BACKGROUNDIn human lupus nephritis (LN), tubulointerstitial inflammation (TII) on biopsy predicts progression to end-stage renal disease (ESRD). However, only about half of patients with moderate-to-severe TII develop ESRD. We hypothesized that this heterogeneity in outcome reflects different underlying inflammatory states. Therefore, we interrogated biopsies from LN longitudinal and cross-sectional cohorts.METHODSData were acquired using conventional highly multiplexed confocal microscopy....

Quantitative assessment of retinal microvasculature in optical coherence tomography angiography (OCTA) images is important for studying, diagnosing, monitoring, and guiding the treatment ocular systemic diseases. However, OCTA user community lacks universal transparent image analysis tools that can be applied to from a range instruments provide reliable consistent microvascular metrics diverse datasets. We present extension Vascular Analyser (OCTAVA) addresses challenges providing robust,...

Abstract The rapid development of highly multiplexed microscopy systems has enabled the study cells embedded within their native tissue, which is providing exciting insights into spatial features human disease [1]. However, computational methods for analyzing these high-content images are still emerging, and there a need more robust generalizable tools evaluating cellular constituents underlying stroma captured by high-plex imaging [2]. To address this need, we have adapted spectral angle...

Tuberculosis is a pulmonary disease with an especially high mortality rate in immuno‐compromised populations, specifically children and HIV positive individuals. The causative agent, Mycobacterium tuberculosis ( Mtb ), very slow growing difficult organism to work with, making both diagnosis development of effective treatments cumbersome. We utilize fiber‐optic fluorescence microendoscope integrated whole‐body imaging system for vivo detection. exploits endogenous enzyme β ‐lactamase, or...

Systemic lupus erythematosus (SLE) is a complex, systemic autoimmune disease with many clinical presentations including nephritis (LuN), or chronic inflammation of the kidneys. Current therapies for SLE are only modestly effective, highlighting need to better understand networks immune cells in and LuN. In this work, we assess performance two convolutional neural network (CNN) architectures –Mask R-CNN U-Net— task instance segmentation 5 immune-cell classes 31 LuN biopsies. Each biopsy was...

Significance: Lupus nephritis (LuN) is a chronic inflammatory kidney disease. The cellular mechanisms by which LuN progresses to failure are poorly characterized. Automated instance segmentation of immune cells in immunofluorescence images can probe these interactions. Aim: Our specific goal quantify how sample fixation and staining panel design impact automated characterization cells. Approach: Convolutional neural networks (CNNs) were trained segment fluorescence confocal biopsies. Three...

SignificanceManual annotations are necessary for training supervised learning algorithms object detection and instance segmentation. These manual difficult to acquire, noisy, inconsistent across readers.AimThe goal of this work is describe demonstrate multireader generalizations the Jaccard Sørensen indices segmentation.ApproachThe described in terms "calls," "objects," number readers. reduce equations defined by confusion matrix variables two-reader case. In a test set 50 cell microscopy...

We employ a concentric sphere Mie scattering model to describe light by pulmonary alveoli and airway surface liquid (ASL). Using this layered model, we compare alveolar at different points along the respiratory cycle observe effect of ASL thickness on in lung. have also extrapolated investigate various animal models disease. This can estimate vivo optical properties for normal pathological states, potentially aiding design systems diagnosis investigation pathologies.

The rapid development of new optical imaging techniques is dependent on the availability low-cost, customizable, and easily reproducible standards. By replicating environment, costly animal experiments to validate a technique may be circumvented. Predicting optimizing performance in vivo ex requires testing samples that are optically similar tissues interest. Tissue-mimicking phantoms provide standard for evaluation, characterization, or calibration an system. Homogenous polymer tissue...

Deep convolutional neural networks (CNNs) have demonstrated high accuracy in a wide range of computer vision applications, including medical and biological imaging. Many CNNs are fully supervised learning algorithms, their performance is directly associated with the quality training data labels, which human-defined. In this work, we investigate fidelity human-defined truth for cell detection, segmentation, classification tasks multiplex microscopy images. We compare manual annotations from...

