University of California, Irvine
  • Irvine, CA, United States
Recent publications
Background Acute respiratory distress syndrome (ARDS) is common in intensive care units with high mortality rate and mechanical ventilation (MV) is the most important related treatment. Early prediction of MV duration has benefit for patients risk stratification and care strategies support. Objective To develop an explainable model for predicting mechanical ventilation (MV) duration in patients with ARDS using the machine learning (ML) approach. Method The number of 1,148, 1,697, and 29 ARDS patients admitted to intensive care units (ICU) in the MIMIC-IV, eICU-CRD, and AmsterdamUMCdb databases were included in the study. Features at MV initiation from the MIMIC-IV dataset were used to train prediction models based on seven supervised machine learning algorithms. After 5-fold cross-validation for hyperparameters tuning, the hyperparameters- optimized model of different algorithms was tested by external datasets extracted from eICU-CRD and Amsterdamumcdb. Finally, three descriptive machine learning explanation methods were conducted for the model explanation. Result The XGBoosting model showed the most stable and accurate performance among two testing datasets (RMSE= 5.57 and 5.46 days in eICU-CRD and AmsterdamUMCdb) and was selected as the optimal model. The model explanation based on SHAP, LIME, and DALEX results showed a consistent result, vasopressor, PH, and SOFA score had the highest effect on MV duration prediction. Conclusion ML models with features at MV initiation can accurate predict MV duration in patients with ARDS in ICUs. Among seven algorithms, XGB models showed the best performance (RMSE= 5.57 and 5.46 in two external datasets). LIME, SHAP, and Breakdown methods showed good performance as AXI methods.
Introduction During the emergent treatment of violently injured patients, law enforcement (LE) officers and health care providers frequently interact. Both have duties to protect patient health, rights, and public health, however, the balance of these duties may feel at odds. The purpose of this study is to assess hospital-based violence intervention program (HVIP) representatives’ experiences with LE officers among survivors of violence and the impact of hospital policies on interactions with LE officers. Materials and methods A nationwide survey was distributed to the 35 HVIPs that form the Health Alliance for Violence Intervention. Data regarding respondent affiliation, programs, and perceptions of hospital policies outlining LE activity were collected. Follow-up video interviews were open coded and qualitatively analyzed using grounded theory. Results Respondents from 32 HVIPs completed the survey (91%), and 22 interviews (63%) were conducted. Common themes from interviews were: police-patient interactions; racism, bias, and victims' treatment as suspects; and training and education. Only 39% of respondents knew that policies existed and were familiar with them. Most representatives believed their hospitals’ existing policies were inadequate, ineffective, or biased. Programs that reported good working relationships with LE officers offered insight on how their programs maintain these partnerships and work with LE officers towards a common goal. Conclusions Unclear or inadequate policies relating to LE activity may jeopardize the health and privacy of violently injured patients. Primary areas identified for improvement include clarifying and revising hospital policies, education of staff and LE officers, and improved communication between health care providers and LE officers to better protect patient rights.
Introduction Early initiation of chemotherapy after surgery for colon cancer has survival benefits. Immediate adjuvant chemotherapy (IAC) involves giving chemotherapy during surgical resection and immediately postoperatively. This novel approach has been shown to be safe, eliminating delays in adjuvant treatment that could increase the risk of micro-metastatic spread. The aim of this study was to assess the willingness of the general public to accept IAC. Materials and methods Between March and April 2021, 800 telephone interviews were conducted with a sample of adult New York State residents. The Survey Research Institute of Cornell University conducted all surveys. Kruskal–Wallis, chi-squared, and Fisher’s tests were conducted using R 4.0.2. Results Three scenarios were presented: (1) receiving IAC resulting in improved survival and quality of life, (2) finishing chemotherapy earlier without survival impact, and (3) finishing chemotherapy earlier but with possible side effects. Respondents with higher education were more likely to accept (1) & (2), males were more likely to accept (2) & (3), higher income respondents were more likely to accept (1) & (3), and those with more work hours were more likely to accept (2). Lastly, 16% responded they would be very or extremely likely, and 52% respondents would be somewhat likely or likely to accept intraoperative chemotherapy, even if it may not be necessary. Conclusions Respondents were likely to accept IAC if offered. Given the known risk of delayed adjuvant chemotherapy (AC) in colon cancer, further research is warranted to determine the survival and quality of life (QOL) benefits of IAC.
