Imaging Of Brain Tissue Research Articles

Magnetic Induction Tomography for Brain Tissue Imaging Based on Conductivity Distribution for Parkinson’s Disease Diagnosis

Parkinson's disease is a prevalent neurodegenerative complication defined by the accumulation of alpha synuclein lewy bodies in the brain. Misdiagnosis results widespread of Parkinson’s disease because clinical diagnosis is challenging, underlining a need of a better detection technique, such as non-invasive magnetic induction tomography (MIT) technique. Non-invasive techniques for biological tissues imaging are becoming popular in biomedical engineering field. Therefore, MIT technology as a non-invasive technique has been encouraged in a medical field due to its advancement of technology in diagnosing diseases. The measurement parameters in MIT are passive electromagnetic properties (conductivity, permittivity, permeability) for biological tissue and the most dominant parameter in MIT is conductivity properties. It is uses a phase shift between a primary magnetic field and an induced field caused by a target object's conductivity. As a function of conductivity, the phase shift between the applied and secondary fields is expressed. Thus, the phase shift can be used to characterize the conductivity of a target object. The phase shift between the excitation and induced magnetic fields (EMF and IMF) reflects the change in conductivity in biological tissues. This paper focuses on the virtual simulation by using COMSOL Multi-physics for the design and development of MIT system that emphasizes on single channel magnetic induction tomography for biological tissue (bran tissue) imaging based on conductivity distribution for Parkinson’s disease diagnosis. The develop system employs the use of excitation coils to induce an electromagnetic field (e.m.f) in the brain tissue, which is then measured at the receiving side by sensors. The proposed system is capable of indicating Parkinson’s disease based on conductivity distribution. This method provides the valuable information of the brain abnormality based on differences of conductivities of normal brain and Parkinson’s disease brain tissues.

Learning precise segmentation of neurofibrillary tangles from rapid manual point annotations.

Accumulation of abnormal tau protein into neurofibrillary tangles (NFTs) is a pathologic hallmark of Alzheimer disease (AD). Accurate detection of NFTs in tissue samples can reveal relationships with clinical, demographic, and genetic features through deep phenotyping. However, expert manual analysis is time-consuming, subject to observer variability, and cannot handle the data amounts generated by modern imaging. We present a scalable, open-source, deep-learning approach to quantify NFT burden in digital whole slide images (WSIs) of post-mortem human brain tissue. To achieve this, we developed a method to generate detailed NFT boundaries directly from single-point-per-NFT annotations. We then trained a semantic segmentation model on 45 annotated 2400μm by 1200μm regions of interest (ROIs) selected from 15 unique temporal cortex WSIs of AD cases from three institutions (University of California (UC)-Davis, UC-San Diego, and Columbia University). Segmenting NFTs at the single-pixel level, the model achieved an area under the receiver operating characteristic of 0.832 and an F1 of 0.527 (196-fold over random) on a held-out test set of 664 NFTs from 20 ROIs (7 WSIs). We compared this to deep object detection, which achieved comparable but coarser-grained performance that was 60% faster. The segmentation and object detection models correlated well with expert semi-quantitative scores at the whole-slide level (Spearman's rho ρ=0.654 (p=6.50e-5) and ρ=0.513 (p=3.18e-3), respectively). We openly release this multi-institution deep-learning pipeline to provide detailed NFT spatial distribution and morphology analysis capability at a scale otherwise infeasible by manual assessment.

