22689 research outputs found
Sort by
Optimized Colitis Model with Peak Colonic Macrophage Infiltration
Get PDFAlthough the etiology of inflammatory bowel disease (IBD), an increasingly popular disease of chronic intestinal inflammation, has not been fully understood, macrophage has been defined as the key mediator in IBD pathological progression. Thus, the development of nanotechnology inspired new nanodrug vehicle to deliver drugs specifically to colonic macrophages as innovative IBD therapeutics, but researchers do not have the colitis model with robust macrophage infiltration in colon to screen the biodistribution and evaluate the macrophage-targeting efficacy of nanodrug vehicles. Here, we referred to the reproducible DSS-induced and TNBS-induced acute colitis mice model and optimized the protocol. Afterward, we determined that the peak colonic macrophage infiltration occurred on Day 2 after TNBS enema and on Day 7 after initiation of DSS feeding based on macroscopic observation and flow cytometry. These optimized mice colitis models can be a useful platform to conduct the nanodrug biodistribution assay which helps assessing delivery efficiency and specificity
Illuminating the Changes of Auditory Cortical Microcircuits From Early Development to Aging
Get PDFHearing is crucial for communication and interaction with the world. The primary auditory cortex (ACtx) consists of a sophisticated neuronal circuit for the precise processing of external auditory stimuli and the generation of appropriate behavioral reactions. The neuronal structure and functional connectivity of ACtx are highly dynamic throughout life. Multiple studies have revealed the general trends of cortical microcircuit alteration, but it is not clear whether other factors, such as molecularly defined subpopulations or sex, have different impacts on the connection changes. To investigate these questions, I used laser-scanning photostimulation (LSPS) coupled with whole cell patch clamp recording to reconstruct the intracortical input connection maps at single neuron level. Layer-specific input changes in intracortical circuits were compared between three major life periods: early development, adulthood, and aging. Subplate neurons and layer 2/3 neurons were selected for analysis for early development and aging, respectively, because of their active participation in intracortical communication and information processing during these time periods. Connective tissue growth factor (CTGF) and dopamine receptor D1 (Drd1) labeled subplate neurons were selected as they consist of different proportions of the overall subplate population. Overall, the results revealed circuit differences between CTGF and Drd1-labeled subplate neurons before the onset of sensory input and that this distinction was eliminated after hearing onset. Drd1 neurons were more inhibited by other cortical layers compared to CTGF neurons before hearing onset. With age, males and females exhibited different trends in the alterations of intracortical connectivity, consistent with more severe inhibition changes in aging males. Therefore, future medical therapeutics of cortical connectivity disorders, such as cochlear implants or deep brain stimulation (DBS), should take these factors as sources of consideration
dVPose: Automated Data Collection and Dataset for 6D Pose Estimation of Robotic Surgical Instruments
Get PDFWe present dVPose, a realistic multi-modality dataset intended for use in the development and evaluation of real-time single-shot deep-learning based 6D pose estimation algorithms on a head mounted display (HMD).
In addition to the dataset, our contribution includes an automated (robotic) data collection platform that integrates an accurate optical tracking system to provide the ground-truth poses.
We collected a comprehensive set of data for vision-based 6D pose estimation, including images and poses of the extra-corporeal portions of the instruments and endoscope of a da Vinci surgical robot.
The images are collected using the multi-camera rig of the Microsoft HoloLens 2 HMD, mounted on a UR10 robot, and the corresponding poses are collected by optically tracking both the instruments/endoscope and HMD.
The intended application is to enable markerless localization of the HMD with respect to the da Vinci robot, considering that the instruments and endoscope are among the few robotic components that are not covered by sterile drapes.
Our dataset features synchronized images from the RGB, depth, and grayscale cameras of the HoloLens 2 device.
It is unique in that it provides medically focused images, provides images from a HoloLens 2 device where object tracking is a fundamental task, and provides data from multiple visible-light cameras in addition to depth.
