22691 research outputs found
Sort by
A ROBOTIC SYSTEM FOR STUDYING HAPTIC SHAPE PERCEPTION
No full textEvery object in the natural world has a three-dimensional shape, which we perceive through our senses of vision and touch. Despite major advances in our understanding of visual shape representation over the past several decades, our understanding of haptic shape encoding—especially at higher cortical areas—has severely lagged behind. This knowledge gap is largely due to a simple experimental challenge: it is hard to present large numbers of tightly-parameterized stimuli during grasping experiments. To overcome this technical barrier, we have developed a robotics-based system for high-throughput stimulus presentation for grasping studies in non-human primates (NHPs). In this work, we describe the design and implementation of the system, as well as demonstrate its use in a novel haptic shape comparison task. Our results show high-performance in discriminating shape, with errors being correlated to a neural network shape similarity score. Further, we present a software package for programmatic generation of large haptic and visual stimulus libraries. This work establishes a new experimental paradigm for studying haptic perception with considerably higher sampling density than has previously been possible. Through our system and accompanying software, we aim to facilitate further research into haptic shape perception, uncovering the brain’s ability to feel the world around it
Characterization of Vestibular Implant Outcomes in Humans
No full textPeople with loss of vestibular sensation in both ears experience diminished reflexes that normally maintain stable vision, posture, gait, and spatial perception. Standard-of-care treatment for bilateral vestibular hypofunction involves cessation of medications that could further suppress vestibular reflexes and recruiting non-vestibular central nervous system pathways like the cervico-ocular reflex to stabilize the eyes and head during motion. However, this proves inadequate for many patients. Prosthetic vestibular stimulation is an emerging alternative.
This work investigates the use of the Multichannel Vestibular Implant in people with severe to profound, adult-onset, bilateral vestibular loss for >1 year and inadequate compensation despite standard-of-care treatment. Active since 2016, this study is an ongoing prospective, nonrandomized, single-center cohort study conducted at Johns Hopkins Hospital in which participants serve as their own controls. As of January 2025, 17 individuals have undergone unilateral vestibular implantation and have been continuously receiving motion-modulated prosthetic vestibular nerve stimulation targeting their semicircular canals.
Prosthetic stimulation of the vestibular nerve was able to elicit reflexive eye movements that targeted participants’ implanted canals with specificity. Vestibular implantation was shown to be able to drive reflexive eye movements for participants with both ototoxic and non-ototoxic loss, 1-23 years of vestibular loss before implantation, and with the return electrode placed either intra-labyrinthine or outside the temporal bone in a subperiosteal pocket. Motion-modulated vestibular stimulation during head motion increased the gain of the vestibulo-ocular reflex during head impulses and during low frequency sinusoids. Additionally, vestibular implant recipients report vestibular symptom improvements not reported by a control group. These vestibulo-ocular reflex responses and patient-reported benefits support the use of vestibular implantation as a treatment for bilateral vestibular hypofunction
Investigation of Tissue Remodeling in Chronic Obstructive Pulmonary Disease
Full text linkChronic obstructive pulmonary disease (COPD) is a debilitating chronic disease of the lungs that impacts millions of people worldwide. Cigarette smoking is one of the leading risk factors for developing COPD, but chronic exposure to other ambient air pollutants is also associated with disease development. The lungs of patients with COPD undergo considerable tissue remodeling as the disease progresses leading to airway obstruction and difficulty breathing. Although COPD is a pressing and well described health risk, there are currently no disease modifying treatments that can reverse COPD associated tissue remodeling to restore optimal lung function. Innovative research is necessary to aid in the understanding of the processes driving tissue remodeling in COPD. In this thesis we explore two distinct angles of lung biology and disease to help enhance our understanding of COPD.