Blind non-linear spectral unmixing with spatial coherence for hyper and multispectral images

Multi and hyperspectral images have become invaluable sources of information, revolutionizing various fields such as remote sensing, environmental monitoring, agriculture and medicine. In this expansive domain, the multi-linear mixing model (MMM) is a versatile tool to analyze spatial and spectral domains by effectively bridging the gap between linear and non-linear interactions of light and matter. This paper introduces an upgraded methodology that integrates the versatility of MMM in non-linear spectral unmixing, while leveraging spatial coherence (SC) enhancement through total variation theory to mitigate noise effects in the abundance maps. Referred to as non-linear extended blind end-member and abundance extraction with SC (NEBEAE-SC), the proposed methodology relies on constrained quadratic optimization, cyclic coordinate descent algorithm, and the split Bregman formulation. The validation of NEBEAE-SC involved rigorous testing on various hyperspectral datasets, including a synthetic image, remote sensing scenarios, and two biomedical applications. Specifically, our biomedical applications are focused on classification tasks, the first addressing hyperspectral images of in-vivo brain tissue, and the second involving multispectral images of ex-vivo human placenta. Our results demonstrate an improvement in the abundance estimation by NEBEAE-SC compared to similar algorithms in the state-of-the-art by offering a robust tool for non-linear spectral unmixing in diverse application domains.

Correlating Histopathological Microscopic Images of Creutzfeldt–Jakob Disease with Clinical Typology Using Graph Theory and Artificial Intelligence

Creutzfeldt–Jakob disease (CJD) is a rare, degenerative, and fatal brain disorder caused by abnormal proteins called prions. This research introduces a novel approach combining AI and graph theory to analyze histopathological microscopic images of brain tissues affected by CJD. The detection and quantification of spongiosis, characterized by the presence of vacuoles in the brain tissue, plays a crucial role in aiding the accurate diagnosis of CJD. The proposed methodology employs image processing techniques to identify these pathological features in high-resolution medical images. By developing an automatic pipeline for the detection of spongiosis, we aim to overcome some limitations of manual feature extraction. The results demonstrate that our method correctly identifies and characterize spongiosis and allows the extraction of features that will help to better understand the spongiosis patterns in different CJD patients.

FLUID: a fluorescence-friendly lipid-compatible ultrafast clearing method.

Many clearing methods achieve high transparency by removing lipid components from tissues, which damages microstructure and limits their application in lipid research. As for methods which preserve lipid, it is difficult to balance transparency, fluorescence preservation and clearing speed. In this study, we propose a rapid water-based clearing method that is fluorescence-friendly and preserves lipid components. FLUID allows for preservation of endogenous fluorescence over 60 days. It shows negligible tissue distortion and is compatible with various types of fluorescent labeling and tissue staining methods. High quality imaging of human brain tissue and compatibility with pathological staining demonstrated the potential of our method for three-dimensional (3D) biopsy and clinical pathological diagnosis.

Associations Between Neuropsychiatric Symptoms and Inflammation in Neurodegenerative Dementia: A Systematic Review.

Neuropsychiatric symptoms are common in dementia and linked to adverse outcomes. Inflammation is increasingly recognized as playing a role as a driver of early disease progression in Alzheimer's disease (AD) and related dementias. Inflammation has also been linked to primary psychiatric disorders, however its association with neuropsychiatric symptoms in neurodegenerative dementias remains uncertain. We conducted a systematic literature review investigating associations between inflammation and neuropsychiatric symptoms in neurodegenerative dementias, including AD, Lewy body, Frontotemporal, Parkinson's (PD) and Huntington's disease dementias. Ninety-nine studies met our inclusion criteria, and the majority (n = 59) investigated AD and/or mild cognitive impairment (MCI). Thirty-five studies included PD, and only 6 investigated non-AD dementias. Inflammation was measured in blood, CSF, by genotype, brain tissue and PET imaging. Overall, studies exhibited considerable heterogeneity and evidence for specific inflammatory markers was inconsistent, with lack of replication and few longitudinal studies with repeat biomarkers. Depression was the most frequently investigated symptom. In AD, some studies reported increases in peripheral IL-6, TNF-a associated with depressive symptoms. Preliminary investigations using PET measures of microglial activation found an association with agitation. In PD, studies reported positive associations between TNF-a, IL-6, CRP, MCP-1, IL-10 and depression. Central and peripheral inflammation may play a role in neuropsychiatric symptoms in neurodegenerative dementias; however, the evidence is inconsistent. There is a need for multi-site longitudinal studies with detailed assessments of neuropsychiatric symptoms combined with replicable peripheral and central markers of inflammation.