Furthermore, the automated data collection platform can be easily adapted to collect images and ground-truth poses of other objects
Transformer Neural Network for Navigation State Estimation
Get PDFNavigation systems have long relied on the Kalman filter for accurate and reliable navigation state estimation, establishing it as a baseline in various applications, from aerospace
to robotics. With the surge of deep learning and its unparalleled capabilities in handling
complex data structures, Transformer Neural Networks (TNNs) have emerged as compelling
timeseries forecasters. While TNNs promise enhanced performance by leveraging spatial and
temporal dependencies in data, it remains uncertain how they compare to the traditional
Kalman filter in terms of estimation accuracy, reliability, and computational efficiency. This
thesis embarks on a comparison between a traditional navigation Kalman filter and a Transformer Neural Network applied as the navigation state estimator. Using simulated navigation
scenarios, we evaluate the performance metrics of both techniques. The key objective is to
ascertain the strengths and limitations of the TNN approach vis-à-vis a baseline navigation
Kalman filter. The thesis findings will offer insights into the feasibility of using a TNN
in a navigation system and the conditions under which it might outperform the Kalman Filter
Essays on Misspecified Leaning and Choice
Get PDFThis dissertation presents four independent essays in economic theory: three on misspecified learning and a fourth on solvability axioms in mathematical psychology. Model misspecification in decision-making environments has been of substantive interest to economists since the early 1970s.1 Motivated by this theme, Chapter 1 presents novel monotone comparative statics results for steady-state behavior in a dynamic optimization environment with misspecified Bayesian learning. It considers a generalized framework, based on Esponda and Pouzo (2021), wherein a Bayesian learner facing a dynamic optimization problem has a prior on a set of parameterized transition probability functions (models) but is misspecified in the sense that the true process is not within this set. In the steady state, the learner infers the model that best fits the data generated by their actions, and in turn, their actions are optimally chosen given their inferred model. I characterize conditions on the primitives of the environment, and in particular, over the set of models under which the steady-state distribution over states and actions and inferred models exhibit monotonic behavior. Further, I offer a new theorem on the existence of a steady state on the basis of a monotonicity argument. Lastly, I provide an upper bound on the cost of misspecification, again in terms of the primitives of the environment. Chapter 2, published in the Journal of Economic Theory (2024), studies the existence of equilibria in misspecified environments. In the context of misspecified Markov Decision Processes, Esponda and Pouzo (2021) defined the notion of Berk-Nash equilibrium and established its existence with finite state and action spaces. However, many substantive applications (including two of the three motivating examples presented by Esponda and Pouzo) involve continuous state or action spaces, and are thus not covered by the Esponda-Pouzo existence theorem. We extend the existence of Berk-Nash equilibrium to compact action spaces and sigma-compact state spaces, with possibly unbounded utility functions. Chapter 3, published in Theory and Decision (2022), presents four theorems that connect continuity postulates in mathematical economics to solvability axioms in mathematical psychology, and ranks them under alternative supplementary assumptions. Theorem 1 connects notions of continuity (full, separate, Wold, weak Wold, Archimedean, mixture) with those of solvability (restricted, unrestricted) under the completeness and transitivity of a binary relation. Theorem 2 uses the primitive notion of a separately continuous function to answer the question when an analogous property on a relation is fully continuous. Theorem 3 provides a portmanteau theorem on the equivalence between restricted solvability and various notions of continuity under weak monotonicity. After a brief excursion into mathematical psychology, I return to the topic of misspecified learning in Chapter 4. However, I diverge from the Bayesian paradigm to explore a specific instance of naive learning, akin to DeGroot’s heuristic, within networks characterized by diverse and potentially misspecified worldviews. Drawing upon one of my papers that was published in Economics Letters (2021), I examine a feasible weighting matrix rooted in information entropy and outline the utility of such a framework in addressing issues related to awareness, consensus, and convergence rates within networks featuring social learning
Quality of Post-Violence Clinical Care Services in Mozambique
No full textWorldwide, 31% of women aged 15-49 years have experienced gender-based violence (GBV) at some point in their lives, leading to both short and long-term health consequences including HIV (WHO, 2021). Over the past decade, Mozambique has established and scaled up GBV prevention and response services within existing HIV clinical services. However, the quality of these post-violence care services had not been systematically assessed.
This exploratory, concurrent mixed-methods study assessed the quality of post-violence care services in Mozambique. Exploratory quantitative data analysis was used to identify the strengths and weaknesses of post-violence clinical care services across 10 domains of service delivery using a structured GBV Quality Assurance Tool in fifty health facilities across eight provinces (Aim 1). In-depth qualitative interviews with 20 health providers further explored perceived programmatic strengths and weaknesses (Aim 2), and actionable recommendations were proposed to address the most substantial weaknesses (Aim 3).
