Multimodal brain imaging and vascular risk factors in classification of people with Alzheimer's dementia

Background: Alzheimer's Disease (AD) is the most common cause of dementia in older people. To date, the gold standard for AD diagnosis is neuropathological examination, which is impractical and using stringent diagnostic criteria are the only viable option for AD diagnosis, resulting inoverlap of different pathologies in people diagnosed with AD. This study is an attempt to classify people with AD through visual interpretation of [18F] fluorodeoxy glucose positron emission tomography (FDG PET) images and to identify and differentiate typical AD from mixed subtype. Material and Methods: The current study is based on FDG PET data collected as part of two large scale Phase 3 clinical trials by Tau Therapeutics (TauRx) pharmaceuticals of a novel tau aggregation inhibitor medication, LMTXÒ in 970 AD participants. All available FDG PET images were visually classified into typical AD and mixed subtype using PMOD Alzheimer's Discrimination Tool (PALZ) and further confirmed using quantitative analysis and automated transfer learning method. Results: Participants with mixed subtype were younger and had more severe disease in comparison to those with typical AD subtype. Hypertension was significantly correlated with typical AD and White Matter Hyperintensities (WMH) were significantly associated with mixed subtype, when compared with mixed subtype and age-matched controls respectively. Left cortical hypo-metabolism and right hypo-metabolism in the cerebellum, typical of crossed cerebellar diaschisis (CCD).