SynBot is an open-source image analysis software for automated quantification of synapses

The formation of precise numbers of neuronal connections, known as synapses, is crucial for brain function. Therefore, synaptogenesis mechanisms have been one of the main focuses of neuroscience. Immunohistochemistry is a common tool for visualizing synapses. Thus, quantifying the numbers of synapses from light microscopy images enables screening the impacts of experimental manipulations on synapse development. Despite its utility, this approach is paired with low-throughput analysis methods that are challenging to learn, and the results are variable between experimenters, especially when analyzing noisy images of brain tissue. We developed an open-source ImageJ-based software, SynBot, to address these technical bottlenecks by automating the analysis. SynBot incorporates the advanced algorithms ilastik and SynQuant for accurate thresholding for synaptic puncta identification, and the code can easily be modified by users. The use of this software will allow for rapid and reproducible screening of synaptic phenotypes in healthy and diseased nervous systems.

Exploring nalbuphine loaded chitosan nanoparticles for effective pain management through intranasal administration: a comparative study

Background Intranasal drug delivery shows potential for brain access via olfactory and trigeminal routes. Purpose This work aimed to ensure brain availability of nalbuphine via the nasal route. Method Chitosan based nanoparticles loaded with nalbuphine were successfully prepared using ionic gelation method and characterised. Result SEM results revealed that the nanoparticles were spherical in shape, with an average size of 192.4 ± 11.6 nm. Zeta potential and entrapment efficiency was found 32.8 mV and 88.43 ± 7.75%, respectively. The X-ray diffractometry and DSC results unravel a profound understanding on the physical and thermal characteristics. The in-vitro release of nalbuphine from the nanoparticles was biphasic, with an initial burst release followed by a slow-release profile. In-vitro cell study on HEK-293 cells and microscopic images of brain tissue confirmed the safety profile of formulation. In-vivo efficacy studies on animal confirmed the effectiveness of developed intranasal formulation as compared to the standard therapy. The in-vivo pharmacokinetic studies showed that the prepared nanoparticles were able to efficiently deliver nalbuphine to the brain in comparison to the other body organs. Gamma scintigraphy images showed retention of the drug in the brain. Furthermore, the efficacy studies confirmed that the nanoparticles were found significantly more effective than the marketed formulation in pain management.

Metabolomic profiling integrated with molecular exploring delineates the action of Ligusticum chuanxiong hort. on migraine

IntroductionLigusticum chuanxiong hort. (Chuanxiong, CX) is a well-known traditional edible-medicinal herb, especially in brain diseases. However, there is a lack of studies focusing on the action of CX in metabolites of migraine. Hypothesis/purposeTo investigate the action of the aqueous extract of CX (LCH) on nitroglycerin (NTG)-induced migraine. MethodsThe effects and mechanisms of LCH on migraine were evaluated in NTG-induced mice and rats. Laser speckle contrast imaging was performed to detect the cerebral blood flow. Metabolomic analysis in serum and mass spectrometry imaging in brain tissue from NTG-induced rats were conducted to explore the metabolites. The techniques including RT-qPCR, immunohistochemistry, and immunofluorescence were employed to estimate the molecular changes. ResultsTotally, 1480 metabolites were identified, among which, 50 and 49 differential metabolites were identified by t-test, fold change, and VIP score for NTG vs. CON and LCH+NTG vs. NTG, respectively. Next, 13 common real difference metabolites were revealed by comparative analysis, and KEGG annotation and enrichment analysis showed that the glutathione (GSH) metabolism pathway played important roles in migraine, whereas the glutamate could be metabolized to γ-glu-cys and converted to GSH. Molecular exploring further confirmed that LCH treatment increased the expression of essential components of GSH synthetase, such as GCLC and GCLM, and elevated the expression levels of Nrf-2 and its downstream targets including HO1 and NQO1. Moreover, the mass spectrometry imaging results found that LCH treatment promoted the synthesis of GSH and the spatial distribution of glucose as well as ATP metabolites to normal levels. ConclusionTo sum up, the present study firstly reveals that LCH plays a therapeutic role for migraine through glucose-glutamate-Nrf-2 axis, which might represent a promising approach in the development of advanced therapeutic strategies for migraine, and the LCH may be an effective drug or dietary supplement for relieving headache.

