22689 research outputs found
Sort by
ARIC COMMUNITY SURVEILLANCE: SOCIODEMOGRAPHIC AND TEMPORAL DIFFERENCES IN INVASIVE CORONARY ANGIOGRAPHY USE ACCORDING TO RISK PROFILE AMONG PATIENTS WITH NSTEMI
No full textBackground: Variability exists in the use of invasive coronary angiography (ICA) for patients presenting with non-ST elevation myocardial infarction (NSTEMI). This dissertation study characterized ICA use among patients from the Atherosclerosis Risk in Communities (ARIC) surveillance study presenting with NSTEMI and examined differences in ICA use by race, sex, and rurality by validated risk scores over time.
Methods: GRACE and TIMI scores were calculated and rates of ICA use were determined for patients hospitalized for NSTEMI from 2004-2014. Data was stratified by risk score category, and subgroup analyses by ICA use status were performed to describe the overall sample. We then performed logistic regression to test if higher risk scores are associated with ICA use; if the relationship between risk score and ICA use differed by race, sex, and rurality; and if ICA use differed according to time.
Results: The sample consisted of 11,501 weighted hospitalizations (Age=65 years; 35% Black; 44% female; 13% rural ) for NSTEMI, and 49% of patients underwent ICA. Patients stratified as intermediate and high risk by GRACE were less likely to undergo ICA as compared to low-risk patients (aOR=0.75 and 0.33 respectively; P<0.05) whereas patients with intermediate or high TIMI risk scores were more likely to undergo ICA (aOR=10.29 and 10.47, respectively; p<0.05). ICA use was lower in Black persons (GRACE: 59%; TIMI:40%; p<0.001) and patients residing in rural communities (GRACE:42%; TIMI:27% p<0.005). ICA use was lower for female patients when compared by TIMI score (34%; p<.02). Differences in ICA use between Black and White patients decreased as risk score increased (p<0.05). Overall, no significant association was evident with respect to time although the study was underpowered to fully explore intersectionality.
Conclusion: ICA use decreased as GRACE score increased, while the inverse was true with the TIMI score. ICA use was inequitable by race and rurality, and disparities in ICA use with respect to race decreased as risk score increased. No difference in ICA use was noted after 2007. Incorporation of risk scores may increase ICA use in a manner consistent with NSTEMI guidelines and narrow gaps in equitable healthcare delivery that have persisted
Predicting Adverse Mental Health Outcomes after Adolescent Psychotic Experiences
No full textBackground: Psychotic experiences are common in the general population, have peak incidence in adolescence, and are associated with multiple adverse mental health outcomes.
Objective: To predict the risk of generalized anxiety disorder, psychotic disorders, and past-year suicide attempts in young adulthood amongst adolescents who report subclinical psychotic experiences. To this end, there are three aims. Aim one estimates the comparative risk of mental disorders and past-year suicide attempts in young adulthood between participants who reported psychotic experiences in adolescence and those who did not, as well as between participants who reported persistent versus transient psychotic experiences. Aim two characterizes latent trajectories of adolescent psychotic experiences and estimates those trajectories’ risk of mental disorders and past-year suicide attempts in young adulthood. Aim three builds and compares predictive models for mental disorders and past-year suicide attempts using machine learning methods and traditional logistic regression.
Methods: Participants are from the birth cohort study the Avon Longitudinal Study of Parents and Children. 8822 participants completed at least one measurement of psychotic experiences between ages 11-16 years. Aim one uses survival analyses to estimate the risk of mental disorders and suicide attempts as a function of adolescent psychotic experiences. Aim two identifies latent trajectories of adolescent psychotic experiences and estimates differences in prevalence of mental disorders and suicide attempts in young adulthood comparing latent trajectories. Aim three compares predictive accuracy of machine learning methods to traditional logistic regression.
Results: Those with psychotic experiences during adolescence had greater risk of subsequent mental disorders and suicide attempts in young adulthood. Five distinct trajectories were identified. Trajectories with increased probability of endorsing psychotic experiences at multiple time points were at highest risk for later mental disorders and suicide attempts. Machine learning methods did not improve upon predictive accuracy of traditional statistical methods.
