Clustering the cortical laminae: in vivo parcellation

A Deep Learning-based Pipeline for Segmenting the Cerebral Cortex Laminar Structure in Histology Images

Characterizing the anatomical structure and connectivity between cortical regions is a critical step towards understanding the information processing properties of the brain and will help provide insight into the nature of neurological disorders. A key feature of the mammalian cerebral cortex is its laminar structure. Identifying these layers in neuroimaging data is important for understanding their global structure and to help understand the connectivity patterns of neurons in the brain. We studied Nissl-stained and myelin-stained slice images of the brain of the common marmoset (Callithrix jacchus), which is a new world monkey that is becoming increasingly popular in the neuroscience community as an object of study. We present a novel computational framework that first acquired the cortical labels using AI-based tools followed by a trained deep learning model to segment cerebral cortical layers. We obtained a Euclidean distance of 1274.750 ± 156.400 μ m for the cortical labels acquisition, which was in the acceptable range by computing the half Euclidean distance of the average cortex thickness ( 1800.630 μ m ). We compared our cortical layer segmentation pipeline with the pipeline proposed by Wagstyl et al. (PLoS biology, 18(4), e3000678 2020) adapted to 2D data. We obtained a better mean95 th percentile Hausdorff distance (95HD) of  92.150 μ m . Whereas a mean 95HD of  94.170 μ m was obtained from Wagstyl et al. We also compared our pipeline's performance against theirs using their dataset (the BigBrain dataset). The results also showed better segmentation quality, 85.318 % Jaccard Index acquired from our pipeline, while 83.000 % was stated in their paper.

Resiliency to Alzheimer's disease neuropathology can be distinguished from dementia using cortical astrogliosis imaging

Despite the presence of significant Alzheimer's disease (AD) pathology, characterized by amyloid β (Aβ) plaques and phosphorylated tau (pTau) tangles, some cognitively normal elderly individuals do not inevitably develop dementia. These findings give rise to the notion of cognitive 'resilience', suggesting maintained cognitive function despite the presence of AD neuropathology, highlighting the influence of factors beyond classical pathology. Cortical astroglial inflammation, a ubiquitous feature of symptomatic AD, shows a strong correlation with cognitive impairment severity, potentially contributing to the diversity of clinical presentations. However, noninvasively imaging neuroinflammation, particularly astrogliosis, using MRI remains a significant challenge. Here we sought to address this challenge and to leverage multidimensional (MD) MRI, a powerful approach that combines relaxation with diffusion MR contrasts, to map cortical astrogliosis in the human brain by accessing sub-voxel information. Our goal was to test whether MD-MRI can map astroglial pathology in the cerebral cortex, and if so, whether it can distinguish cognitive resiliency from dementia in the presence of hallmark AD neuropathological changes. We adopted a multimodal approach by integrating histological and MRI analyses using human postmortem brain samples. Ex vivo cerebral cortical tissue specimens derived from three groups comprised of non-demented individuals with significant AD pathology postmortem, individuals with both AD pathology and dementia, and non-demented individuals with minimal AD pathology postmortem as controls, underwent MRI at 7 T. We acquired and processed MD-MRI, diffusion tensor, and quantitative T 1 and T 2 MRI data, followed by histopathological processing on slices from the same tissue. By carefully co-registering MRI and microscopy data, we performed quantitative multimodal analyses, leveraging targeted immunostaining to assess MD-MRI sensitivity and specificity towards Aβ, pTau, and glial fibrillary acidic protein (GFAP), a marker for astrogliosis. Our findings reveal a distinct MD-MRI signature of cortical astrogliosis, enabling the creation of predictive maps for cognitive resilience amid AD neuropathological changes. Multiple linear regression linked histological values to MRI changes, revealing that the MD-MRI cortical astrogliosis biomarker was significantly associated with GFAP burden (standardized β=0.658, pFDR<0.0001), but not with Aβ (standardized β=0.009, p FDR =0.913) or pTau (standardized β=-0.196, p FDR =0.051). Conversely, none of the conventional MRI parameters showed significant associations with GFAP burden in the cortex. While the extent to which pathological glial activation contributes to neuronal damage and cognitive impairment in AD is uncertain, developing a noninvasive imaging method to see its affects holds promise from a mechanistic perspective and as a potential predictor of cognitive outcomes.

BANKSY unifies cell typing and tissue domain segmentation for scalable spatial omics data analysis

Spatial omics data are clustered to define both cell types and tissue domains. We present Building Aggregates with a Neighborhood Kernel and Spatial Yardstick (BANKSY), an algorithm that unifies these two spatial clustering problems by embedding cells in a product space of their own and the local neighborhood transcriptome, representing cell state and microenvironment, respectively. BANKSY's spatial feature augmentation strategy improved performance on both tasks when tested on diverse RNA (imaging, sequencing) and protein (imaging) datasets. BANKSY revealed unexpected niche-dependent cell states in the mouse brain and outperformed competing methods on domain segmentation and cell typing benchmarks. BANKSY can also be used for quality control of spatial transcriptomics data and for spatially aware batch effect correction. Importantly, it is substantially faster and more scalable than existing methods, enabling the processing of millions of cell datasets. In summary, BANKSY provides an accurate, biologically motivated, scalable and versatile framework for analyzing spatially resolved omics data.