Fast imaging of millimeter-scale areas with beam deflection transmission electron microscopy

Serial section transmission electron microscopy (TEM) has proven to be one of the leading methods for millimeter-scale 3D imaging of brain tissues at nanoscale resolution. It is important to further improve imaging efficiency to acquire larger and more brain volumes. We report here a threefold increase in the speed of TEM by using a beam deflecting mechanism to enable highly efficient acquisition of multiple image tiles (nine) for each motion of the mechanical stage. For millimeter-scale areas, the duty cycle of imaging doubles to more than 30%, yielding a net average imaging rate of 0.3 gigapixels per second. If fully utilized, an array of four beam deflection TEMs should be capable of imaging a dataset of cubic millimeter scale in five weeks.

Very-long T2-weighted imaging of the non-lesional brain tissue in multiple sclerosis patients.

The purpose of this study is to demonstrate that T2-weighted imaging with very long echo time (TE > 300 ms) can provide relevant information in neurodegenerative/inflammatory disorder. Twenty patients affected by relapsing-remitting multiple sclerosis with stable disease course underwent 1.5T 3D FLAIR, 3D T1-weighted, and a multi-echo sequence with 32 echoes (TE = 10-320 ms). Focal lesions (FL) were identified on FLAIR. T1-images were processed to segment deep gray matter (dGM), white matter (WM), FL sub-volumes with T1 hypo-intensity (T1FL), and dGM volumes (atrophy). Clinical-radiological parameters included Expanded Disability Status Scale (EDSS), disease duration, patient age, T1FL, and dGM atrophy. Correlation analysis was performed between the mean signal intensity (SI) computed on the non-lesional dGM and WM at different TE versus the clinical-radiological parameters. Multivariable linear regressions were fitted to the data to assess the association between the dependent variable EDSS and the independent variables obtained by T1FL lesion load and the mean SI of dGM and WM at the different TE. A clear trend is observed, with a systematic strengthening of the significance of the correlation at longer TE for all the relationships with the clinical-radiological parameters, becoming significant (p < 0.05) for EDSS, T1FL volumes, and dGM atrophy. Multivariable linear regressions show that at shorter TE, the SI of the T2-weighted sequences is not relevant for describing the EDSS variability while the T1FL volumes are relevant, and vice versa, at very-long TEs (around 300 ms); the SI of the T2-weighted sequences significantly (p < 0.05) describes the EDSS variability. By very long TE, the SI primarily originates from water with a T2 longer than 250 ms and/or free water, which may be arising from the perivascular space (PVS). Very-long T2-weighting might detect dilated PVS and represent an unexplored MR approach in neurofluid imaging of neurodegenerative/inflammatory diseases.

NnSegNeXt: A 3D Convolutional Network for Brain Tissue Segmentation Based on Quality Evaluation.