About half of the quantitative weaknesses (7 of the top 15), and most of the qualitative weaknesses (11 of 15) pertained to a lack of sufficient inputs (materials, infrastructure and/or human resources). Only 10% of health facilities had at least 3 months stock of medicines and consumables to respond to cases of violence and only 34% of health facilities had essential equipment. About half of the quantitative weaknesses (8 of 15) and 27% of the qualitative weaknesses (4 of 15) pertained to a lack of appropriate processes. Only 60% of providers properly documented information in patient files and 74% of health facilities implemented annual supervision plans.
These findings suggest that additional, targeted investment of resources is necessary to address inadequate inputs. The Ministry of Health GBV Program is working at national, regional, provincial and district levels to address process deficiencies. Additional recommendations include the need for dedicated GBV staff; training for all health staff as part of their mandatory health professional curriculum; stronger multisectoral response via robust, integrated one-stop centers and regular multisectoral meetings; temporary shelters and tailored safety plans; and emphasis on community demand creation for both GBV and HIV services (including HIV post-exposure prophylaxis)
Prioritization of causal genes from genome-wide association studies by Bayesian data integration across loci
Get PDFGenome-wide association studies (GWAS) have identified genetic variants, usually single-nucleotide polymorphisms (SNPs), associated with human traits, including disease and disease risk. These variants (or causal variants in linkage disequilibrium with them) usually affect the regulation or function of a nearby gene. A GWAS locus can span many genes, however, and prioritizing which gene or genes in a locus are most likely to be causal remains a challenge. Better prioritization and prediction of causal genes could reveal disease mechanisms and suggest interventions.
We describe a new Bayesian method, termed SigNet for significance networks, that combines information both within and across loci to identify the most likely causal gene at each locus. The SigNet method builds on existing methods that focus on individual loci with evidence from gene distance and expression quantitative trait loci (eQTL) by sharing information across loci using protein-protein and gene regulatory interaction network data.
In an application to cardiac electrophysiology with 225 GWAS loci, only 46 (20%) have strong within-locus evidence. At the remaining 179 loci, SigNet selects 63 genes other than the minimum distance gene, equal to 35% of the information-poor loci and 28% of the GWAS loci overall. Assessment by both pathway enrichment demonstrates improved performance by SigNet, and review of individual loci suggests PMP22 as a novel causal gene candidate
BUILDING CONSUMER ENGAGEMENT INTO THE BALTIMORE URBAN FOOD DISTRIBUTION MOBILE APPLICATION: IMPROVING FOOD ACCESS ACROSS THE BALTIMORE CITY FOOD SYSTEM
Get PDFBackground: Diet-related chronic disease disproportionately affects under-resourced urban communities across the United States. Evidence-based solutions targeting multiple levels of the food system are needed for increased reach, impact and sustainability. The Baltimore Urban food Distribution (‘BUD’) study leverages a mobile application (app) to facilitate distribution of healthy foods and beverages between small retailers and local suppliers in Baltimore, Maryland. This dissertation project builds upon this work by engaging consumers in the app’s supply-demand feedback loop.
Objectives: To: (1) develop the BUD app for retailers and suppliers; (2) conduct multi-methods research to characterize target end consumers, their food sourcing and consumption patterns; and (3) design a consumer-engagement module (called ‘BUDConnect’) for future integration into the app.
Methods: In Phase 1, formative research informed programming of the BUD app, and a preliminary pilot was conducted in three corner stores. Quantitative and qualitative data were collected from Baltimore adults (N=127) via an online cross-sectional survey. Individual- and household-level factors of dietary quality were analyzed using a linear regression model-building approach and concurrent thematic analysis. Phase 2 comprised the initial design of a BUDConnect interface, thematically-analyzed feedback from target consumers (N=20), and subsequent refinement.
Results: An alpha version of the BUD app was achieved to meet the needs of the urban food retailer-supplier system in Baltimore. The pilot demonstrated high acceptability and operability for corner store ordering and delivery of local, fresh foods. A concurrent nested triangulation analysis highlighted associations between consumer-level factors such as being Black, home ownership and WIC enrollment, and dietary quality. Consumer perceptions pointed to pandemic-related shifts in food sourcing and consumption. BUDConnect was conceptualized and iteratively refined based on local consumer’s’ desire for real-time crowd-sourced information, in-app reviews, ratings and rewards, and interoperability with existing digital tools.