Conclusions: Psychotic experiences increase the risk for mental disorders and suicide attempts, suggesting psychotic experiences are a non-specific risk factor for psychopathology. Establishing the timing and persistence of adolescent psychotic experiences can help determine who is most at risk of subsequent mental disorders and past-year suicide attempts in young adulthood
EFFECTS OF A SCHOOL-BASED MINDFULNESS INTERVENTION ON STRESS AND EMOTIONS ON STUDENTS ENROLLED IN AN INDEPENDENT SCHOOL
No full textStudents enrolled in high-achieving schools are under tremendous pressure to perform at high levels inside and outside the classroom. Achievement pressure is a prevalent source of stress for students enrolled in high-achieving schools, and female students in particular experience a higher frequency and higher levels of stress compared to their male peers. The practice of mindfulness in a school setting is one tool that has been linked to improved self-regulation of emotions, increased positive emotions, and stress reduction. This study, a mixed methods randomized pretest-posttest no-treatment control trial, evaluated the effects of a six-session mindfulness intervention taught during a regularly scheduled life skills period in an independent day school, one type of high-achieving school. Twenty-nine students in Grades 10 and 11 were randomized by class where Grade 11 students were in the intervention group (n = 14) and Grade 10 students were in the control group (n = 15). Findings from the study produced mixed results. There was no evidence that the mindfulness program reduced participants’ stress levels and negative emotions. In fact, contrary to what was expected, students enrolled in the intervention group experienced higher levels of stress and increased negative emotions at posttreatment when compared to pretreatment. Neither the within-group nor the between-groups changes in stress level were statistically significant, p < .05, and the between-groups effect size was small, d = .2. The study found evidence that the mindfulness program may have had a positive impact on students’ ability to regulate their emotions. The within-group comparison and the between-groups comparison at posttreatment found that students in the mindfulness course experienced statistically significant improvement in the in their ability to regulate their emotions at posttreatment, p = .009 < .05 and p =. 034 < .05, respectively. The between-groups effect size was medium, d =.7, suggesting that the positive differences in emotion regulation difficulties were substantial and have practical implications. The analysis of gender differences as they relate to stress and emotions revealed that female students perceive higher levels of stress and report experiencing stress more often than males. There were no gender differences when analyzing sources of stress experienced by the student participants. Both females and males experience regular achievement pressures related to their school performance and worry about their future, college acceptance, grades, and parental expectations. Females reported an increased awareness of their stress and actively engaged in practicing mindfulness to manage their stress. Students in the treatment group expressed that the practice of mindfulness resulted in feelings of relaxation and calmness
SURVIVAL AND GROWTH OF NONPROFIT ORGANIZATIONS: A MIXED METHODS STUDY ON SMALL, RESEARCH FOCUSED NONPROFITS IN WASHINGTON, DC.
Get PDFNonprofit organizations (NPOs) are created to carry out a beneficent purpose. Unlike the private sector, nonprofit organizations are purpose-driven so that profit is not the primary goal. Nevertheless, nonprofits require financial resources to operate and carry out their mission. The more financial resources a nonprofit has at its disposal the more mission-oriented success it can achieve. Nonprofits fundraise from a variety of sources and methods, including individual donations, government contracts, corporate giving, and many other sources. Although an abundance of literature exists on nonprofit management and governance, there is little on fundraising management, revenue optimization, and the drivers of financial success for nonprofits. This is surprising given the wealth of research focused on growing revenue and income in the private sector. The goal of this study was to investigate small, research-focused nonprofits based in Washington D.C. that significantly increased revenue in a short time. Qualitative and quantitative data was utilized in combination with mixed methods. Financial data alone was not enough to understand how nonprofits with substantial increases in annual revenue were able to achieve such fundraising success. Semi-structured interviews with nonprofit leaders were conducted to learn about fundraising strategies and best practices for income growth. The mixed methods of this two-part study revealed several common strategies and best practices among small, research-focused nonprofits in Washington, D.C. However, it was also apparent that each nonprofit required an individualized approach to fundraising based on the unique needs and structure of the organization
AN INVESTIGATION INTO POTENTIAL NON-ANTIBIOTIC CONTRIBUTORS TO ANTIMICROBIAL RESISTANCE AND NON-ANTIBIOTIC THERAPEUTICS FOR ANTIBIOTIC-RESISTANT E. COLI
No full textAntimicrobial resistance is a threat to human health worldwide. It is estimated that by the year 2050, up to ten million deaths every year will be due to antibiotic-resistant bacterial infections and antimicrobial resistance. Many non-antibiotic drugs have been shown to have antibacterial properties, but it is unknown whether non-antibiotic drugs can contribute to antimicrobial resistance. They may also serve as alternative therapeutics for resistant infections.
