Dataset Information

Gaussian Mixture Models and Model Selection for [18F] Fluorodeoxyglucose Positron Emission Tomography Classification in Alzheimer's Disease.

ABSTRACT: We present a method to discover discriminative brain metabolism patterns in [18F] fluorodeoxyglucose positron emission tomography (PET) scans, facilitating the clinical diagnosis of Alzheimer's disease. In the work, the term "pattern" stands for a certain brain region that characterizes a target group of patients and can be used for a classification as well as interpretation purposes. Thus, it can be understood as a so-called "region of interest (ROI)". In the literature, an ROI is often found by a given brain atlas that defines a number of brain regions, which corresponds to an anatomical approach. The present work introduces a semi-data-driven approach that is based on learning the characteristics of the given data, given some prior anatomical knowledge. A Gaussian Mixture Model (GMM) and model selection are combined to return a clustering of voxels that may serve for the definition of ROIs. Experiments on both an in-house dataset and data of the Alzheimer's Disease Neuroimaging Initiative (ADNI) suggest that the proposed approach arrives at a better diagnosis than a merely anatomical approach or conventional statistical hypothesis testing.

SUBMITTER: Li R

PROVIDER: S-EPMC4412726 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Gaussian Mixture Models and Model Selection for [18F] Fluorodeoxyglucose Positron Emission Tomography Classification in Alzheimer's Disease.

Li Rui R Perneczky Robert R Yakushev Igor I Förster Stefan S Kurz Alexander A Drzezga Alexander A Kramer Stefan S

PloS one 20140101 4

We present a method to discover discriminative brain metabolism patterns in [18F] fluorodeoxyglucose positron emission tomography (PET) scans, facilitating the clinical diagnosis of Alzheimer's disease. In the work, the term "pattern" stands for a certain brain region that characterizes a target group of patients and can be used for a classification as well as interpretation purposes. Thus, it can be understood as a so-called "region of interest (ROI)". In the literature, an ROI is often found b ...[more]

PMID: 25919662

Similar Datasets

Project description:Positron emission tomography (PET) visualizes increased cellular [F]fluorodeoxyglucose ([F]FDG) uptake. Pulmonary hypertension (PH) is conceived of a proliferative disease of the lung vessels. Increased glucose uptake can be quantified as pulmonary [F]FDG uptake via PET imaging. Because the angioproliferative mechanisms in PH are still in need of further description, the aim of the present study was to investigate whether [F]FDG PET/CT imaging can elucidate these pathophysiologic mechanisms in different etiologies of PH.Patients (n?=?109) with end-stage pulmonary disease being evaluated for lung transplant were included in this observational study. Mean standardized uptake value (SUVmean) of predefined regions of interest in lung parenchyma (LP), left (LV), and right ventricle (RV) of the heart, and SUVmax in pulmonary artery (PA) were determined and normalized to liver uptake. These SUV ratios (SUVRs) were compared with results from right heart catheterization (mean pulmonary artery pressure [mPAP], pulmonary vascular resistance [PVR]), and serum N-terminal pro-brain natriuretic peptide. Group comparisons were performed and Pearson correlation coefficients (r) were calculated.The [F]FDG uptake ratios in LP, RV, RV/LV, and PA, but not in LV, were found to be significantly higher in both patients with mPAP ?25?mm Hg (P?=?0.013, P?=?0.006, P?=?0.049, P?=?0.002, P?=?0.68, respectively) and with PVR ?480?dyn·s/cm (P?<?0.001, P?=?0.045, P?<?0.001, P?<?0.001, P?=?0.26, respectively). The [F]FDG uptake in these regions positively correlated also with mPAP, PVR, and N-terminal pro-brain natriuretic peptide. The SUVR of PA positively correlated with the SUVR of LP and RV (r?=?0.55, r?=?0.42, respectively).Pulmonary and cardiac [F]FDG uptake in PET imaging positively correlated with the presence and severity of PH in patients with end-stage pulmonary disease. Increased glucose metabolism in the central PAs seems to play a certain role in terms of severity of PH. These results suggest that [F]FDG-PET imaging can help understand the pathophysiology of PH as a proliferative pulmonary disease.

Project description:COPD and smoking are characterised by pulmonary inflammation. 18F-fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) imaging may improve knowledge of pulmonary inflammation in COPD patients and aid early development of novel therapies as an imaging biomarker. To evaluate pulmonary inflammation, assessed by FDG uptake, in whole and regional lung in "usual" (smoking-related) COPD patients, alpha-1 antitrypsin deficiency (α1ATD) COPD patients, smokers without COPD and never-smokers using FDG PET/CT. Secondly, to explore cross-sectional associations between FDG PET/CT and systemic inflammatory markers in COPD patients and repeatability of the technique in COPD patients. Data from two imaging studies were evaluated. Pulmonary FDG uptake (normalised Ki; nKi) was measured by Patlak graphical analysis in four subject groups: 84 COPD patients, 11 α1ATD-COPD patients, 12 smokers and 10 never-smokers. Within the COPD group, associations between nKi and systemic markers of inflammation were assessed. Repeatability was evaluated in 32 COPD patients comparing nKi values at baseline and at 4-month follow-up. COPD patients, α1ATD-COPD patients and smokers had increased whole lung FDG uptake (nKi) compared with never-smokers (0.0037±0.001, 0.0040±0.001, 0.0040±0.001 versus 0.0028±0.001 mL·cm-3·min-1, respectively, p<0.05 for all). Similar results were observed in upper and middle lung regions. In COPD participants, plasma fibrinogen was associated with whole lung nKi (β=0.30, p=0.02) in multivariate analysis adjusted for current smoking, forced expiratory volume in 1 s % predicted, systemic neutrophils and C-reactive protein levels. Mean percentage difference in nKi between the baseline and follow-up was 3.2%, and the within subject coefficient of variability was 7.7%. FDG PET/CT has potential as a noninvasive tool to enable whole lung and regional quantification of FDG uptake to assess smoking- and COPD-related pulmonary inflammation.

Project description:Fluorine-18-fluoro-2-deoxy-D-glucose (FDG) is widely used as positron-emission-tomography (PET) radiotracer for the detection and staging of human cancer. Tumor uptake of FDG varies substantially between different cancer types and between patients with the same tumor type. The molecular basis for this heterogeneity is unknown. Using cancer cell lines and primary human tumors of distinct histologic origins, we here show that increased FDG uptake is universally associated with coordinate upregulation of genes within the glycolysis, pentose-phosphate, and other related metabolic pathways. In primary human breast cancers, this FDG signature shows significant overlap with established breast cancer signatures for the “basal-like” disease subtype and “poor prognosis”. FDG high breast cancer showed significantly more gene copy number alterations genome wide than FDG low cancers. About 50 % of primary breast cancers with high FDG uptake and FDG gene expression signature show DNA copy gain encompassing the c-myc gene locus and express gene sets regulated by the transcription factor MYC. Our data shows that FDG-PET marks a distinct subset of “basal-like” human breast cancer which is characterized by MYC and prognostically unfavorable gene expression signatures, suggesting that FDG-PET imaging may be useful to risk-stratify patients with locally advanced breast cancer. Our general strategy to identify molecular determinants of FDG-retention through a genome-wide approach consisted of FDG uptake measurements in cancer cell lines and primary human tumors, the selection of samples with particularly high versus low FDG-retention, and subsequent pathway-based analysis of RNA expression data collected from these samples. Cell line RNA was extracted from a 10 cm plate seeded simultaneously and at the same density as the wells for FDG-uptake assays. For primary human tumor samples, RNA was extracted from macrodissected frozen tumor tissue. All cell line RNA samples and the astrocytoma RNA aliquots were hybridized to Affymetrix U133 Plus 2.0 arrays. All primary breast cancer RNA samples were hybridized to Affymetrix U133A arrays.

Project description:BackgroundPheochromocytomas and paragangliomas (PPGLs) are rare tumors of the adrenal medulla and extra-adrenal sympathetic chromaffin tissues; their anatomical and functional imaging are critical to guiding treatment decisions. This study aimed to compare the sensitivity and specificity of (18)F-fluorodeoxyglucose positron emission tomography with computed tomography ((18)F-FDG PET/CT) for tumor localization and staging of PPGLs with that of conventional imaging by [(123)I]-metaiodobenzylguanidine single photon emission CT ((123)I-MIBG SPECT), CT, and magnetic resonance imaging (MRI).MethodsA total of 216 patients (106 men, 110 women, aged 45.2 ± 14.9 years) with suspected PPGL underwent CT or MRI, (18)F-FDG PET/CT, and (123)I-MIBG SPECT/CT. Sensitivity and specificity were measured as endpoints and compared by the McNemar test, using two-sided P values only.ResultsSixty (28%) of patients had nonmetastatic PPGL, 95 (44%) had metastatic PPGL, and 61 (28%) were PPGL negative. For nonmetastatic tumors, the sensitivity of (18)F-FDG was similar to that of (123)I-MIBG but less than that of CT/MRI (sensitivity of (18)F-FDG = 76.8%; of (123)I-MIBG = 75.0%; of CT/MRI = 95.7%; (18)F-FDG vs (123)I-MIBG: difference = 1.8%, 95% confidence interval [CI] = -14.8% to 14.8%, P = .210; (18)F-FDG vs CT/MRI: difference = 18.9%, 95% CI = 9.4% to 28.3%, P < .001). The specificity was 90.2% for (18)F-FDG, 91.8% for (123)I-MIBG, and 90.2% for CT/MRI. (18)F-FDG uptake was higher in succinate dehydrogenase complex- and von Hippel-Lindau syndrome-related tumors than in multiple endocrine neoplasia type 2 (MEN2) related tumors. For metastases, sensitivity was greater for (18)F-FDG and CT/MRI than for (123)I-MIBG (sensitivity of (18)F-FDG = 82.5%; of (123)I-MIBG = 50.0%; of CT/MRI = 74.4%; (18)F-FDG vs (123)I-MIBG: difference = 32.5%, 95% CI = 22.3% to 42.5%, P < .001; CT/MRI vs (123)I-MIBG: difference = 24.4%, 95% CI = 11.3% to 31.6%, P < .001). For bone metastases, (18)F-FDG was more sensitive than CT/MRI (sensitivity of (18)F-FDG = 93.7%; of CT/MRI = 76.7%; difference = 17.0%, 95% CI = 4.9% to 28.5%, P = .013).ConclusionsCompared with (123)I-MIBG SPECT and CT/MRI, both considered gold standards for PPGL imaging, metastases were better detected by (18)F-FDG PET. (18)F-FDG PET provides a high specificity in patients with a biochemically established diagnosis of PPGL.

Dataset Information

Gaussian Mixture Models and Model Selection for [18F] Fluorodeoxyglucose Positron Emission Tomography Classification in Alzheimer's Disease.

Publications

Gaussian Mixture Models and Model Selection for [18F] Fluorodeoxyglucose Positron Emission Tomography Classification in Alzheimer's Disease.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets