p p 0.003). During ischemic stroke NaF was concentrated in brain tissue undergoing acute cell death and its uptake intensity was correlated with neurological severity, suggesting that NaF could reflect acute ischemic cell death after stroke.

Association of androgen receptor expression with glucose metabolic features in triple-negative breast cancer.

BACKGROUND: Androgen receptor (AR) is a potential therapeutic target in triple-negative breast cancer (TNBC). We aimed to elucidate the association of AR expression with glucose metabolic features in TNBC. METHODS: Two independent datasets were analyzed: FDG PET data of our institution and a public dataset of GSE135565. In PET analysis, patients with TNBC who underwent pretreatment PET between Jan 2013 and Dec 2017 were retrospectively enrolled. Clinicopathologic features and maximum standardized uptake value (SUVmax) of tumors were compared with AR expression. In GSE135565 dataset, glycolysis score was calculated by the pattern of glycolysis-related genes, and of which association with SUVmax and AR gene expression were analyzed. RESULTS: A total of 608 female patients were included in the PET data of our institution. SUVmax was lower in AR-positive tumors (P < 0.001) and correlated with lower AR expression (rho = -0.26, P < 0.001). In multivariate analysis, AR was a deterministic factor for low SUVmax (P = 0.012), along with other key clinicopathologic features. In the GSE135565 dataset, AR expression also exhibited a negative correlation with SUVmax (r = -0.34, P = 0.001) and the glycolysis score (r = -0.27, P = 0.013). CONCLUSIONS: Low glucose metabolism is a signature of AR expression in TNBC. It is suggested that evaluation of AR expression status needs to be considered in clinical practice particularly in TNBC with low glucose metabolism.

Aripiprazole Improved Post-Streptococcal Dyskinesia by Modulating Dopaminergic Activity: A Case Report.

This case report aimed to discuss the efficacy of aripiprazole for dyskinesia in patients with functional movement disorder after streptococcal infection, with its biological action of modulating dopamine hyperactivity in the basal ganglia as a dopamine partial agonist. This report has shown that the [18F] N -(3-Fluoropropyl)-2ß-carbon ethoxy-3ß-(4-iodophenyl) nortropane positron emission tomography findings of the patient revealed that the dopamine hyperactivity in the basal ganglia at baseline was normalized after aripiprazole treatment due to its balancing effect as a dopamine partial agonist.

Prediction of post-stroke cognitive impairment using brain FDG PET: deep learning-based approach.

PURPOSE: Post-stroke cognitive impairment can affect up to one third of stroke survivors. Since cognitive function greatly contributes to patients' quality of life, an objective quantitative biomarker for early prediction of dementia after stroke is required. We developed a deep-learning (DL)-based signature using positron emission tomography (PET) to objectively evaluate cognitive decline in patients with stroke. METHODS: We built a DL model that differentiated Alzheimer's disease (AD) from normal controls (NC) using brain fluorodeoxyglucose (FDG) PET from the Alzheimer's Disease Neuroimaging Initiative database. The model was directly transferred to a prospectively enrolled cohort of patients with stroke to differentiate patients with dementia from those without dementia. The accuracy of the model was evaluated by the area under the curve values of receiver operating characteristic curves (AUC-ROC). We visualized the distribution of DL-based features and brain regions that the model weighted for classification. Correlations between cognitive signature from the DL model and clinical variables were evaluated, and survival analysis for post-stroke dementia was performed in patients with stroke. RESULTS: The classification of AD vs. NC subjects was performed with AUC-ROC of 0.94 (95% confidence interval [CI], 0.89-0.98). The transferred model discriminated stroke patients with dementia (AUC-ROC = 0.75). The score of cognitive decline signature using FDG PET was positively correlated with age, neutrophil-lymphocyte ratio and platelet-lymphocyte ratio and negatively correlated with body mass index in patients with stroke. We found that the cognitive decline score was an independent risk factor for dementia following stroke (hazard ratio, 10.90; 95% CI, 3.59-33.09; P < 0.0001) after adjustment for other key variables. CONCLUSION: The DL-based cognitive signature using FDG PET was successfully transferred to an independent stroke cohort. It is suggested that DL-based cognitive evaluation using FDG PET could be utilized as an objective biomarker for cognitive dysfunction in patients with cerebrovascular diseases.

Comparison of endoscopically determined gross tumor volume and metabolic tumor volume in esophageal cancer.