Conclusions: This project is the first of its kind to work at multiple levels of an urban food system by building a digital network to support access for local healthy foods. Consumer-engaged approaches are crucial for improving food access and equity, and sustaining impact of interventions. Future steps will entail development and testing of the expanded app in urban and rural settings
Electrophysiological and Functional Imaging Approaches to SUDEP Mechanisms and Treatment
No full textSudden Unexpected Death in Epilepsy (SUDEP) is an often-unwitnessed mortality which occurs under benign circumstances in patients with epilepsy. Due to the sudden nature of SUDEP, the etiology behind SUDEP remain largely unknown. Recent work has suggested a mechanism of central respiratory control failure within the brainstem.
One of the only effective therapies to reduce SUDEP risk is vagus nerve stimulation (VNS), which involves stimulating the vagus nerve. However, the mode of action by which VNS leads to these outcomes is also unknown. This dissertation attempts to advance our understanding of SUDEP mechanisms and the dynamics of VNS-evoked central responses to begin addressing these unknowns.
The first contribution of this dissertation is a mechanistic interrogation of ventral respiratory circuits that may underly the breakdown in respiratory control during a SUDEP event using single microwire recordings. We focused on irregularities in the respiratory brainstem during apnea in seizure. We found that before seizure, gasps - which are the respiratory network’s attempt to restart inspiration after hypoxia - are tightly correlated with population activity in the solitary nucleus. However, after seizure, these gasps become decoupled and no longer fire concomitantly with solitary nucleus activity.
The second contribution of this dissertation used the Neuropixels multielectrode arrays to expand our examination of the respiratory brainstem circuits involved in SUDEP. We found significant changes to the ictal firing rate in several anatomic bands within the respiration. Notably, while apnea increased the firing rate of the solitary nucleus before seizure, these dynamics flipped during seizure, where apnea significantly inhibited the firing rate.
The third contribution of this dissertation is the adaptation and use of a holographic optical imaging system to visualize the functional effects of vagus nerve stimulation, which is one of the few factors through which SUDEP risk is mitigated. We find that VNS-evoked neural response dynamics are dependent on amplitude and pulse width, but not frequency; additionally, some responses demonstrate localization and differing temporal dynamics. We present noninvasive results with the future goal of using this imaging modality to further understand the broader respiratory brainstem in seizure
CHARACTERIZATION OF INCLUSION BODY MYOSITIS PROGENITOR CELLS USING IN VITRO METHODOLOGIES AND NEXT GENERATION SEQUENCING
Get PDFIn a mouse xenograft model of sporadic inclusion body myositis (IBM) depleted of T cells with anti-CD3 (OKT3), myofiber regeneration occurs normally, but myofibers retain degenerative pathologic features such as rimmed vacuoles and loss of TDP-43 function (Britson et al 2022). These findings have led us to hypothesize that newly regenerated myoblasts are genetically or epigenetically programmed to develop cell-autonomous pathology. Indeed, prior studies have suggested that cultured IBM myoblasts undergo premature senescence (Morosetti et al 2010). The goal of this study was to use transcriptomic, epigenomic, and in vitro methods to characterize regenerating myoblasts from IBM patient muscle biopsies compared with controls to better understand IBM pathogenesis. Muscle biopsy samples were obtained from patients who were diagnosed with IBM or other diseases and myoblasts were isolated using FACS. RNA was isolated from sorted muscle stem cells (MuSCs) for bulk RNA sequencing. In parallel, dissociated muscle cells were allowed to proliferate before myoblast isolation via flow cytometry and then differentiated into myotubes. Myotubes were stained for TDP-43, γH2AX, p16INK4A, and myosin heavy chain to investigate myotube maturation and degenerative phenotypes. Myotubes were also stained for beta-galactosidase to investigate senescence. Nuclei were isolated from these myotubes for 10X Genomics transcriptomic and epigenetic profiling. IBM myoblasts proliferate at a slower rate and generate fewer differentiated myotubes compared to non-IBM muscle. IBM myotubes show increased -galactosidase staining, suggesting senescence. IBM myotubes also show evidence of cryptic exon incorporation and loss of nuclear TDP-43. RNA sequencing data from both bulk and 10X sequencing show upregulated degenerative and immunological pathways in IBM myoblasts and myotubes compared with controls. Myoblasts can be directly isolated from human IBM muscle biopsies for multiomic analyses and differentiation into myotubes. However, cultured myoblasts from IBM patients proliferate at much slower rates and undergo early senescence compared with non-IBM controls, in agreement with prior studies. In the absence of immune cells, IBM myotubes show signs of degenerative pathology in vitro including the loss of nuclear TDP-43 and cryptic exon incorporation. This data suggests that cell-autonomous alterations in myofiber progenitor cells help drive IBM pathology