First, we used precision-cut lung slices (PCLS) exposed to cigarette smoke ex vivo to effectively model COPD. Our findings revealed that repeated, physiologically relevant doses of cigarette smoke mimic the tissue remodeling seen in COPD, offering a scalable and more time-efficient alternative to traditional in vivo models while maintaining the complex 3D architecture and cell-cell relationships absent in in vitro models. Second, we investigate the role of planar cell polarity in the process of epithelial wound closure using primary bronchial epithelial cells derived from COPD patients. Understanding the mechanisms of planar cell polarity can provide valuable insights into epithelial integrity and repair in the context of lung disease. Using primary bronchial epithelial cells from COPD patients, we identified mislocalization of the core planar cell polarity protein, Vangl1, and a significant reduction in the protein ARFRP1, which is known to target Vangl1 to the cell membrane. We propose that loss of ARFRP1 is a key driver in COPD associated epithelial disruption by limiting epithelial cell’s ability to sense their surroundings and migrate towards a wound
GLUTAMINE BLOCKADE STALLS TUMOR GROWTH AND REPROGRAMS TUMOR INFILTRATING MYELOID CELLS IN MOUSE MODELS OF SOFT TISSUE SARCOMAS
No full textThe endogenous immune system possesses the ability to eliminate tumors; however, this potential is often hindered by immunosuppressive mechanisms that tumors exploit. While immunotherapeutic strategies, such as immune checkpoint blockade, have successfully lifted some of these barriers, many patients fail to respond. Myeloid cells have been strongly implicated in this resistance across multiple cancer types, including sarcomas. These cells exhibit remarkable plasticity, enabling them to adapt to therapeutic interventions and sustain an immunosuppressive tumor microenvironment. Attempts to deplete myeloid cells or disrupt their function have largely been unsuccessful due to their resilience and ability to compensate for therapeutic pressures. We hypothesized targeting the metabolic dependencies of myeloid cells might steer them into an anti-tumor cell type. Glutamine antagonism, particularly through the novel prodrug JHU083, has emerged as a promising approach. By inhibiting glutamine metabolism, this strategy not only restricts tumor growth but also reprograms myeloid cells from an immunosuppressive to a tumoricidal phenotype, ultimately enhancing anti-tumor immune responses. In this study, we performed a comprehensive immune profiling of Undifferentiated Pleomorphic Sarcomas (UPS), a subtype of sarcoma that remains in dire need of effective therapeutic interventions. Our findings revealed a predominant immunosuppressive myeloid cell infiltrate, likely contributing to the poor efficacy of current immunotherapies. To overcome this resistance, we investigated the effects of glutamine blockade in a genetically engineered mouse model of immunotherapy-resistant soft tissue sarcomas. Treatment with JHU083 not only suppressed tumor growth but also led to significant changes in myeloid cell abundance, transcriptomic profiles, and metabolic states. Notably, our observations support that JHU083 reprogrammed these cells to support anti-tumor immunity rather than tumor progression, reinforcing the potential of metabolic interventions in overcoming myeloid-driven resistance. Overall, our findings support the clinical evaluation of glutamine antagonism, particularly in sarcomas and other malignancies characterized by myeloid cell enrichment. By reshaping the immune microenvironment, this approach holds promise for improving immunotherapy efficacy in historically resistant cancer
Autoencoder-based deep learning methods for single cell and spatial transcriptomics
No full textThe advent of single-cell and spatial transcriptomics has provided unprecedented resolution into cellular heterogeneity but introduced major computational challenges for data integration and interpretation. In this thesis, we develop and explore deep learning-based methods to address these challenges through the lens of autoencoder architectures.
In Chapter 1, we benchmark Variational Autoencoders (VAEs) against traditional linear decomposition methods, demonstrating that VAEs more effectively capture nonlinear biological variation in single-cell RNA-seq datasets. However, we also highlight inherent limitations in interpretability of VAE latent spaces and show through simulation studies that interpretability cannot be guaranteed without additional model constraints.
Building on these findings, Chapter 2 addresses the critical problem of cross-sample integration in spatial transcriptomics. Using the human dorsolateral prefrontal cortex dataset, we evaluate the performance of adversarial domain adaptation methods and show that standard domain classifier approaches may falter when scaling to multiple samples, motivating the need for more stable integration frameworks.