ABSTRACT: The purpose of this study is to compare the longitudinal location of endoscopically-defined gross tumor volume (GTV) and positron emission tomography-based metabolic tumor volume (MTV) of esophageal cancer.A retrospective review of medical records was performed of the nine patients who underwent endoscopic placement of fiducial markers for radiotherapy of esophageal squamous cell carcinoma. Endoscopic hemoclips were used as the fiducial markers, and GTV was newly delineated solely based on the locations of the fiducial markers. The standardized uptake value (SUV) threshold corresponding to the superior and inferior borders of GTV was defined as the highest threshold that made MTV reach each border of GTV.The median fixed relative and absolute thresholds were 32% and 3.8, respectively. The coefficient of variation of the threshold values was 0.781 for the fixed relative threshold method and 0.400 for the fixed absolute threshold method, indicating more consistent results from the fixed absolute threshold method. All but two GTV borders were included in MTV with a SUV threshold of 2.5. Esophageal tumors with a maximum SUV > 20 tended to have closer threshold values corresponding to the GTV borders to 2.5 (median 2.8 vs 3.6, P = .069).The fixed absolute threshold method was suitable for determining the MTV threshold for esophageal lesions. A SUV of 2.5 was appropriate for esophageal tumors with a maximum SUV > 20. Endoscopic hemoclips were stable enough for using as the fiducial marker.

Variability of FP-CIT PET Patterns Associated With Clinical Features of Multiple System Atrophy.

OBJECTIVE: To validate the role of the dopamine transporter (DAT) imaging as a biomarker in multiple system atrophy (MSA), we analyzed the association between spatial patterns of [18F]fluoro-propyl-carbomethoxy-iodophenyl-tropane ([18F]FP-CIT) PET and the clinical characteristics of MSA. METHODS: Sixty-five patients with MSA who underwent [18F]FP-CIT PET between 2009 and 2018 were retrospectively enrolled. To identify spatial patterns of [18F]FP-CIT PET, principal component (PC) analysis was used and correlated with the clinical presentation. RESULTS: Of the 65 patients, 42 presented with parkinsonian subtype of MSA, and 23 presented with cerebellar subtype of MSA (mean age 63.7 ± 9.3 years; disease duration, 1.8 ± 1.8 years). Each PC represents a specific pattern of degeneration: PC1 and PC2 were associated with the DAT binding of the entire putamen and the posterior putamen, respectively. PC3 was associated with increased [18F]FP-CIT uptake of the caudate and decreased uptake of the dorsal pons. PC2 was significantly correlated with the presence of parkinsonism (p = 5.34 × 10-5) and a positive levodopa response (p = 0.044), with age as a cofactor. PC3 was correlated with the presence of urinary incontinence (p = 0.036). Onset age was significantly correlated with both PC2 (R = 0.48, p = 5.0 × 10-5) and PC3 (R = -0.39, p = 0.0013). CONCLUSIONS: The spatial pattern of DAT binding can reflect distinct clinical features of MSA and provides insight into the underlying pathophysiology of a broad spectrum of clinical features in MSA.

An Update on [18F]Fluoride PET Imaging for Atherosclerotic Disease.

Atherosclerosis is the leading cause of life-threatening morbidity and mortality, as the rupture of atherosclerotic plaques leads to critical atherothrombotic events such as myocardial infarction and ischemic stroke, which are the 2 most common causes of death worldwide. Vascular calcification is a complicated pathological process involved in atherosclerosis, and microcalcifications are presumed to increase the likelihood of plaque rupture. Despite many efforts to develop novel non-invasive diagnostic modalities, diagnostic techniques are still limited, especially before symptomatic presentation. From this point of view, vulnerable plaques are a direct target of atherosclerosis imaging. Anatomic imaging modalities have the limitation of only visualizing macroscopic structural changes, which occurs in later stages of disease, while molecular imaging modalities are able to detect microscopic processes and microcalcifications, which occur early in the disease process. Na[18F]-fluoride positron emission tomography/computed tomography could allow the early detection of plaque instability, which is deemed to be a primary goal in the prevention of cardiac or brain ischemic events, by quantifying the microcalcifications within vulnerable plaques and evaluating the atherosclerotic disease burden.

Evaluation of Hot Nodules of Thyroid Gland Using Tc-99m Pertechnetate: a Novel Approach Using Quantitative Single-Photon Emission Computed Tomography/Computed Tomography.