Detection of the QRS complex is the most important step in analyzing ECG signals for heart monitoring and diagnosis. There have been several QRS-peak detection methods reported in the literature. Most of these methods have low performance under noisy conditions. In this paper, we propose a novel QRS detection algorithm based on a new Permutation Entropy (PE) method that we developed and referred to as the Adaptive Improved Permutation Entropy (AIPE) method. The parameters of the AIPE method are determined based on the specific signal properties. Implementing the AIPE method leads to prominently preserving the QRS complex and eliminating noises of the ECG signal without smoothing the ECG signal. Our simulations show that the proposed QRS detection algorithm is effective and robust under noisy conditions. The algorithm is validated on the MIT-BIH Noise Stress Test Database for various SNR values. In addition, we examined the algorithm’s performance under motion noise conditions, mimicking a practical scenario. We used the metrics of sensitivity, positive predictive, and F1 score to evaluate the performance of our algorithm and compare it with several other algorithms explained in the literature. Our investigation shows that the proposed algorithm is efficient and effective. More importantly, it is robust under noisy conditions providing superior performance over other recent and popular QRS detection algorithms, including the popular Pan–Tompkins and the recent Advanced Adaptive Multilevel Thresholding (AAMT) algorithms.
Following the first stay-at-home order in March 2020, many Californians responded with panic buying: they stocked up on masks, and hoarded toilet paper, hand sanitizer, canned food, bread, and pasta. Restaurants had to stop on-site dining services, and some even closed permanently. Californians adapted to the evolving restrictions imposed by the pandemic by switching to alternative channels for their groceries and experimenting with meal deliveries from participating restaurants. The purpose of this chapter is to analyze the impacts of the COVID-19 pandemic on how Californians shopped for groceries and prepared meals before and during the pandemic based on a random survey of 1,026 Californians conducted at the end of May 2021. To better contextualize observed changes in California, we also investigated grocery and prepared meals purchases in China and South Korea, two countries at the forefront of online grocery shopping and meal deliveries that provide a window into possible alternative futures for e-grocery and meal deliveries in California. We conclude by reflecting on how our increasing dependence on online grocery shopping and meal deliveries may impact travel.
This chapter examines changes in telecommuting and the resulting activity-travel behavior during the COVID-19 pandemic, with a particular focus on California. A geographical approach was taken to “zoom in” to the county level and to major regions in California and to “zoom out” to comparable states (New York, Texas, Florida). Nearly one-third of the domestic workforce worked from home during the pandemic, a rate almost six times higher than the pre-pandemic level. At least one member from 35% of U.S. households replaced in-person work with telework; these individuals tended to belong to higher income, White, and Asian households. Workplace visits have continued to remain below pre-pandemic levels, but visits to non-work locations initially declined but gradually increased over the first nine months of the pandemic. During this period, the total number of trips in all distance categories except long-distance travel decreased considerably. Among the selected states, California experienced a higher reduction in both work and non-workplace visits and the State’s urban counties had higher reductions in workplace visits than rural counties. The findings of this study provide insights to improve our understanding of the impact of telecommuting on travel behavior during the pandemic.