Finally, in Chapter 3, we introduce a novel autoencoder-based model that explicitly disentangles latent representations into common and dataset-specific components. We propose using the Sliced Wasserstein Distance as a stable and interpretable alternative to adversarial losses for multi-sample integration. Proof-of-concept experiments on synthetic datasets show that our approach successfully recovers and disentangles shared and residual signals, laying the groundwork for future applications to real-world biological data.
Overall, this thesis advances the application of deep learning for high-dimensional biomedical data analysis by proposing interpretable, stable, and scalable representation learning frameworks
Phenomenological shift in Bioethics Methodology
No full textAs the most used tool, principlism is a valuable and efficient methodology for bioethics. The scientific progress in medical care should mean there needs to be more tools in bioethics to better address people’s issues and reevaluate our understanding of care.
In this paper, I argue that phenomenology can help enrich the use of principlism to analyze ethical issues. Specifically, phenomenology can help us understand people’s embodied experiences, see how embodied experience may be shaped by social structures, and help healthcare providers to problematize their patients’ statements, which helps discover the hidden values related to their ethical decisions and reflect upon our own values and assumptions. Those methods could potentially provide us with a new perspective on the form of ethical questions healthcare providers should focus on and develop better understanding between doctors and patients
DIFFERENCES IN HEMOGLOBIN A1C 12 MONTHS AFTER DIRECT ACTING ANTIVIRALS TO TREAT HEPATITIS C VIRUS INFECTION IN ADULTS WITH HIV
Full text linkBackground: Previous studies suggested improved glycemic outcomes in hepatitis C virus
(HCV)-infected adults after direct-acting antivirals (DAAs) treatment. However, the impact of DAAs on glycated hemoglobin (HbA1c) levels in people with HIV/HCV coinfection is relatively unknown. This study aimed to determine the 12-month difference in HbA1c level between HIV/HCV coinfected adults on antiretroviral therapy (ART) who received a DAA prescription and those who did not receive a prescription in a large North American cohort of people living with
HIV.
Methods: The study was nested in the NA-ACCORD, a collaboration of observational HIV clinical cohort studies. Adults (>18 years old) with HIV/HCV coinfection who were on ART and prescribed DAAs between January 1, 2014, and December 31, 2022 ("case") were individually matched by year of observation, 10-year age group, sex, and race and ethnicity with one HIV/HCV coinfected participant who was under observation and not prescribed DAAs within one year of the DAA prescription date for the case (baseline). Multivariable linear regression models were used to evaluate the difference in HbA1c levels 12 months post-DAA prescription after
adjusting for baseline demographic and clinical characteristics, including baseline HbA1c.
Results: The 1,149 individuals with HIV/HCV coinfection and DAA prescription were matched to
1,100 similar individuals without DAA prescription (N= 2,249). The median HbA1c values in individuals with DAA prescription and those without DAA prescription at 12 months were 5.7% (interquartile range [IQR]: 5.3%-6.3%) and 5.6% (IQR: 5.2%-6.2%), respectively. There was no difference in the average HbA1c value at 12 months by DAA prescription (adjusted mean difference = 0.01%, 95% confidence interval: -0.09%, 0.08%) after adjusting for age, baseline
CD4 counts, baseline HbA1c, race and ethnicity, HIV risk groups, history of statin use, diabetes, chronic kidney disease > stage 3, and clinical cohorts; further controlling for sex did not change the results.
Conclusions: The study did not find a significant difference in HbA1c 12 months after DAA prescription in HIV/HCV coinfected adults on ART with DAA prescription compared to similar adults without DAA prescription. Studies investigating DAA's long-term (>12 months) effects on diabetes and cardiovascular risk factors are needed and may yield different findings
ASSOCIATION BETWEEN METFORMIN USE AND MORTALITY AMONG INDIVIDUALS WITH NON-SMALL CELL LUNG CANCER RECEIVING IMMUNE CHECKPOINT INHIBITORS: A RETROSPECTIVE COHORT STUDY
Full text linkIntroduction. Metformin has potential to synergistically enhance the effect of immune checkpoint inhibitors (ICI) in non-small cell lung cancer (NSCLC). We evaluated the association between metformin use prior to ICI initiation and cancer-specific and all-cause mortality among NSCLC patients.
