Objective: To identify brain regions whose metabolic impairment contributes to dementia with Lewy bodies (DLB) clinical core features expression and to assess the influence of severity of global cognitive impairment on the DLB hypometabolic pattern. Methods: Brain fluorodeoxyglucose positron emission tomography and information on core features were available in 171 patients belonging to the imaging repository of the European DLB Consortium. Principal component analysis was applied to identify brain regions relevant to the local data variance. A linear regression model was applied to generate core-feature-specific patterns controlling for the main confounding variables (Mini-Mental State Examination [MMSE], age, education, gender, and center). Regression analysis to the locally normalized intensities was performed to generate an MMSE-sensitive map. Results: Parkinsonism negatively covaried with bilateral parietal, precuneus, and anterior cingulate metabolism; visual hallucinations (VH) with bilateral dorsolateral-frontal cortex, posterior cingulate, and parietal metabolism; and rapid eye movement sleep behavior disorder (RBD) with bilateral parieto-occipital cortex, precuneus, and ventrolateral-frontal View this article online at wileyonlinelibrary.com.

Background Striatal dopamine deficiency and metabolic changes are well‐known phenomena in dementia with Lewy bodies and can be quantified in vivo by 123I‐Ioflupane brain single‐photon emission computed tomography of dopamine transporter and 18F‐fluorodesoxyglucose PET. However, the linkage between both biomarkers is ill‐understood. Objective We used the hitherto largest study cohort of combined imaging from the European consortium to elucidate the role of both biomarkers in the pathophysiological course of dementia with Lewy bodies. Methods We compared striatal dopamine deficiency and glucose metabolism of 84 dementia with Lewy body patients and comparable healthy controls. After normalization of data, we tested their correlation by region‐of‐interest–based and voxel‐based methods, controlled for study center, age, sex, education, and current cognitive impairment. Metabolic connectivity was analyzed by inter‐region coefficients stratified by dopamine deficiency and compared to healthy controls. Results There was an inverse relationship between striatal dopamine availability and relative glucose hypermetabolism, pronounced in the basal ganglia and in limbic regions. With increasing dopamine deficiency, metabolic connectivity showed strong deteriorations in distinct brain regions implicated in disease symptoms, with greatest disruptions in the basal ganglia and limbic system, coincident with the pattern of relative hypermetabolism. Conclusions Relative glucose hypermetabolism and disturbed metabolic connectivity of limbic and basal ganglia circuits are metabolic correlates of dopamine deficiency in dementia with Lewy bodies. Identification of specific metabolic network alterations in patients with early dopamine deficiency may serve as an additional supporting biomarker for timely diagnosis of dementia with Lewy bodies. © 2019 The Authors. Movement Disorders published by Wiley Periodicals, Inc. on behalf of International Parkinson and Movement Disorder Society.

Purpose The purpose of this study is to develop and validate a 3D deep learning model that predicts the final clinical diagnosis of Alzheimer’s disease (AD), dementia with Lewy bodies (DLB), mild cognitive impairment due to Alzheimer’s disease (MCI-AD), and cognitively normal (CN) using fluorine 18 fluorodeoxyglucose PET (18F-FDG PET) and compare model’s performance to that of multiple expert nuclear medicine physicians’ readers. Materials and methods Retrospective 18F-FDG PET scans for AD, MCI-AD, and CN were collected from Alzheimer’s disease neuroimaging initiative (556 patients from 2005 to 2020), and CN and DLB cases were from European DLB Consortium (201 patients from 2005 to 2018). The introduced 3D convolutional neural network was trained using 90% of the data and externally tested using 10% as well as comparison to human readers on the same independent test set. The model’s performance was analyzed with sensitivity, specificity, precision, F1 score, receiver operating characteristic (ROC). The regional metabolic changes driving classification were visualized using uniform manifold approximation and projection (UMAP) and network attention. Results The proposed model achieved area under the ROC curve of 96.2% (95% confidence interval: 90.6–100) on predicting the final diagnosis of DLB in the independent test set, 96.4% (92.7–100) in AD, 71.4% (51.6–91.2) in MCI-AD, and 94.7% (90–99.5) in CN, which in ROC space outperformed human readers performance. The network attention depicted the posterior cingulate cortex is important for each neurodegenerative disease, and the UMAP visualization of the extracted features by the proposed model demonstrates the reality of development of the given disorders. Conclusion Using only 18F-FDG PET of the brain, a 3D deep learning model could predict the final diagnosis of the most common neurodegenerative disorders which achieved a competitive performance compared to the human readers as well as their consensus.