Accurate and automated segmentation of brain tissue images can significantly streamline clinical diagnosis and analysis. Manual delineation needs improvement due to its laborious and repetitive nature, while automated techniques encounter challenges stemming from disparities in magnetic resonance imaging (MRI) acquisition equipment and accurate labeling. Existing software packages, such as FSL and FreeSurfer, do not fully replace ground truth segmentation, highlighting the need for an efficient segmentation tool. To better capture the essence of cerebral tissue, we introduce nnSegNeXt, an innovative segmentation architecture built upon the foundations of quality assessment. This pioneering framework effectively addresses the challenges posed by missing and inaccurate annotations. To enhance the model's discriminative capacity, we integrate a 3D convolutional attention mechanism instead of conventional convolutional blocks, enabling simultaneous encoding of contextual information through the incorporation of multiscale convolutional features. Our methodology was evaluated on four multi-site T1-weighted MRI datasets from diverse sources, magnetic field strengths, scanning parameters, temporal instances, and neuropsychiatric conditions. Empirical evaluations on the HCP, SALD, and IXI datasets reveal that nnSegNeXt surpasses the esteemed nnUNet, achieving Dice coefficients of 0.992, 0.987, and 0.989, respectively, and demonstrating superior generalizability across four distinct projects with Dice coefficients ranging from 0.967 to 0.983. Additionally, extensive ablation studies have been implemented to corroborate the effectiveness of the proposed model. These findings represent a notable advancement in brain tissue analysis, suggesting that nnSegNeXt holds the promise to significantly refine clinical workflows.

Deep Learning With Physics-Embedded Neural Network for Full Waveform Ultrasonic Brain Imaging.

The convenience, safety, and affordability of ultrasound imaging make it a vital non-invasive diagnostic technique for examining soft tissues. However, significant differences in acoustic impedance between the skull and soft tissues hinder the successful application of traditional ultrasound for brain imaging. In this study, we propose a physics-embedded neural network with deep learning based full waveform inversion (PEN-FWI), which can achieve reliable quantitative imaging of brain tissues. The network consists of two fundamental components: forward convolutional neural network (FCNN) and inversion sub-neural network (ISNN). The FCNN explores the nonlinear mapping relationship between the brain model and the wavefield, replacing the tedious wavefield calculation process based on the finite difference method. The ISNN implements the mapping from the wavefield to the model. PEN-FWI includes three iterative steps, each embedding the F CNN into the ISNN, ultimately achieving tomography from wavefield to brain models. Simulation and laboratory tests indicate that PEN-FWI can produce high-quality imaging of the skull and soft tissues, even starting from a homogeneous water model. PEN-FWI can achieve excellent imaging of clot models with constant uniform distribution of velocity, randomly Gaussian distribution of velocity, and irregularly shaped randomly distributed velocity. Robust differentiation can also be achieved for brain slices of various tissues and skulls, resulting in high-quality imaging. The imaging time for a horizontal cross-sectional imag e of the brain is only 1.13 seconds. This algorithm can effectively promote ultrasound-based brain tomography and provide feasible solutions in other fields.

A modular framework for multi-scale tissue imaging and neuronal segmentation.

The development of robust tools for segmenting cellular and sub-cellular neuronal structures lags behind the massive production of high-resolution 3D images of neurons in brain tissue. The challenges are principally related to high neuronal density and low signal-to-noise characteristics in thick samples, as well as the heterogeneity of data acquired with different imaging methods. To address this issue, we design a framework which includes sample preparation for high resolution imaging and image analysis. Specifically, we set up a method for labeling thick samples and develop SENPAI, a scalable algorithm for segmenting neurons at cellular and sub-cellular scales in conventional and super-resolution STimulated Emission Depletion (STED) microscopy images of brain tissues. Further, we propose a validation paradigm for testing segmentation performance when a manual ground-truth may not exhaustively describe neuronal arborization. We show that SENPAI provides accurate multi-scale segmentation, from entire neurons down to spines, outperforming state-of-the-art tools. The framework will empower image processing of complex neuronal circuitries.

Photonic neural probe enabled microendoscopes for light-sheet light-field computational fluorescence brain imaging.