In this work, we investigated the potential for non-antibiotic drugs to contribute to antibiotic resistance. We examined the antibacterial properties of antiviral drugs in vitro and demonstrated that repeated antiviral exposure can result in antibiotic cross-resistance in Escherichia coli and Bacillus cereus. Whole genome sequencing of antiviral-resistant E. coli revealed that antiviral exposure led to mutations in genes with known roles in antimicrobial resistance.
We explored the repurposing potential of some non-antibiotic drugs using clinical E. coli isolates. Our results showed that two drugs in particular—antiviral nucleoside analog zidovudine and antineoplastic pyrimidine analog fluorouracil—were effective at inhibiting the growth of multidrug-resistant E. coli within relevant physiological concentration ranges of these drugs. In an exploration into the potential for bacteriophages (phages) to serve as non-drug treatments for antibiotic-resistant bacteria, we isolated phages from environmental soil samples and challenged clinical E. coli isolates. Even multidrug-resistant E. coli were susceptible to the isolated phages. To explore the role that non-antibiotic drugs may play in either augmenting or impeding the efficacy of phage therapy, we treated E. coli with the antiviral zidovudine and demonstrated that zidovudine exposure altered phage susceptibility.
To better understand the presence and concentrations of non-antibiotic drugs such as antivirals in environmental water, we quantified antiviral drugs in wastewater samples and assessed their removal through wastewater treatment. We conducted acute aquatic toxicity tests using luminescent Aliivibrio fischeri and concluded that several non-antibiotic drugs exhibit high acute aquatic toxicity and can have additive effects as mixtures.
These findings indicate that non-antibiotics not only have the potential to contribute to antimicrobial resistance and acute aquatic toxicity but also may provide alternative therapeutics to combat antimicrobial resistant infections, all of which warrant further investigation
The Illusion of the “Public” Conservatory: A Contextual Analysis of the Peabody Institute’s Founding
Get PDFAn analysis of the early development of the Peabody Conservatory from its founding in 1857 until the end of Asger Hamerik’s directorship in 1898 in the context of public perception. The founding letter by George Peabody suggests that the Peabody Institute would become a space accepting of anyone who wished to receive a serious, elevated education in the fine arts, regardless of gender or social background. As the new Conservatory grew, however, its leadership developed a more elitist institution modeled after European conservatories rather than a “popular” one for people of Baltimore without prior musical education. The paper situates the Peabody Conservatory within the broader context of certain historical themes of 19th -century America like such as moral reform and social control. It concludes with commentary on a struggle within classical music today, the urge to “distance itself from the public, while still encouraging new audiences to attend performances.
Investigating Metabolic Pathways and Proteins Important for Fueling Sporozoite Motility
No full textMalaria persists as a major life-threatening disease, owing to the evolution of drug resistant parasites. Our ability to develop new drugs is dependent on the discovery of new parasite-specific biological targets. In general, antimalarial drugs and drug development have primarily focused on blood stage parasites. However, to block malaria transmission, targeting the infective stage of the parasite—sporozoites, is required. A key characteristic of sporozoites is their capacity to perform gliding motility. Without motility, sporozoites would be unable to exit the inoculation site and initiate infection, thereby highlighting the transmission-blocking potential of targeting motility. Like all living organisms, sporozoites need to produce and use energy to move. In this study, we sought to characterize metabolic pathways that are important for producing the energy that fuels sporozoite motility. Using a moderate throughput motility assay, we found that sporozoites can move in the absence of exogenous carbohydrates, and that P. berghei and P. falciparum motility can be abrogated by both oxidative phosphorylation and glycolysis inhibitors. Moreover, sporozoites express Plasmodium berghei vacuolar pyrophosphatase 2 (PbVP2) and vacuolar ATPases (V-ATPases) to pump protons across membranes. However, treatment with vacuolar pyrophosphatase and ATPase inhibitors did not prevent sporozoites from moving. To further elucidate the role and importance of PbVP2, we developed a PbVP2 knockout line. Lastly, we show that motility is significantly inhibited in the presence of fatty-acid free BSA, indicating that fatty acids are also important motility mediators. Our findings bring us closer to understanding sporozoite motility and which metabolic pathways have potential to be targeted in a transmission-blocking approach
Controllable Simulation of Deformable Motion in Cone-Beam Computed Tomography with Generative Learned Models
No full textCone-beam computed tomography (CBCT) has emerged as an omnipresent instrument for guidance in interventional radiology over the past two decades. However, it is susceptible to patient movement, manifested by artifacts and morphological distortions, which stem from a blend of quasi-periodic global factors, such as respiratory motion, and irregular, localized sources, including peristalsis. Recent advances in deep autofocus techniques have demonstrated potential in deformable motion compensation utilizing learned metrics, yet these methods depend on the accessibility of extensive and varied datasets comprising motion-corrupted CBCT, coupled with the originating motion field. The credible emulation of such intricate motion continues to pose a formidable challenge. This study presents a generative model for simulating realistic deformable motion of adjustable magnitude, through unsupervised training using the observation of unpaired, motion-corrupted data.
