RESULTS: Our models learned several syntactic, lexical, and n-gram linguistic biomarkers to distinguish the probable AD group from the healthy group. In contrast to the healthy group, we found that the probable AD patients had significantly less usage of syntactic components and significantly higher usage of lexical components in their language. Also, we observed a significant difference in the use of n-grams as the healthy group were able to identify and make sense of more objects in their n-grams than the probable AD group. As such, our best diagnostic model significantly distinguished the probable AD group from the healthy elderly group with a better Area Under the Receiving Operating Characteristics Curve (AUC) using the Support Vector Machines (SVM).
CONCLUSIONS: Experimental and statistical evaluations suggest that using ML algorithms for learning linguistic biomarkers from the verbal utterances of elderly individuals could help the clinical diagnosis of probable AD. We emphasise that the best ML model for predicting the disease group combines significant syntactic, lexical and top n-gram features. However, there is a need to train the diagnostic models on larger datasets, which could lead to a better AUC and clinical diagnosis of probable AD.
METHODS: We recruited 164 healthy controls (HC) and 120 cognitively impaired (CI) subjects- 47 mild cognitive impairment (MCI) and 73 mild Alzheimer's disease (AD) dementia participants, from four countries between January 2015 and August 2016 to determine the usefulness of a single version of the VCAT, without translation or adaptation, in a multinational, multilingual population. The VCAT was administered along with established cognitive evaluation.
RESULTS: The VCAT, without local translation or adaptation, was effective in discriminating between HC and CI subjects (MCI and mild AD dementia). Mean (SD) VCAT scores for HC and CI subjects were 22.48 (3.50) and 14.17 (5.05) respectively. Areas under the curve for Montreal Cognitive Assessment (0.916, 95% CI 0.884-0.948) and the VCAT (0.905, 95% CI 0.870-0.940) in discriminating between HCs and CIs were comparable. The multiple languages used to administer VCAT in four countries did not significantly influence test scores.
CONCLUSIONS: The VCAT without the need for language translation or cultural adaptation showed satisfactory discriminative ability and was effective in a multinational, multilingual Southeast Asian population.
OBJECTIVE: We aimed to correlate the ability of these modalities to differentiate Probable AD and Possible AD using the clinical diagnosis as a gold standard. We also investigated the correlation of severity of amyloid deposit in the brain with the diagnosis of AD.
METHODS: A retrospective study of 47 subjects (17 Probable AD and 30 Possible AD) who were referred for PET/CT amyloid scans to our centre was conducted. Hypoperfusion in the temporo-parietal lobes on Tc99m-HMPAO SPECT and loss of grey-white matter contrast in cortical regions on PET/CT Amyloid scans indicating the presence of amyloid β deposit were qualitatively interpreted as positive for AD. SPECT and PET/CT were also read in combination (Combo reading). The severity of amyloid β deposit was semiquantitatively assessed in a visual binary method using a scale of Grade 0-4. The severity of amyloid β deposit was assessed in a visual binary method and a semi-quantitative method using a scale of Grade 0-4.
RESULTS: There was significant correlation of Tc99m-HMPAO SPECT, PET/CT amyloid findings and Combo reading with AD. The sensitivity, specificity, PPV and NPV were 87.5%, 73.7%, 58.3% and 93.3% (SPECT); 62.5%, 77.4%, 58.8% and 80.0% (PET/CT) and 87.5%, 84.2%, 70.0% and 30.0% (Combo reading) respectively. The grade of amyloid deposition was not significantly correlated with AD (Spearman's correlation, p=0.687).
CONCLUSION: There is an incremental benefit in utilizing PET/CT amyloid imaging in cases with atypical presentation and indeterminate findings on conventional imaging of Alzheimer's disease.
METHODS: Further to informed consent from 39 healthy subjects and 39 probable AD patients, 8.5 mL of peripheral blood was collected and serum was extracted. The differential levels of 12 serum cytokines extracted from peripheral blood samples were measured using Procarta Multiplex Cytokine and enzyme-linked immunoassay kits. Concentrations of cytokines were measured at 615 nm using a fluorometer.
RESULTS: Except for tumor necrosis factor-α, all classical pro-inflammatory cytokines (interleukin [IL]-1β, IL-6, IL-12 and interferon-γ) were found to be significantly upregulated (P 53.65 ρg/mL and <9.315 ρg/mL, respectively).
CONCLUSIONS: Both the non-classical pro-inflammatory CXCL-10 and anti-inflammatory IL-13 cytokines showed promising potential as blood-based cytokine biomarkers for AD. This is the first study of non-classical cytokine profiles of Malaysian AD patients. Geriatr Gerontol Int 2017; 17: 839-846.