22689 research outputs found
Sort by
Understanding the Cellular Mechanisms of Parkinson's Disease: An Interactive Learning Module
No full textAs the second most common neurodegenerative disorder, Parkinson’s disease (PD) is still being thoroughly researched. The complex cellular mechanisms of PD are responsible for its many symptoms, such as tremors, difficulties with balance, and constipation. An immense amount of information about PD has been discovered, but much of it remains among the scientific research community. There is a need for resources to educate PD patients, caretakers, and students about specific reasons behind diagnoses and symptoms. In addition, existing resources also do not allow audiences to customize their learning process.
This project aims to provide an effective interactive learning module that allows users to choose the order and rate at which they absorb information. In particular, it aims to teach users about the cellular mechanism of parthanatos that causes many of the symptoms of PD. Using 3D models and animations, it was developed as a WebGL (Web Graphics Library) web application created with Unity. Cinema4D was used to convert molecular structure data into 3D models.
The goal of the interactive learning modules is to provide the first step in filling the gap in existing resources and educating all users about parthanatos, whether it is to supplement an existing knowledge background or start from the beginning. In future additions to the modules, other cellular pathways can be added to expand the educational content of the project
The fish, lake, and groundwater: Driving towards disparities reduction: Equitable approaches to social care at an integrated health system in Northern California
No full textIntroduction: The growing body of evidence shows associations between upstream social determinants of health and downstream health outcomes. Studies are increasingly documenting race and ethnic differences in social risk factors across all social determinants. Thus, healthcare industry stakeholders are joining the charge to support social needs. In response to these trends, the National Academies of Sciences, Engineering, and Medicine (NASEM) put forth a five-domain framework for health system integration of social care activities into the delivery of care.
Methods: Set within the Northern California Integrated Health System (NCIHS), this dissertation employed a mixed-methods design. Manuscript 1 is a literature review that examines multilevel social care programs influenced by the 2019 NASEM framework. Manuscript 2 sources equitable social care activities through qualitative interviews with NCIHS patients, clinicians, and operational managers. Manuscript 3 utilized the Modified Delphi Method to gain leadership consensus and prioritization of said equitable social care activities.
Results: Manuscript 1 describes 21 multilevel social care programs aligned with the NASEM framework. Findings questioned NASEM framework completeness, calling for it to address the family level of the socioecological model and more foundationally, structural determinants of health. In Manuscript 2, patients, frontline clinicians, and operational managers detailed equitable social care activities across the five domains of the NASEM framework: awareness, adjustment, assistance, alignment, and advocacy. Four equitable care principles emerged as foundational to the NASEM framework and called for: anti-racism, cultural responsivity, healing-centered engagement, and a whole-family approach in social care. Manuscript 3 included leadership review of 52 equitable social care activities and resulted in 20 activities gaining consensus for prioritization at NCIHS. Results of these inquiries informed the development of an "Equitable Social Care Integration Proposal" for NCIHS consideration.
