The prevalence, correlation, and co-occurrence of neuropathology in old age: harmonisation of 12 measures across six community-based autopsy studies of dementia.

BACKGROUND: Population-based autopsy studies provide valuable insights into the causes of dementia but are limited by sample size and restriction to specific populations. Harmonisation across studies increases statistical power and allows meaningful comparisons between studies. We aimed to harmonise neuropathology measures across studies and assess the prevalence, correlation, and co-occurrence of neuropathologies in the ageing population. METHODS: We combined data from six community-based autopsy cohorts in the US and the UK in a coordinated cross-sectional analysis. Among all decedents aged 80 years or older, we assessed 12 neuropathologies known to be associated with dementia: arteriolosclerosis, atherosclerosis, macroinfarcts, microinfarcts, lacunes, cerebral amyloid angiopathy, Braak neurofibrillary tangle stage, Consortium to Establish a Registry for Alzheimer's disease (CERAD) diffuse plaque score, CERAD neuritic plaque score, hippocampal sclerosis, limbic-predominant age-related TDP-43 encephalopathy neuropathologic change (LATE-NC), and Lewy body pathology. We divided measures into three groups describing level of confidence (low, moderate, and high) in harmonisation. We described the prevalence, correlations, and co-occurrence of neuropathologies. FINDINGS: The cohorts included 4354 decedents aged 80 years or older with autopsy data. All cohorts included more women than men, with the exception of one study that only included men, and all cohorts included decedents at older ages (range of mean age at death across cohorts 88·0-91·6 years). Measures of Alzheimer's disease neuropathological change, Braak stage and CERAD scores, were in the high confidence category, whereas measures of vascular neuropathologies were in the low (arterioloscerosis, atherosclerosis, cerebral amyloid angiopathy, and lacunes) or moderate (macroinfarcts and microinfarcts) categories. Neuropathology prevalence and co-occurrence was high (2443 [91%] of 2695 participants had more than one of six key neuropathologies and 1106 [41%] of 2695 had three or more). Co-occurrence was strongly but not deterministically associated with dementia status. Vascular and Alzheimer's disease features clustered separately in correlation analyses, and LATE-NC had moderate associations with Alzheimer's disease measures (eg, Braak stage ρ=0·31 [95% CI 0·20-0·42]). INTERPRETATION: Higher variability and more inconsistency in the measurement of vascular neuropathologies compared with the measurement of Alzheimer's disease neuropathological change suggests the development of new frameworks for the measurement of vascular neuropathologies might be helpful. Results highlight the complexity and multi-morbidity of the brain pathologies that underlie dementia in older adults and suggest that prevention efforts and treatments should be multifaceted. FUNDING: Gates Ventures.

Brain SegNet: 3D local refinement network for brain lesion segmentation.

MR images (MRIs) accurate segmentation of brain lesions is important for improving cancer diagnosis, surgical planning, and prediction of outcome. However, manual and accurate segmentation of brain lesions from 3D MRIs is highly expensive, time-consuming, and prone to user biases. We present an efficient yet conceptually simple brain segmentation network (referred as Brain SegNet), which is a 3D residual framework for automatic voxel-wise segmentation of brain lesion. Our model is able to directly predict dense voxel segmentation of brain tumor or ischemic stroke regions in 3D brain MRIs. The proposed 3D segmentation network can run at about 0.5s per MRIs - about 50 times faster than previous approaches Med Image Anal 43: 98-111, 2018, Med Image Anal 36:61-78, 2017. Our model is evaluated on the BRATS 2015 benchmark for brain tumor segmentation, where it obtains state-of-the-art results, by surpassing recently published results reported in Med Image Anal 43: 98-111, 2018, Med Image Anal 36:61-78, 2017. We further applied the proposed Brain SegNet for ischemic stroke lesion outcome prediction, with impressive results achieved on the Ischemic Stroke Lesion Segmentation (ISLES) 2017 database.