Light-sheet fluorescence microscopy is widely used for high-speed, high-contrast, volumetric imaging. Application of this technique to in vivo brain imaging in non-transparent organisms has been limited by the geometric constraints of conventional light-sheet microscopes, which require orthogonal fluorescence excitation and collection objectives. We have recently demonstrated implantable photonic neural probes that emit addressable light sheets at depth in brain tissue, miniaturizing the excitation optics. Here, we propose a microendoscope consisting of a light-sheet neural probe packaged together with miniaturized fluorescence collection optics based on an image fiber bundle for lensless, light-field, computational fluorescence imaging. Foundry-fabricated, silicon-based, light-sheet neural probes can be packaged together with commercially available image fiber bundles to form microendoscopes for light-sheet light-field fluorescence imaging at depth in brain tissue. Prototype microendoscopes were developed using light-sheet neural probes with five addressable sheets and image fiber bundles. Fluorescence imaging with the microendoscopes was tested with fluorescent beads suspended in agarose and fixed mouse brain tissue. Volumetric light-sheet light-field fluorescence imaging was demonstrated using the microendoscopes. Increased imaging depth and enhanced reconstruction accuracy were observed relative to epi-illumination light-field imaging using only a fiber bundle. Our work offers a solution toward volumetric fluorescence imaging of brain tissue with a compact size and high contrast. The proof-of-concept demonstrations herein illustrate the operating principles and methods of the imaging approach, providing a foundation for future investigations of photonic neural probe enabled microendoscopes for deep-brain fluorescence imaging in vivo.

An actively stabilized, miniaturized epi-fluorescence widefield microscope for real-time observation in vivo.

Recent developments in real-time, in vivo micro-imaging have allowed for the visualization of tissue pathological changes, facilitating rapid diagnosis. However, miniaturization, magnification, the field of view, and in vivo image stabilization remain challenging factors to reconcile. A key issue for this technology is ensuring it is user friendly for surgeons, enabling them to use the device manually and obtain instantaneous information necessary for surgical decision-making. This descriptive study introduces a handheld, actively stabilized, miniaturized epi-fluorescence widefield microscope (MEW-M) for real-time observation in vivo with high resolution. The methodology of MEW-M system includes high resolution microscopy miniaturization technology, thousandfold shaking suppression (actively stabilized), ultra-photosensitivity, and tailored image signal processing cell image capture and processing technology, which support for the excellent real-time imaging performance of MEW-M system in brain, mammary, liver, lung, and kidney tissue imaging of rats in vivo. With a single-objective and high-frame-rate imaging, the MEW-M system facilitates roving image acquisition, enabling contiguous analysis of large tissue areas. RESEARCH HIGHLIGHTS: A handheld, actively stabilized MEW-M system was introduced. Excellent real-time, in vivo imaging with high resolution and active stabilization in brain, mammary, liver, lung, and kidney tissue of rats.

Brain MRI detection and classification: Harnessing convolutional neural networks and multi-level thresholding.

Brain tumor detection in clinical applications is a complex and challenging task due to the intricate structures of the human brain. Magnetic Resonance (MR) imaging is widely preferred for this purpose because of its ability to provide detailed images of soft brain tissues, including brain tissue, cerebrospinal fluid, and blood vessels. However, accurately detecting brain tumors from MR images remains an open problem for researchers due to the variations in tumor characteristics such as intensity, texture, size, shape, and location. To address these issues, we propose a method that combines multi-level thresholding and Convolutional Neural Networks (CNN). Initially, we enhance the contrast of brain MR images using intensity transformations, which highlight the infected regions in the images. Then, we use the suggested CNN architecture to classify the enhanced MR images into normal and abnormal categories. Finally, we employ multi-level thresholding based on Tsallis entropy (TE) and differential evolution (DE) to detect tumor region(s) from the abnormal images. To refine the results, we apply morphological operations to minimize distortions caused by thresholding. The proposed method is evaluated using the widely used Harvard Medical School (HMS) dataset, and the results demonstrate promising performance with 99.5% classification accuracy and 92.84% dice similarity coefficient. Our approach outperforms existing state-of-the-art methods in brain tumor detection and automated disease diagnosis from MR images.