Hotel room cleaners are a vulnerable population at risk for cardiovascular disease. To evaluate their workload heart rate (HR), % heart rate reserve (%HRR), blood pressure (BP), metabolic equivalent (MET), and energy expenditure (EE) were measured over two workdays and two off-workdays. The mean age was 45.5 (SD 8.2) years with a mean 10.4 (SD 7.8) years of work experience. Mean average and peak HR, %HRR, MET, and EE were significantly higher during a workday than an off-workday for the entire work shift, first and last hour of work. Mean average HR and %HRR saw the largest increase between the lunch and post-lunch interim. One-fourth of subjects exceeded the recommended 30% HRR threshold for 8-hour shifts. Some workers experienced a substantial increase in HR and DBP over a workday indicating physiologic fatigue and thus may be at increased risk for cardiovascular disease and premature death due to excessive physical work demands.
As discussed throughout this volume, febrile seizures (FS) remain the most common seizures in infants and children worldwide. This fact has provided the impetus to study them and their consequences and consider their treatment, the focus of the first edition of this book (Baram and Shinnar, 2002 [1]). The 20 years since the publication of this first edition have witnessed an explosion of new information about FS, meriting this new edition. Key advances have occurred in the genetics and neurobiological underpinnings of FS and febrile status epilepticus (FSE), the role of neuroinflammatory factors in the emergence of FS and their consequences, the demonstration of unique clinical and neuroradiological aspects of FSE, and the prospect of predictive (bio)markers to identify and characterize cognitive and pro-epileptogenic outcomes. Here, we review some of these developments and speculate about the next 20 years of the field.
The contribution of neuroinflammatory processes to epilepsy that may follow febrile status epilepticus (FSE) is crucial, as inflammatory mediators offer targets for prevention and intervention strategies. Experimental models suggest that neuroinflammation contributes intrinsically to the generation of fever-related seizures in children. FSE, defined as prolonged febrile seizures (FS), often precedes and likely contributes to a significant subset of adult temporal lobe epilepsy (TLE). Because TLE may be associated with cognitive and emotional problems and is commonly refractory to treatment, unraveling the connections between FSE and TLE and specifically the role of neuroinflammation and related epigenomic mechanisms is of significant translational value. This chapter discusses the contribution of inflammatory mediators to FS and FSE and then focuses on evidence from experimental models for involvement of inflammatory pathways in FSE-related epileptogenesis. Capitalizing on information from both human studies and animal models, we discuss the potential contributions of several salient neuroinflammatory cascades, as well as of blood-brain barrier disruption, microRNAs, and downstream transcriptional regulators, in the context of FSE and epileptogenesis in general. Finally, we highlight the potential efficacy of both pathway-specific blockers and global antiinflammatory strategies for mitigating FSE-related epileptogenesis.
Febrile seizures (FSs) are the most common type of seizures in humans. Yet, many of the questions associated with FS remain unresolved. Why do FSs arise? Why do they involve specific developmental ages? Is a genetic predisposition required? A specific type of systemic infection? Does temperature or the rate of change in temperature trigger the seizures? Which fever or other immune mediators are involved? What governs seizure duration? When do the seizures become deleterious? How do we treat them? Similar and additional questions relate to febrile status epilepticus (FSE): Why do some children develop FSE? What are the direct consequences in an otherwise normal brain? In individuals with genetic predisposition (including specific mutations)? What is the mechanism of epileptogenesis? Of cognitive problems? How do we treat them? Can we prevent them? Many of these questions are difficult to resolve in studies involving children. Clear answers require experimental models in which both genetics and environmental factors can be controlled. In addition, the short life span of rodents allows rapid prospective analyses, and the ability to directly study the brain tissue enables mechanistic studies and interventions. Therefore, several models for FS and FSE have been created and employed. These address several issues: the role of specific human mutations in generating FS and FSE, underlying biologic mechanisms that contribute to generation of FS and FSE, the consequences of FSE including epileptogenesis and cognitive deficits, and the pathophysiology of FSE-related epileptogenesis and related comorbidities. Models are also used to test the efficacy of pharmacologic interventions incorporating specific and global immune modulators as well as antagonism of transcriptional and epigenetic pathways. They have also been used to study alternative therapies including microRNA and metabolic approaches. Models have been created in multiple species, from Drosophila and fish to mice and rats. This chapter concludes with the discoveries of the most salient genetic and imposed models which, in the aggregate, enable important contributions to current and future understanding of FS and FSE.