Evaluation of this model in a proof-of-concept study demonstrated three main features of the generated motion fields. First, the agreement between the spatial distribution of the synthetic motion and of the controlled training data; specifically, the magnitude of motion and anatomical distribution showed good agreement, and anatomical structures that consistently remained static in the training set were spared in the generated synthetic motion fields. Second, the prevalent direction of motion showed good agreement with trends in the training data. Furthermore, the proposed model was able to synthesize controlled amplitude of motion while yielding random deformable motion trajectories preserving consistency with the source anatomy
Understanding and Optimizing Communication Overhead in Distributed Training
No full textIn recent years, Deep Learning models have shown great potential in many areas, including Computer Vision, Speech Recognition, Information Retrieval, etc. This results in a growing interest in applying Deep Learning models in academia and industry. Using Deep Learning models on a specific task requires training. With the recent trends of the rapid growth of the size of the Deep Learning models and datasets, training on a single accelerator can take years. To complete the training within a reasonable amount of time, people start using multiple accelerators to speed up training (i.e., distributed training). Using distributed training requires additional communications to coordinate all accelerators. In many cases, communications become the bottleneck of distributed training. In this thesis, we study and optimize the communication overhead in distributed training.
In the first part of the thesis, we conduct measurement studies and what-if analyses to understand the relationship between the network and communication overhead. We design a trace-based simulation algorithm and test it with various network assumptions. We found that the network is under-utilized, and achieving gradient compression ratios up to hundreds of times is often unnecessary for data center networks.
The second part of the thesis optimizes the communication overhead of distributed training without changing the semantics of the training algorithm. We design and implement system MiCS that significantly reduces the communication overhead in public cloud environments by minimizing the communication scale. The evaluation shows that MiCS outperforms existing partitioned data-parallel systems significantly.
In the last part of the thesis, we further improve the system performance of MiCS for more challenging cases, e.g., long input sequences. We combine pipeline parallelism with MiCS to further reduce the overhead of inter-node communications in MiCS. Besides, we propose two memory optimizations to improve memory efficiency. System MiCS has been adopted by several teams inside Amazon and is available at Amazon SageMaker
Gene dynamics of maturation in endogenous and pluripotent stem cell-derived cardiomyocytes
No full textA primary limitation in the clinical application of pluripotent stem cell-derived cardiomyocytes (PSC-CMs) is the failure of these cells to achieve full functional maturity. In vivo, cardiomyocytes undergo numerous adaptive changes during perinatal maturation. By contrast, PSC-CMs fail to fully undergo these developmental processes, instead remaining arrested at an embryonic stage of maturation. To date, however, the precise mechanisms by which directed differentiation differs from endogenous development, leading to consequent PSC-CM maturation arrest, are unknown. The advent of single cell RNA-sequencing (scRNA-seq) has offered great opportunities for studying CM maturation at single cell resolution. However, postnatal cardiac scRNA-seq has been limited owing to technical difficulties in the isolation of single CMs. Additionally, cross-study comparison is limited by dataset specific batch effects. In this dissertation, I first established large particle fluorescence-activated cell sorting (LP-FACS) for isolation of viable single adult CMs. I secondly developed transcriptomic entropy as a robust, batch effect-resistant approach to quantifying CM maturation. With these and other computational tools, I investigated gene expression trends in endogenous and PSC-derived CMs. I first generated an scRNA-seq reference of mouse in vivo CM maturation with extensive sampling of perinatal time periods. I subsequently generated isogenic embryonic stem cells and created an in vitro scRNA-seq reference of PSC-CM directed differentiation. Through computational analysis, I identified a perinatal iimaturation program in endogenous CMs that is poorly recapitulated in vitro. By comparison of these trajectories with previously published human datasets, I identified a network of nine transcription factors (TFs) whose targets are consistently dysregulated in PSC-CMs across species. Notably, I demonstrated that these TFs are only partially activated in common ex vivo approaches to engineer PSC-CM maturation. This dissertation represents the first direct comparison of CM maturation in vivo and in vitro at the single cell level. Moreover, the findings and tools developed here can be leveraged towards improving the clinical viability of PSC-CMs