Conclusion: NCIHS can lean on the "Equitable Social Care Integration Proposal" to strengthen its social care programming. The four equitable care principles now informing the NASEM framework help recalibrate social care initiatives with the equity lens. The collective action from our patients, clinical frontline, operational managers, and senior leaders detailed in the final proposal can help drive for race/ethnic disparities reduction and promote health equity for NCIHS’ diverse members and communities
Soluble lymphocyte activation gene-3 contributes to α-synucleinopathy and related therapeutic development
No full textPathogenetic α-synuclein (α-Syn) misfolded preformed fibrils (PFF) in Parkinson’s disease (PD) spreads from cell-to-cell in a prion-like manner has been evident in many studies. However, the spreading mechanism remains unclear. We have determined that α-Syn PFF but not α-Syn monomer, can specifically bind with transmembrane Lag3 (lymphocyte-activation gene 3), and the binding domain is the extracellular domain 1 (D1). Depletion of Lag3 can significantly inhibit pathogenic α-Syn cell-to-cell transmission. Membrane-bound Lag3 can be cleaved and soluble Lag3 (sLag3) can shed from the Lag3-expressing cell. Abundant sLag3 can be found in the plasma, however, the role of sLag3 in mediating α-synucleinopathy is completely unknown. Here, we used sLag3 lacking mice (membrane-bound Lag3 still exists) can reduce α-Syn pathology spreading from the gut to the brain, following by the discovery of facilitation of α-Syn PFF uptake in primary neuron by sLag3, which indicates the promoting role of sLag3 in the pathogenic α-Syn spreading. Moreover, we investigated the interaction of sLag3 to the immune system, in which we found both in vivo and in vitro that sLag3 facilitates the immune response. To explore the underlying molecular mechanism, we have identified several sLag3 binding receptors, which needs further validation. Insighted by the specific interaction between Lag3-D1 and α-Syn PFF, we further apply proteolysis targeting chimeric (PROTAC) technology to link a E3 ligase with the extracellular domain 1 (D1) of Lag3. The new agent, D1-E3 can significantly reduce α-Syn pathology induced by α-Syn PFF. In these studies, we determine that depletion of sLag3 can inhibit α-Syn gut-to-brain spreading and explored the underlying mechanism. By using the specific interaction with prion-like α-Syn, we show the efficacy of D1-E3 in clearance of α-Syn pathology, which provide new agents for therapeutic development against PD and related α-synucleinopathies
EFFECT OF SUBSTANCE USE ON RISK OF DEATH AND RECURRENCE IN HEAD AND NECK CANCER
No full textBackground
While predictors of outcome in head and neck cancer (HNC) have been extensively studied, less is known about whether substance use modifies the effect of these risk factors. This thesis studies the effect of substance use, including alcohol, tobacco, and marijuana, on risk of death and recurrence. This research also evaluates joint effect of substance use and 1) human papillomavirus (HPV) infection and 2) tumor subsite on overall and recurrence free survival time. This study aims to provide information on the role of substance use on HNC survival to aid in risk assessment and lifestyle recommendations.
Methods
448 incident HNC cases and a subset of 245 cases with oropharyngeal cancer (OPC) were identified from the Papillomavirus Role in Oral Cancer Viral Etiology (PROVE) study. HNC cases were enrolled from 2013 to 2018. Demographic, socioeconomic, and behavioral variables were evaluated, including age, ECOG performance status, race, sex, education, income, body mass index, HPV status, cancer stage, and substance-related variables. Cox proportional-hazards models were used to estimate hazard ratio (HR) of death. Competing-risks regression models according to Fine and Gray’s method were used for time-to-recurrence analyses. Violation of proportionality was assessed with complementary log-log plots and testing of scaled Schoenfeld residuals. Four multivariable models were explored to evaluate predictors of death and recurrence among two populations (all HNC cases and subcohort OPC cases).
Results
The findings suggested higher education level (≥ some college versus ≤ high school graduate, HNC: aHR=0.31, 95% CI: 0.14-0.66, p=0.003; OPC: aHR=0.11, 95% CI: 0.02-0.55, p=0.007) and tumor HPV positivity (HNC: aHR=0.29, 95% CI: 0.12-0.72, p=0.007; OPC: 0.04, 95% CI: 0.04-0.96, p=0.045) were associated with reduced risk of death in both HNC and OPC patients. In bivariate analysis, tobacco was associated with increased risk of death; marijuana was related to reduced risk of recurrence. However, these associations were not observed in multivariable models. No significant interaction was found between substance use and 1) HPV status and 2) tumor subsite for the risk of death and recurrence.