Technologies are increasingly being used to provide mental health interventions — either as conduits to professionally delivered care or as technological interventions. Despite a rich history of the concept of therapeutic alliance in psychotherapy interventions, researchers have struggled to define and measure digital therapeutic alliance (DTA) in a way that accounts for the unique constraints and opportunities of digital mental health interventions (DMHIs). We define DTA as a user-perceived alliance with a DMHI, including both human and technological components of the intervention. While people are sometimes averse to the idea that they could have a relationship or alliance with a digital tool, they may become more open to this idea as technological advances, such as artificial intelligence, make DMHIs more responsive and tailored to particular individuals. A more comprehensive understanding of DTA could help to design DMHIs that are more engaging and effective, increasing their impact.
Research on automated social media rumour verification, the task of identifying the veracity of questionable information circulating on social media, has yielded neural models achieving high performance, with accuracy scores that often exceed 90%. However, none of these studies focus on the real-world generalisability of the proposed approaches, that is whether the models perform well on datasets other than those on which they were initially trained and tested. In this work we aim to fill this gap by assessing the generalisability of top performing neural rumour verification models covering a range of different architectures from the perspectives of both topic and temporal robustness. For a more complete evaluation of generalisability, we collect and release COVID-RV, a novel dataset of Twitter conversations revolving around COVID-19 rumours. Unlike other existing COVID-19 datasets, our COVID-RV contains conversations around rumours that follow the format of prominent rumour verification benchmarks, while being different from them in terms of topic and time scale, thus allowing better assessment of the temporal robustness of the models. We evaluate model performance on COVID-RV and three popular rumour verification datasets to understand limitations and advantages of different model architectures, training datasets and evaluation scenarios. We find a dramatic drop in performance when testing models on a different dataset from that used for training. Further, we evaluate the ability of models to generalise in a few-shot learning setup, as well as when word embeddings are updated with the vocabulary of a new, unseen rumour. Drawing upon our experiments we discuss challenges and make recommendations for future research directions in addressing this important problem.
Accounting for the ecosystem service values (ESVs) and discussing the relationship between the ESVs and economic development can help achieve sustainable ecological development. Therefore, this paper evaluates the county-level ESVs of various land types in China, and depicts the distribution of ESVs in various urban agglomerations. In addition, the nonlinear relationship between ESVs and economic development is revealed. The main findings are as follows: (1) From 2000 to 2018, the ESVs in China decreased, and the decline rate of ESVs in urban agglomerations is much higher than that of China as a whole. (2) The decline rate of ESVs in core cities is much higher than in urban agglomerations, and the decline rate of ESVs is higher in areas close to core cities and lower in areas far from core cities. (3) The ecological Kuznets curve of China has a positive “U” shape, and the ecological Kuznets curve of urban agglomerations has an “N” shape; the ecological Kuznets curve of core cities has a positive “U” shape, while the ESVs of other cities decreases monotonically with the increase of the economic level.
The rigidity of the spacetime positive mass theorem states that an initial data set (M, g, k) satisfying the dominant energy condition with vanishing mass can be isometrically embedded into Minkowski space. This has been established by Beig–Chruściel and Huang–Lee under additional decay assumptions for the energy and momentum densities μ and J. In this note, we give a new and elementary proof in dimension 3 which removes these additional decay assumptions. Our argument uses spacetime harmonic functions and Liouville’s theorem. We also provide an alternative proof based on the Killing development of (M, g, k).