Planar scintigraphy using Tc-99m pertechnetate is useful for snapshot evaluation of hot thyroid nodules, which are pathologically follicular adenoma and seldom, if ever, malignant. The autonomy of the hot nodules has been demonstrated by the presence of thyroid-stimulating hormone-dependent extra-nodular thyroid tissue besides the hot nodules. Here, we present two cases of hot thyroid nodules in patients who underwent quantitative single-photon emission computed tomography/computed tomography (SPECT/CT). In addition to the nodules, contralateral normal thyroid parenchyma was evaluated based on standardized uptake values. One patient had a traditional follicular adenoma suppressing other thyroid tissue, whereas the other patient seemed to have a nodule erupting from underlying hyperfunctioning, not suppressed, thyroid tissue. This novel approach using quantitative SPECT/CT unveils a new pathology of hot thyroid nodule that does not suppress, but coincides with hyperfunctioning thyroid tissue.

Application of Quantitative Indexes of FDG PET to Treatment Response Evaluation in Indolent Lymphoma.

PURPOSE: Although 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) is a standard imaging modality for response evaluation in FDG-avid lymphoma, there is a controversy using FDG PET in indolent lymphoma. The purpose of this study was to investigate the effectiveness of quantitative indexes on FDG PET in response evaluation of the indolent lymphoma. METHODS: Fifty-seven indolent lymphoma patients who completed chemotherapy were retrospectively enrolled. FDG PET/computed tomography (CT) scans were performed at baseline, interim, and end of treatment (EOT). Response was determined by Lugano classification, and progression-free survival (PFS) by follow-up data. Maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were measured in the single hottest lesion (target A) or five hottest lesions (target B). Their efficacies regarding response evaluation and PFS prediction were evaluated. RESULTS: On EOT PET, SUVmax, and MTV of both targets were well associated with visual analysis. Changes between initial and EOT PET were not significantly different between CR and non-CR groups. On interim PET, SUVmax, and %ΔSUVmax in both targets were significantly different between CR and non-CR groups. For prediction of PFS, most tested indexes were significant on EOT and interim PET, with SUVmax being the most significant prognostic factor. CONCLUSION: Quantitative indexes of FDG PET are well associated with Lugano classification in indolent lymphoma. SUVmax measured in the single hottest lesion can be effective in response evaluation and prognosis prediction on interim and EOT PET.

Measurement of 68Ga-DOTATOC Uptake in the Thoracic Aorta and Its Correlation with Cardiovascular Risk.

PURPOSE: 68Ga-labeled 1,4,7,10-tetraazacyclododecane-N,N',N″,N‴-tetraacetic acid-d-Phe1-Tyr3-octreotide (68Ga-DOTATOC) is taken up by activated macrophages, which accumulate in active inflammatory lesions. The purpose of this study was to investigate the feasibility of 68Ga-DOTATOC PET/CT for assessment of vulnerable plaque, by evaluating correlation between aortic uptake of 68Ga-DOTATOC and cardiovascular risk factors. METHODS: Fifty patients with neuroendocrine tumors who underwent 68Ga-DOTATOC PET/CT were retrospectively enrolled. The uptakes in the thoracic aorta were measured by two methods: multi-sample region-of-interest (ROI) method and single volume-of-interest (VOI) method. TBRmax-avg, TBRmean-avg, TBRmax-VOI, and TBRmean-VOI were defined by maximum and mean target-to-background ratio (TBR) from the multi-sample ROI method and the single VOI method, respectively. RESULTS: Framingham risk score (FRS) exhibited significant correlations with TBRmax-avg and TBRmean-avg, as well as TBRmax-VOI (r = 0.3389-0.4593, P < 0.05 for all). TBRmax-avg and TBRmax-VOI were significantly higher in high FRS group than in low FRS group (1.48 ± 0.21 vs. 1.70 ± 0.17, P < 0.001 for TBRmax-avg and 1.90 ± 0.33 vs. 2.25 ± 0.36, P = 0.002 for TBRmax-VOI). TBR exhibited high correlations between the two measuring methods (r = 0.9684, P < 0.001 for TBRmean-avg and TBRmean-VOI and r = 0.8681, P < 0.001 for TBRmax-avg and TBRmax-VOI). CONCLUSIONS: 68Ga-DOTATOC uptake in the thoracic aorta exhibited a significant correlation with cardiovascular risk factors, which suggests the feasibility of 68Ga-DOTATOC PET for vulnerable plaque imaging, with a simple measurement of the single VOI method that is comparable to the multi-sample ROI-based approach.