Search

Dementia is a type of brain disease that affects the mental abilities. Various studies utilize PET features or some two-dimensional brain perspectives to diagnose dementia. In this study, we have proposed an ensemble approach, which employs volumetric and axial perspective features for the diagnosis of Alzheimer's disease and the patients with mild cognitive impairment. We have employed deep learning models and constructed two disparate networks. The first network evaluates volumetric features, and the second network assesses grid-based brain scan features. Decisions of these networks were combined by an adaptive majority voting algorithm to create an ensemble learner. In the evaluations, we compared ensemble networks with single ones as well as feature fusion networks to identify possible improvement; as a result, the ensemble method turned out to be promising for making a diagnostic decision. The proposed ensemble network achieved an average accuracy of 91.83% for the diagnosis of Alzheimer's disease; to the best of our knowledge, it is the highest diagnosis performance in the literature. [ABSTRACT FROM AUTHOR]

Objective: Niemann Pick disease type C (NPC) is a rare progressive neurovisceral lysosomal disorder caused by autosomal recessive mutations in the NPC1 or NPC2 genes. 18F-fluorodeoxyglucose (FDG) is a positron-emitting glucose analogue for non-invasive imaging of brain metabolism. FDG PET is commonly used for dementia imaging but its specific application to NPC is rarely described. Methods: This is a retrospective study of all baseline brain FDG PET performed for NPC patients. Images were assessed using a normal database statistical comparison of metabolic changes expressed in standard deviations and three-dimensional Stereotactic Surface Projection maps. Typical hypometabolic patterns in NPC were identified. We further investigated any correlation between the degree of regional brain hypometabolism and the Iturriaga clinical severity scale. Results: Brain FDG PET images of 14 adolescent–adult NPC patients were analysed, with mean age of 35 years. We found significant frontal lobe hypometabolism in 12 patients (86%), thalamic hypometabolism in eight patients (57%) and variable parietal lobe hypometabolism in 13 patients (93%). Hypometabolic changes were usually bilateral and symmetric. Ten out of 13 ataxic patients showed cerebellar or thalamic hypometabolism (sensitivity 77%, specificity 100%). Linear regression analysis showed frontal lobe hypometabolism to have the best correlation with the Iturriaga clinical scale (R2 = 0.439; p = 0.01). Conclusions: We found bilateral symmetric hypometabolism of the frontal lobes, thalami and parietal lobes (especially posterior cingulate gyrus) to be typical of adolescent–adult NPC. Ataxia was commonly associated with cerebellar or thalamic hypometabolism. Frontal lobe hypometabolism showed the best inverse correlation with clinical severity. [ABSTRACT FROM AUTHOR]

Background: Prior to interpreting PET/CT, it is crucial to understand the normal biodistribution of fluorodeoxyglucose (FDG). It is also important to realize that the normal biodistribution can vary between adults and children. Although many studies have defined normal patterns of pediatric FDG uptake in structures like the thymus, brown fat and bone marrow, patterns of normal pediatric bowel activity, specifically uptake within the appendix, have not been well described. Active lymphoid tissue has increased FDG uptake when compared with inactive tissue. Since children have more active lymphoid tissue than adults, and because the appendix contains aggregated lymphoid tissue, we postulated that appendiceal uptake may be increased in pediatric patients. Objective: To define the normal level of appendiceal FDG activity in children by evaluating a series of consecutive FDG PET/CT scans performed for other indications. Materials and methods: After obtaining IRB approval, we retrospectively reviewed 128 consecutive whole-body pediatric FDG PET/CT examinations obtained for a variety of clinical indications. CT scans on which the appendix could not be visualized were excluded from analysis. CT scans on which the appendix could be visualized were evaluated for underlying appendiceal pathology. Studies with appendiceal or periappendiceal pathology by CT criteria were excluded. A region of interest (ROI) was placed over a portion of each appendix and appendiceal maximum standardized uptake value (SUVmax) was calculated. If an adjacent loop of bowel activity interfered with accurate measurements of the appendix SUVmax, the scan was excluded from the analysis. A chart review was performed on patients with elevated appendiceal SUVmax values to ensure that the patients did not have clinical symptomatology suggestive of acute appendicitis. When the appendix or a portion of the appendix could be visualized and accurately measured, the SUVmax was determined. SUVmax of the appendix was compared to the SUVmax of normal liver and ratios were recorded. Results: A total of 128 scans were reviewed, patient ages 1 month to 21 years (mean age: 11.6 years). Thirty-one scans were excluded because of inability to visualize the appendix on CT. No scans were excluded for appendiceal/periappendiceal pathology on CT or chart review. No scans had to be excluded for inability to obtain an accurate SUVmax measurement because measurements were calculated on portions of the appendix separate from adjacent bowel using small ROIs. Maximum appendiceal SUVs ranged from 0.5 to 9.4 (mean: 2.2) with an appendix-to-liver background ratio ranging from 0.3 to 3.1 (mean: 1.1). Conclusion: FDG uptake in the appendix is typically similar to that of background activity. However, slight variations in appendiceal FDG uptake do occur, which should not be misinterpreted as pathological. [ABSTRACT FROM AUTHOR]