How do multijurisdictional political fields impact the strategies of pro- and anti-immigrant advocates? The geographical literature on immigration demonstrates that immigration policy in the United States has become decentralized, federalized, and fragmented. However, scholars studying immigration politics and activism continue to conceptualize mobilizations as unfolding within jurisdictional containers. This paper examines how advocates on both sides of the issue develop strategies in response to multiple entangled jurisdictions. It does so through a case study of contentious immigration politics in Orange County, California during the 2010s. The paper maintains that a multijurisdictional field distributes political opportunities unevenly to opposing advocates. How advocates respond to these opportunities depends on the distribution of resources to each side across this complex political space. The combination of political opportunities and resources determine the strategies that pro- and anti-immigrant advocates pursue. The paper derives its data from regional newspapers and employs a “claims analysis” to analyze waves of mobilization, actors, attitudes, locations, and strategies (Koopmans & Paul, 1999). The paper shows that advocates on both sides were not contained within single jurisdictional walls. Instead, they developed complex geographical strategies that sought to exploit opportunities in friendly jurisdictions to combat threats from unfriendly jurisdictions.
Background For years, US medical schools have relied on community-based, private clinicians to educate medical students. There has been a steady decline in the number of physicians willing to take on medical students in their clinical practices. Recent issues related to the pandemic raise questions about how many patients students should see to have a meaningful clinical experience. Methods As part of a 16-week longitudinal clinical experience, medical students spend 2 days each week in a family medicine or internal medicine clinic. As repetition enhances learning, maximizing the number of patients students see is important. Using a mixed integer linear program, we sought to determine the optimal schedule that maximizes the number of patients whom students see during a rotation. Patient visits were collected from January to April 2018 for clinics used by the medical school. By maximizing the minimum number of patients per learner over all non-empty day-clinic combinations, we deliver equitable rotation plans based on our assumptions. Results For this pilot study, multiple experiments were performed with different numbers of students assigned to clinics. Each experiment also generated a weekly rotation plan for a given student. Based on this optimization model, the minimum number of patients per student over 16 weeks was 87 (3 patients per day) and actually increased the number of students who could be assigned to one of the clinics from 1 student per rotation to 8 students. Conclusions The mixed integer linear program assigned more students to clinics that have more total visits in order to achieve the optimal and fairest learning quality. In addition, by conducting various experiments on different numbers of students, we observed that we were able to allocate more students without affecting the number of patients students see.
This study examines how changes to patients’ financial responsibility affect physicians’ behavior. This is achieved by examining a health insurance reform that changes patients’ relative financial responsibilities for a medical service that can be received at one of two locations. In particular, this study examines how physicians’ treatment location decisions change after the reform. This study finds that physicians who previously work across the two locations are increasingly observed working at the location that becomes cheaper for patients. Thus, physicians’ responsiveness to new policies may be an important lever by which certain demand-side health insurance reforms successfully operate.
Human APOE4 (apolipoprotein E4 isoform) is a powerful genetic risk factor for late-onset Alzheimer disease (AD). Many groups have investigated the effect of APOE4 on the degradation of amyloid β (Aβ), the main component of plaques found in the brains of AD patients. However, few studies have focused on the degradation of APOE itself. We investigated the lysosomal trafficking of APOE in cells and found that APOE from the post-Golgi compartment is degraded through an autophagic process requiring the lysosomal membrane protein LAMP2A. We found that APOE4 accumulates in enlarged lysosomes, alters autophagic flux, and changes the proteomic contents of lysosomes following internalization. This dysregulated lysosomal trafficking may represent one of the mechanisms that contributes to AD pathogenesis.
Emerging studies highlight the Hippo pathway as an important player in organ size control, tissue homeostasis, regeneration, development, and diseases, but our understanding of its roles and regulations remains incomplete. Our recent work reported a functional interplay between the Hippo pathway and heavy metals, providing new insights into this key signaling pathway.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.
14,235 members
Sunny Jiang
  • Department of Civil and Environmental Engineering
Benjamin Colby
  • Department of Anthropology
Hazem Orabi
  • Department of Urology
Michael David Lee
  • Department of Cognitive Sciences
UC Irvine Medical Center, 92868, Irvine, CA, United States