Methods. We conducted a retrospective cohort study using the Surveillance, Epidemiology, and End Results (SEER)-Medicare data (2013 to 2019), including NSCLC patients with type 2 diabetes who newly initiated ICI therapy and had prior antidiabetic medication use. The exposure was metformin monotherapy versus sulfonylurea and/or dipeptidyl peptidase-4 (DPP-4) inhibitors. The primary outcome was cancer-specific mortality, and the secondary outcome was all-cause mortality. We used stabilized inverse probability of treatment weighting (sIPTW) to adjust for confounders. Fine-Gray competing risk model estimated cancer-specific mortality, while Cox proportional hazards model evaluated all-cause mortality.
Results. We included 1,123 metformin users and 362 sulfonylurea/DPP-4 users. Although baseline characteristics differed, groups were well balanced after weighting. The adjusted incidence rate (aIR) of cancer-specific mortality was 82 vs. 81 (aIR difference = 1, 95% CI: -13 – 16), and all-cause mortality was 71 vs. 67 (aIR difference = 4, 95% CI: -6 – 15) per 100 person-years for metformin and sulfonylurea/DPP-4 users, respectively. Metformin use was not significantly associated with cancer-specific mortality (adjusted hazard ratio (aHR) = 1.08, 95% CI: 0.88–1.33) and all-cause mortality (aHR = 1.07, 95% CI: 0.90–1.26).
Conclusions. In this large, diverse cohort of individuals with NSCLC using ICI, there was no statistically significant association between metformin use and cancer-specific or all-cause mortality
3D Self-Folding Shell Microfluidic Arrays (SMFAs) for Spatiotemporal Chemistry
No full textConventional microfluidic systems have traditionally been confined to planar, two-dimensional geometries, limiting their ability to generate complex chemical patterns in truly 3D contexts. This limitation becomes particularly significant when studying spherical biological models such as organoids, which have emerged as powerful tools for investigating developmental processes and disease mechanisms. To address this challenge, we have designed and fabricated biocompatible self-folding shell microfluidic arrays (SMFAs) with embedded microchannels that can encapsulate and interface with 3D spherical structures. Using conventional photolithography, we created gradient cross-linked SU8 films with embedded sacrificial fillers that dissolve to form self-folded microchannels around spherical objects. Through a combination of computational fluid dynamics simulations and experimental validation, we demonstrate the versatility of our system in generating chemical gradients, performing multi-chemical patterning, and facilitating complex chemical reactions within agarose gel beads. Moreover, we establish the biocompatibility of SMFAs and their potential for biological applications through successful spatiotemporal patterning of live stem cell-derived neural organoids. By enabling precise control over the 3D chemical microenvironment of spherical objects, this work provides a versatile platform for investigating fundamental questions in developmental biology, advancing tissue engineering approaches, and creating novel patchy particles with complex 3D chemical patterns
Exploring the Epigenetic Regulation Using a CRISPR Edited Cell Line
No full textThe ATRX (Alpha Thalassemia/Mental Retardation X-linked) protein is a critical chromatin remodeler with multifaceted roles in genome maintenance. This study explores ATRX’s functional domains, emphasizing its ability to recognize repressive histone marks and restructure chromatin through its SWI/SNF helicase activity. Following sessions highlight ATRX’s role in silencing transposable elements via H3.3 deposition, as well as its contributions to replication stress resolution and DNA repair, where it stabilizes stalled forks and suppresses aberrant recombination. To investigate ATRX’s biological significance, I generated an ATRX knockout (KO) cell line, validated by Western blot (WB) and immunofluorescence (IF). These findings underscore ATRX’s importance in maintaining genomic stability and cellular homeostasis. Our work provides insights into the molecular mechanisms underlying ATRX-related disorders, including neurodevelopmental syndromes and cancer