Conclusion
Among 448 HNC patients in the US, socioeconomic factor and HPV status (but not substance use) predicted risk of death
IMPROVING TRUST AND USE OF EDUCATION RESEARCH
No full textDespite over two decades of steady improvement in research on what works in education policy and practice, the evidence on how best to deploy these findings is still very weak. While anecdotal data and professional expertise are inherently valuable – use of evidence and professional expertise are not mutually exclusive. On the contrary, by relying on high quality evidence for decisions where the data exists, education practitioners can be freed to refocus their mental energy to other day-to-day decisions. The purpose of this present triad of studies is (1) to better understand the landscape of obstacles that impede research use in education, (2) to test the impact of a strategy to mitigate one such obstacle, and (3) to identify how publication bias inflates the effects of education outcomes. The findings herein indicate that trust is an underlying issue in research use, and that interventions may support improvements in research use if meta-analysts work to insure more reliable, uninflated results. This work may be used to guide research translation efforts, identify ways to increase replicability of effects, and ultimately to improve uptake of high-quality research in education
CENTROSOME LOSS IS SYNTHETIC LETHAL WITH AURORA KINASE A/TPX2 AXIS DYSFUNCTION
No full textCell division is an essential process for tissue/cellular growth and repair. However, rapid and uncontrolled cell division is a hallmark of cancer, making cell division an attractive target for cancer therapy. Unfortunately, the current drugs targeting this pathway are unable to distinguish malignant cells from rapidly proliferating healthy cells. Synthetic Lethality-based therapies allow for selectively targeting the cancer cells containing specific genomic perturbation while sparing the healthy ones. Polo-like kinase 4 inhibitor (PLK4i) prevents cell proliferation by inducing centrosome loss in cells and has been recently shown to induce synthetic lethality in cells with overexpression of a pericentriolar material (PCM) negative regulator. The presence of specific PCM scaffolds varies between cancer cell lines, and this is likely caused by the alterations in the pathways that regulate their formation. Those pathway alterations can render a cancer cell sensitive to PLK4i since PCM foci become crucial upon centrosome loss. Here we explored the mechanisms that confer PLK4i sensitivity in HCC1806 cells. We found that the acentrosomal microtubule organizing center (MTOC) in HCC1806 cells constitutes only nuclear mitotic apparatus (NuMA) and no other PCM scaffolding proteins. We showed that the knockdown of PPP6C, a phosphatase that negatively regulates mitotic kinase Aurora Kinase A (AURKA), can rescue cell division in acentrosomal HCC1806 cells through the increase in AURKA localization and activity on both the spindle and the acentrosomal poles. Moreover, AURKAi treatment resensitized HCC1806 PPP6C knockdown cells to centrosome loss. These results demonstrate that low AURKA activity is synthetically lethal with centrosome loss
CURA PERSONALIS: CATHOLIC HEALTH CARE SYSTEMS & THE MORAL OBLIGATION TO ADDRESS RELIGION IN CLINICAL ENCOUNTERS TO AVOID IATROGENIC HARMS
No full textGlobally, there are over 4000 recognized religions. Throughout the history of the Western Medical Tradition, religion/spirituality has played a significant role and continues to do so in a clinical context. In this paper, I argue that because a failure to address religion/spirituality in a clinical context may lead to iatrogenic harms in patients, all health care systems have a baseline level of duty to address religion/spirituality in their clinical settings. Furthermore, I argue that although both secular and faith-based health care systems have a baseline level of duty to address the role religion/spirituality may play in patient care, faith-based health care systems have a more stringent obligation as various components of faith traditions influence the harms patients encounter. In particular, I will focus on Catholic health care systems, utilizing examples in which Catholic beliefs must be addressed in order to mitigate potential harms
EXPLORING CHANGES IN TEACHERS’ SELF-EFFICACY AND CULTURAL COMPETENCE THROUGH A CULTURALLY RESPONSIVE PROFESSIONAL LEARNING MODEL
No full textThe national increase in student diversity within primary and secondary educational institutions is not reflected in the current faculty population of predominately White females. The difference in demographics between student and faculty population inspires this research study to shift focus from the gaps of student outcomes to the development of teachers’ efficacy and cultural competence when teaching students from diverse backgrounds. While teachers may receive multicultural education within their schooling, ongoing, job-embedded training provides the content focused, modeling, collaboration, reflection, and feedback required when creating a learning environment demonstrating awareness of and value for culturally diverse students. This research explores the training and development of teachers’ culturally responsive