Image Enhancement Method for Photoacoustic Imaging of Deep Brain Tissue

Photoacoustic imaging (PAI) is an emerging biomedical imaging modality, offering numerous advantages, including high resolution and high contrast. In its application to brain imaging, however, the photoacoustic (PA) signals from brain tissue weaken considerably due to the distortion effects of the skull. This attenuation reduces the resolution and contrast significantly. To address this issue, here we describe a Log-MSR algorithm that combines the logarithmic depth logarithmic enhancement (Log) algorithm and the multi-scale Retinex (MSR) algorithm. In this method, the Log algorithm performs local weighted compensation based on signal attenuation for different depths, while the MSR algorithm improves the contrast of the image. The proposed Log-MSR algorithm was tested and validated using several phantom and in vivo experiments. The enhanced images constructed by the Log-MSR algorithm were qualitatively and quantitatively analyzed in terms of brain structure and function. Our results show that the Log-MSR algorithm may provide a significant enhancement to photoacoustic imaging of deep brain tissue.

Explainable AI and its instantiation in Computational Pathology for a better understanding of Alzheimer’s disease

AbstractBackgroundEXplainable Artificial Intelligence (XAI) has an essential role to accelerate the AI adoption by empowering biomedical experts and catalyzing modern research through the emergence of a new generation of virtual actors, enabling traceable knowledge‐based quantification, as a guided qualification support. XAI instantiation is shown through our study, hypothesizing that rapid Alzheimer’s Disease (rAD) brains display subtle histological changes that would be undercovered by high‐throughput automated microscopic analysis. We study the topography and morphology of tau lesions’ aggregates to better understand the morphological substratum of Alzheimer’s Disease (AD) heterogeneity.MethodTo address this question at a large scale, we designed, tested and implemented a software for automatic segmentation, annotation and quantitation of brain lesions in histopathological whole slide images (WSI). A dataset of 15 whole slide images of postmortem human brain tissue is used in this study. These are fully annotated by neuropathologists yielding a set of more than 30,000 annotated plaques and tangle objects. A deep learning pipeline consisting of an attention‐UNet model with explainability features is trained for segmentation of tau aggregates in WSIs. The attention maps corresponding to segmentation predictions from the model are used by pathologists as a comparable reference to assist their decision making process.ResultThe DL‐based WSI analysis framework allows detecting a majority of the tau aggregates and refining tau aggregate boundaries with higher accuracy compared to manual annotations. Besides, with such a DL model based tau identification, the pathologist needs to annotate only a few samples for the DL model training. Overall, DL assisted analysis of tau aggregates could ease the effort of pathologists and facilitate analysis of hundreds of WSIs which are vital in AD research areas such as identifying different AD forms and AD patient stratification.ConclusionXAI is not only providing traceability by opening the way to an effective adoption, but also allows optimizing the ML/DL design, by keeping high performances. The green footprint is, therefore, considered, as well as the GDPR (European General Data Protection Regulation), towards Responsible Artificial Intelligence.

Multi-layered maps of neuropil with segmentation-guided contrastive learning

Maps of the nervous system that identify individual cells along with their type, subcellular components and connectivity have the potential to elucidate fundamental organizational principles of neural circuits. Nanometer-resolution imaging of brain tissue provides the necessary raw data, but inferring cellular and subcellular annotation layers is challenging. We present segmentation-guided contrastive learning of representations (SegCLR), a self-supervised machine learning technique that produces representations of cells directly from 3D imagery and segmentations. When applied to volumes of human and mouse cortex, SegCLR enables accurate classification of cellular subcompartments and achieves performance equivalent to a supervised approach while requiring 400-fold fewer labeled examples. SegCLR also enables inference of cell types from fragments as small as 10 μm, which enhances the utility of volumes in which many neurites are truncated at boundaries. Finally, SegCLR enables exploration of layer 5 pyramidal cell subtypes and automated large-scale analysis of synaptic partners in mouse visual cortex.