We describe a Monte Carlo model of the mouse torso to optimize illumination lung for fluorescence detection low levels pulmonary pathogens, specifically Mycobacterium tuberculosis. After validation simulation with an internally illuminated optical phantom, entire was simulated compare external and internal techniques. Measured properties deflated lungs were scaled mimic diffusive inflated in vivo. Using full-torso model, 2 × 3 improvement average fluence rate seen dorsal compared ventral...

Computer vision and deep learning are integral tools in the improvement of high-throughput analysis cellular images. Specifically, optimization algorithms for object detection instance segmentation tasks important image to segment classify multi-object, multi-class In this work, we employ an pipeline with Mask RCNN, using a ResNet-101 Feature Pyramid Network convolutional backbone T cells antigen presenting (APCs) multi-channel fluorescence confocal images lupus nephritis biopsies. This task...

We demonstrate an instance segmentation method with Mask R-CNN using a ResNet-101 plus Feature Pyramid Network convolutional backbone to segment and classify T cells antigen presenting (APCs) in multi-channel fluorescence confocal images. This network improves on our previous cell distance mapping (CDM) pipeline, which used custom 10- layer neural for segmentation. have validated two independent datasets of images: 1) mouse lymph node tissue, 2) human lupus nephritis (LuN) biopsies. For...

Highly multiplexed fluorescence microscopy is an emerging technology that allows for spatial analysis of increasingly more classes cells within human tissue—state-of-the-art methods are now probing up to 60 different protein markers image. This level phenotypic resolution ideal uncovering the underpinnings immune cell interactions. However, defining types from this high-plex data non-trivial. We present a method borrows hyperspectral image improve accuracy and efficiency classification in...

Lupus nephritis (LN) is a severe manifestation of systemic lupus erythematosus, with up to 30% LN patients progressing end-stage kidney disease within ten years diagnosis. Spatial relationships between specific types immune cells and structures hold valuable information clinically biologically. Thus, we develop modular computational pipeline analyze the spatially resolved molecular features from high-plex immunofluorescence imaging data. Here, present three modules pipeline, goal achieving...

Single-cell sequencing and proteomics have been critical for the study of human disease. However, highly multiplexed microscopy has revolutionized spatial biology by measuring cell expression from ~50 proteins while maintaining locations cells. This presents unique computational challenges; acquiring manual annotations across so many image channels is challenging, therefore supervised learning methods classification are undesirable. To overcome this limitation we developed a decision-tree...

The goal of this work is to reduce the complexity cell and neighborhood annotations in studies for spatial immunity. Specifically, we use a method inspired by spectral angle mapping collapse multichannel images into class-level representations. We will demonstrate that these class maps assist characterizing immune infiltration renal pathologies.

Tuberculosis is one of the deadliest infectious diseases worldwide.New tools to study pathogenesis and monitor subjects in pre-clinical studies develop treatment regimens are critical for progress.We developed an improved optical system detecting bacteria lungs mice using internal illumination.We present a computational model full mouse torso characterize system.Simulated theoretical limits lowest detectable bacterial load support experimental improvements with illumination source, suggest...

The rapid development of new optical imaging techniques is dependent on the availability low-cost, customizable, and easily reproducible standards. By replicating environment, costly animal experiments to validate a technique may be circumvented. Predicting optimizing performance in vivo ex requires testing samples that are optically similar tissues interest. Tissue-mimicking phantoms provide standard for evaluation, characterization, or calibration an system. Homogenous polymer tissue...

Germinal center (GC) B cells segregate into three subsets that compartmentalize the antagonistic molecular programs of selection, proliferation, and somatic hypermutation. In bone marrow, epigenetic reader BRWD1 orchestrates insulates sequential stages cell proliferation

Several disease states, including cancer and autoimmunity, are characterized by the infiltration of large populations immune cells into organ tissue. The degree composition these invading have been correlated with patient outcomes, suggesting that intercellular interactions occurring in inflamed tissue play a role pathology. Immunofluorescence staining paired confocal microscopy produce detailed visualizations interactions. Applying computer vision machine learning methods to resulting...

Triple-negative breast cancer (TNBC) is an aggressive subtype of defined by the lack hormone receptor overexpression. TNBC patients are at a higher risk recurrence than with other cancers. As this disease disproportionally affects young women color, there urgent need to address health inequity improving detection, prognostication, and therapy guidance. Currently, expected have better prognosis if biopsy analysis shows more tumor-infiltrating immune cells. However, immunotherapies such as...