practices (CRP). Additionally, the research examines teachers’ self-efficacy in having the ability to acknowledge and demonstrate value in cultural differences while creating effective learning environments for culturally diverse students. This study addresses teachers’ cultural competence and self-efficacy in CRP within a K12 private, Southern college preparatory school, Southern Christian Academy (SCA). At SCA teachers are required to engage in job-embedded professional learning and the participants of this study (N = 30) selected the topic of CRP as the focus of their first professional learning rotation. The five-month, 3 session professional learning structure guided participants through an examination of their teacher self-efficacy and cultural competence as it relates to implementation of CRP. Analysis of the participants’ efficacy and competence in cultural knowledge, beliefs, and practices indicate significant increase in stages of self-identity and competence in dimensions of culturally responsive teaching. The qualitative journal reflection responses
iii
also indicated progression in teachers’ awareness of student culture. However, teachers’ general and personal self-efficacies did not significantly increase. While professional learning interventions help teachers develop skills and apply strategies for culturally responsive practice, implications for future research may include longer professional learning for extended opportunities to engage in verbal persuasion, vicarious experiences, mastery experiences, and physiological affect. Since implicit bias and institutional culture shape instructional practices and educational standards, ongoing professional learning opportunities will provide intentionally structured, content-specific training, for a sustained duration, to develop the skills necessary to address the needs of students, and teachers, in a culturally diverse classroom
RISK CONVERGENCE AND ALGORITHMIC REGULARIZATION OF DISCRETE-STEPSIZE (STOCHASTIC) GRADIENT DESCENT
No full textThis dissertation delves into the theoretical analysis of gradient descent (GD) and stochastic gradient descent (SGD), the fundamental algorithms for optimizing machine learning models, particularly in the context of deep learning. While these algorithms have demonstrated remarkable practical success, certain observed behaviors cannot be fully explained by classic optimization and statistical learning theories. This research aims to provide new theoretical insights into GD and SGD by analyzing their risk convergence and algorithmic regularization effects within modern settings that align with their applications in deep learning.
The first part of the dissertation focuses on the optimization behaviors of GD, specifically in scenarios where the stepsizes are large enough to cause non-monotonic decreases in the induced risks during the GD iterations. Our analysis centers on constant-stepsize batch GD applied to logistic regression on linearly separable data. Despite the potential presence of local oscillations, we prove that GD with any constant stepsize will ultimately minimize the logistic loss over an extended time scale. This theory offers a new understanding of the unstable risk convergence phenomenon commonly observed in practical implementations.
The second part of the dissertation investigates the generalization benefits of SGD in the overparameterized regime, where the number of model parameters exceeds the number of training samples. For overparameterized linear regression problems, we establish nearly matching upper and lower bounds on the excess risk for online SGD with geometrically decaying stepsizes. Our risk bounds reveal that SGD achieves a small excess risk if and only if an ``effective'' dimension is small compared to the sample size, while allowing for a large ambient dimension. Furthermore, we extend these results to learning an overparameterized single rectified linear unit (ReLU) neuron using a variant of the online SGD algorithm. These findings provide novel insights into the implicit regularization mechanism of SGD, enabling generalization even in overparameterized scenarios.
By investigating risk convergence and algorithmic regularization effects in modern deep learning applications, this dissertation sheds light on the theoretical foundations of GD and SGD, offering new perspectives and explanations for their observed behaviors
China’s High-Altitude Heliports: Examining PLA Helicopter Force Changes
No full textThe Chinese People’s Liberation Army (PLA) continues to expand its heliport infrastructure throughout the Western Theater Command (WTC). The Western Theater, particularly Tibet and Xinjiang, host the highest average elevation in the world. Extreme elevation, coupled with harsh geographic and environmental conditions, demand a robustly equipped, trained, and resourced vertical lift program to be effective. While the PLA has begun to build the infrastructure to support a budding domestic helicopter fleet designed to face these challenges and provide an alternative to Russian equipment, it is unclear if the PLA can employ its equipment and train its force to execute its vision. Furthermore, the peculiar design and location of its heliports raise new questions about PLA capabilities in the region