Dataset Information

Deep Learning With 18F-Fluorodeoxyglucose-PET Gives Valid Diagnoses for the Uncertain Cases in Memory Impairment of Alzheimer's Disease.

ABSTRACT: Objectives: Neuropsychological tests are an important basis for the memory impairment diagnosis in Alzheimer's disease (AD). However, multiple memory tests might be conflicting within-subjects and lead to uncertain diagnoses in some cases. This study proposed a framework to diagnose the uncertain cases of memory impairment. Methods: We collected 2,386 samples including AD, mild cognitive impairment (MCI), and cognitive normal (CN) using 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) and three different neuropsychological tests (Mini-Mental State Examination, Alzheimer's Disease Assessment Scale-Cognitive Subscale, and Clinical Dementia Rating) from the Alzheimer's Disease Neuroimaging Initiative (ADNI). A deep learning (DL) framework using FDG-PET was proposed to diagnose uncertain memory impairment cases that were conflicting between tests. Subsequent ANOVA, chi-squared, and t-test were used to explain the potential causes of uncertain cases. Results: For certain cases in the testing set, the proposed DL framework outperformed other methods with 95.65% accuracy. For the uncertain cases, its positive diagnoses had a significant (p < 0.001) worse decline in memory function than negative diagnoses in a longitudinal study of 40 months on average. In the memory-impaired group, uncertain cases were mainly explained by an AD metabolism pattern but mild in extent (p < 0.05). In the healthy group, uncertain cases were mainly explained by a non-energetic mental state (p < 0.001) measured using a global deterioration scale (GDS), with a significant depression-related metabolism pattern detected (p < 0.05). Conclusion: A DL framework for diagnosing uncertain cases of memory impairment is proposed. Proved by longitudinal tracing of its diagnoses, it showed clinical validity and had application potential. Its valid diagnoses also provided evidence and explanation of uncertain cases based on the neurodegeneration and depression mental state.

SUBMITTER: Zhang W

PROVIDER: S-EPMC8715958 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:There is a lack of imaging markers revealing the functional characteristics of different brain regions in paediatric dystonia. In this observational study, we assessed the utility of [18F]2-fluoro-2-deoxy-D-glucose (FDG)-PET in understanding dystonia pathophysiology by revealing specific resting awake brain glucose metabolism patterns in different childhood dystonia subgroups. PET scans from 267 children with dystonia being evaluated for possible deep brain stimulation surgery between September 2007 and February 2018 at Evelina London Children's Hospital (ELCH), UK, were examined. Scans without gross anatomical abnormality (e.g. large cysts, significant ventriculomegaly; n = 240) were analysed with Statistical Parametric Mapping (SPM12). Glucose metabolism patterns were examined in the 144/240 (60%) cases with the 10 commonest childhood-onset dystonias, focusing on nine anatomical regions. A group of 39 adult controls was used for comparisons. The genetic dystonias were associated with the following genes: TOR1A, THAP1, SGCE, KMT2B, HPRT1 (Lesch Nyhan disease), PANK2 and GCDH (Glutaric Aciduria type 1). The acquired cerebral palsy (CP) cases were divided into those related to prematurity (CP-Preterm), neonatal jaundice/kernicterus (CP-Kernicterus) and hypoxic-ischaemic encephalopathy (CP-Term). Each dystonia subgroup had distinct patterns of altered FDG-PET uptake. Focal glucose hypometabolism of the pallidi, putamina or both, was the commonest finding, except in PANK2, where basal ganglia metabolism appeared normal. HPRT1 uniquely showed glucose hypometabolism across all nine cerebral regions. Temporal lobe glucose hypometabolism was found in KMT2B, HPRT1 and CP-Kernicterus. Frontal lobe hypometabolism was found in SGCE, HPRT1 and PANK2. Thalamic and brainstem hypometabolism were seen only in HPRT1, CP-Preterm and CP-term dystonia cases. The combination of frontal and parietal lobe hypermetabolism was uniquely found in CP-term cases. PANK2 cases showed a distinct combination of parietal hypermetabolism with cerebellar hypometabolism but intact putaminal-pallidal glucose metabolism. HPRT1, PANK2, CP-kernicterus and CP-preterm cases had cerebellar and insula glucose hypometabolism as well as parietal glucose hypermetabolism. The study findings offer insights into the pathophysiology of dystonia and support the network theory for dystonia pathogenesis. 'Signature' patterns for each dystonia subgroup could be a useful biomarker to guide differential diagnosis and inform personalized management strategies.

Project description:Inflammation and neovascularization in vulnerable atherosclerotic plaques are key features for severe clinical events. Dynamic contrast-enhanced (DCE) MRI and FDG PET are two noninvasive imaging techniques capable of quantifying plaque neovascularization and inflammatory infiltrate, respectively. However, their mutual role in defining plaque vulnerability and their possible overlap has not been thoroughly investigated. We studied the relationship between DCE-MRI and (18)F-FDG PET data from the carotid arteries of 40 subjects with coronary heart disease (CHD) or CHD risk equivalent, as a substudy of the dal-PLAQUE trial (NCT00655473).The dal-PLAQUE trial was a multicenter study that evaluated dalcetrapib, a cholesteryl ester transfer protein modulator. Subjects underwent anatomical MRI, DCE-MRI and (18)F-FDG PET. Only baseline imaging and biomarker data (before randomization) from dal-PLAQUE were used as part of this substudy. Our primary goal was to evaluate the relationship between DCE-MRI and (18)F-FDG PET data. As secondary endpoints, we evaluated the relationship between (a) PET data and whole-vessel anatomical MRI data, and (b) DCE-MRI and matching anatomical MRI data. All correlations were estimated using a mixed linear model.We found a significant inverse relationship between several perfusion indices by DCE-MRI and (18)F-FDG uptake by PET. Regarding our secondary endpoints, there was a significant relationship between plaque burden measured by anatomical MRI with several perfusion indices by DCE-MRI and (18)F-FDG uptake by PET. No relationship was found between plaque composition by anatomical MRI and DCE-MRI or (18)F-FDG PET metrics.In this study we observed a significant, weak inverse relationship between inflammation measured as (18)F-FDG uptake by PET and plaque perfusion by DCE-MRI. Our findings suggest that there may be a complex relationship between plaque inflammation and microvascularization during the different stages of plaque development. (18)F-FDG PET and DCE-MRI may have complementary roles in future clinical practice in identifying subjects at high risk of cardiovascular events.

Project description:IntroductionWe assessed the impact of [(18)F]-fluorodeoxyglucose (FDG)-positron emission tomography (PET) on intended management of patients in the National Oncologic PET Registry (NOPR) for three different diagnostic indications: (a) determining whether a suspicious lesion is cancer (Dx), (b) detecting an unknown primary tumor site when there is confirmed or strongly suspected metastatic disease (cancer of unknown primary origin [CUP]), and (c) detecting a primary tumor site when there is a presumed paraneoplastic syndrome (PNS).MethodsWe reviewed a sample of randomly selected reports of NOPR subjects who underwent PET for Dx and CUP and all reports for PNS to find subjects for analysis. For these studies, we evaluated the impact of PET on referring physicians' intended management, based on their management plans reported before and after PET.ResultsIntended management was changed more frequently in the CUP group (43.1%) than in the Dx (23.9%) and PNS (25.4%) groups (CUP vs. Dx, p < .0001; PNS vs. Dx, p < .0001; CUP vs. PNS, p < .0002). Referring physicians reported that, in light of PET results, they were able to avoid further testing in approximately three-fourths of patients (71.8%-74.6%). At the time when the post-PET forms were completed, biopsies of suspicious sites had been performed in 21.2%, 32.4%, and 23.2%, respectively, of Dx, CUP, and PNS cases.ConclusionOur analysis of NOPR data shows that PET appears to have a substantial impact on intended management when used for three common diagnostic indications.Implications for practice[(18)F]-fluorodeoxyglucose-positron emission tomography appears to have a substantial impact on intended management when used for three targeted diagnostic indications: (a) determining whether a suspicious lesion is cancer, (b) detecting an unknown primary tumor site in a patient with confirmed or strongly suspected metastatic disease, and (c) detecting a primary tumor site in a patient with a presumed paraneoplastic syndrome.

Dataset Information

Deep Learning With 18F-Fluorodeoxyglucose-PET Gives Valid Diagnoses for the Uncertain Cases in Memory Impairment of Alzheimer's Disease.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets