Chronic pain affects 30% of children and adolescents, putting them at risk for physical and psychological impairments. Given poor access to pediatric pain care, psychological treatments such as cognitive behavioral therapy are more accessible through digital health interventions. One intervention our lab created is Web-based Management of Adolescent Pain (WebMAP). In this mobile app, youth with chronic pain develop pain management strategies by progressing through interactive modules. Although digital health interventions like WebMAP are transforming accessibility to chronic pain care, limited information is available to understand how best to enhance youth engagement in a mobile app or whether there are differences in engagement among youth with different background characteristics. Our project aimed to 1) identify barriers to engagement with WebMAP among subgroups defined by user demographics and 2) develop recommendations for enhancing engagement with WebMAP and extending its reach to target underserved populations. In a previous cluster randomized controlled trial, participants assigned to the WebMAP intervention were 73 youth aged 10 to 17 (84.3% female) with various chronic pain conditions. We analyzed the app metadata to assess module completion rates across various subgroups, including groups defined by age, race, sex, and annual household income. We also coded interview and questionnaire responses from users to identify app preferences and areas of improvement. Results indicated that although users liked WebMAP’s design and content, they suggested changes to its display and features. By viewing our findings on usage and perception through a health equity lens, we summarized the accessibility of WebMAP and outlined recommendations to enhance its cultural relevance to different groups. Ultimately, our research identified factors that impacted engagement with WebMAP and offered valuable insight into optimizing digital health psychological interventions for chronic pain management. Moving forward, we hope to apply these findings as we develop the next release of WebMAP.

Teaching robots to efficiently search for a target object (such as medication) in cluttered environments (such as a house) with limited prior information is a challenging yet important task, with applications ranging from home assistance to search and rescue. An ideal search policy, i.e. strategy, maximizes the accumulation of the target object while travelling along an efficient search route. We formulate object-search as a combinatorial-optimization problem known as the Stochastic Orienteering Problem (SOP), which is a graph traversal problem where an agent must identify and traverse a subset of nodes (such as furniture items, in our case) in a graph data-structure, with each edge associated with some cost and each node associated with some expected reward. The agent must choose a path that maximizes expected reward and keeps total travel cost under some prespecified bound, where the bound is informed by an unavoidable real-world constraint such as battery-usage. In our formulation, we call each node a "container", a catch-all term for any distinct area that can hold an object (such as a cabinet), and the edge costs represent the distances between each container. In this work, we mathematically show how this SOP can describe object-search tasks at a high-level, and present a simulated agent trained in a basic grid-world environment. We leverage the powerful reward-maximization capabilities of deep reinforcement-learning (a subfield of machine learning) to achieve near-optimal performance for solving this object-search SOP. The broader implication of this work is that real-world robotic object-search tasks are well-described by SOPs, since the multi-objective nature of SOPs forces the agent to choose search policies that both have a high-likelihood of finding the object(s) and do not exceed a hard constraint, such as energy expenditure.

Despite the brain being a 3D structure with a complex topography and spatial relationships, neuroscientists currently rely on 2D visualizations. These less informative visualizations obscure the distances between 3D regions, and hinder scientists’ ability to perceive functional correlations and anatomical connections. To provide a more decipherable method of exploring the structure and function of the brain, we built neuroscience tools specifically aimed toward exploratory 3D data visualization. I worked on the development of a Universal Renderer for Neuroscience (Urchin) that lets users plot their data in its original 3D anatomical context. Urchin can perform a variety of different functions such as displaying certain features (e.g. neurons, brain regions, or contextual objects such as probes), or interactively exploring the data within context of brain location via mouse and keyboard navigation. This not only paves the way for new methods of data analysis but also creates a deeper understanding of the structure and patterns found within the data. I worked on building Urchin within the Unity platform, implementing features to enhance data exploration and analysis via scripting in C. Some examples of functionality that I built include implementing 3D mesh rendering for brain regions, primitive models, and changing materials. I also established a proxy server that allows for secure communication between client side browser applications and python notebooks. Along with this, I developed a more intuitive and efficient python API that allows people with minimal coding experience to run the renderer visualizations with ease. Urchin enhances neuroscience research and education by providing a more interactive and immersive experience, allowing students and the public to directly engage with diverse data sets and investigate different aspects and features of the brain.

Electrophysiology experiments targeting deep brain structures require extensive training and expertise. However, even experienced researchers face challenges in placing electrodes precisely within a target location, particularly when using multiple electrodes simultaneously. On average, there is a 400-um (standard deviation) of human error when targeting Bregma and navigating to insertion coordinates. Slow setup time and human error can lead to unnecessary stress in experimental animals and prevent scientists from focusing on data collection. Our laboratory developed an experiment planning tool called Pinpoint to address these challenges. However, even with interactive tools, a typical two-probe experiment setup can take over an hour, increasing as more probes are added in complex experiments. To reduce time inefficiencies and lower the risk of human error, we developed an electrode manipulator automation platform for Pinpoint. Our platform consists of a server application called Ephys Link, which unifies communication between Pinpoint and various electrode manipulator platforms. With Ephys Link, scientists can view the electrodes they are using in their experiment live inside the virtual brain and pre-plan insertions for multiple probes. They can then simply press a button to have their probes automatically move to their chosen targets. We expect our automation pipeline to make multi-probe electrophysiology an easier and more accessible task for researchers, enabling them to focus on gathering high-quality data rather than managing the geometry of their experiments. To measure the impact of our automation platform, we plan to use positional logging, timed recordings, and researcher feedback to evaluate the efficacy of the pipeline in speeding up electrophysiology experiments. We expect to see increased targeting precision, reduced time setup time, and overall productivity boosts for researchers. By reducing electrophysiology's difficulty and time-consuming nature, our automation pipeline helps researchers alleviate cumbersome experiment setups and prevent unnecessary stress on experimental animals.

Photovoice is a research methodology that uses photography as a tool for participants to express their thoughts on a topic of interest. Our project is specified towards marginalized communities and to give them a voice regarding issues each group faces. The main purpose of photovoice is to create change in our community by analyzing photographs provided by the participants.

Recent publications of photovoice projects at universities have yet to focus on a number of groups and their experiences on campus. We have decided to create a photovoice project specifically tailored to this area of interest.

This project focuses on four groups in the University of Washington Bothell campus: Students of Color, Muslim students, LGBTQ students, and first-generation students. 15 students participated in this study, with around 2 to 4 participants in each group. Our goal is to learn how these groups experience college/university differently compared to students outside of these groups. We asked participants to answer prompts that discuss how they experienced college academically, socially, and emotionally. Thus, we analyzed their experiences and formatted them into differing “codes” and “themes”.

Most themes in the project overlapped between the groups. The three notable themes were (1) The disconnect between faculty and students, (2) Expectations of what students should have and have access to, and (3) Struggling in finding ways to express oneself. Two more themes are to be created from the data we collected.

The project is still under development, and we are expecting to analyze these themes more in-depth. The goal of this project is to highlight disparities in the universities’ process of matriculation. Moreover, we hope to use this research to revise the assumptions and rules regarding resources and general student welfare to create more accessible resources and a streamlined transition from primary to secondary education.

Racial and ethnic minority groups are less likely to seek or have access to mental health support services. Despite an increase in the quantity of mental health services nationwide, disparities in usage across racial and ethnic minority groups still exist, leaving resources underutilized and many individuals with unmet needs. Previous studies highlight both structural and non-structural barriers, such as income and stigma, which prevent certain racial groups from fully utilizing health care services. Studies often overlook health outcomes associated with immigrant status and overgeneralize findings to all populations. While scholars have studied the structural barriers to accessing mental health care services, there is a need to understand the cultural contexts preventing racial and ethnic groups from reaching out to these services. Using data from the National Health Interview Survey (NHIS), this study examines differences in access to and utilization of mental health care services of U.S born and immigrant populations from the years 2010-2020. The NHIS allows for the control and determination of effects of both structural and cultural factors such as health insurance coverage, financial instability, racial and ethnic characteristics, and migrant status to further understand the dynamics of access to and use of mental health care services. Foreign-born individuals are expected to show lower access to health care resources with varying levels of moderate mental distress. These differences are expected to vary for those who have spent more time in the U.S with more access to resources and higher levels of mental distress. Understanding trends related to mental health care can help develop better public policy responses and improve the promotion of health care services. This study will help address why services are underutilized and how to minimize health service inequalities among racial and ethnic populations. 

The drug rapamycin can increase lifespan in a variety of model organisms by repressing the activity of the mTOR complex, a cellular component required for growth and development. Other than one small study that looked at genetic variation in the effect of rapamycin on lifespan in fruit flies (Rohde et al., 2021), little is known about how natural genetic variation affects the response to rapamycin. Previous work by the Promislow lab, utilizing developmental time to indicate rapamycin’s affect, has shown that some strains of fruit flies are completely resistant to rapamycin while others are sensitive. These genotypic differences are also reflected in the metabolome, the complete set of small molecules and metabolites present within cells. Metabolome analysis of these strains revealed significant differences in metabolite concentrations between resistant and sensitive lines. Interestingly, when we treated sensitive strains with rapamycin, their metabolome profiles were like those of starved larvae. I hypothesize that rapamycin is affecting larval ability to take up nutrients and that the starved metabolome is a result of actual starvation. To test my hypothesis, I am designing a starvation assay to compare the death rate of sensitive and resistant larvae. Measurement of resistance to starvation is taken two days after rapamycin treatment by transferring larvae to nutrient deficient food. The duration of time for individual larvae to die is recorded, and the death rates between the two populations are compared. If my hypothesis is true, the sensitive larvae will have a higher death rate than the resistant larva. If death rates are similar however, that could mean that rapamycin does not cause a nutrient deficit and there is another explanation for its effects on the metabolome. This study will provide insights to the underlying mechanisms of sensitivity to rapamycin, and why it might differ between individuals.

The Pacific Northwest is home to a lush lowland forest that continually changes. Quantitative monitoring is essential in understanding forest health and climate change. The primary task was to help create a comprehensive, multi-year, dataset of floral phenology for Green River College’s learning forest; botany and ecology classes participated together in this endeavor. This quantitative data set allowed us to establish baselines for our forest so we could ask questions about how our forest changed over time. How do the native species compare to the invasive species? How many species are there, and which is most abundant? Using the ArcGIS Survey123 app, we measured floral species, abundance, GPS location, and phenology stages (buds, flower, seed/fruit, and leaves). The data was then analyzed using Qlik, an online visualization software. In 2022 alone, we compiled ~13,000 observations, estimating ~3.2 million inflorescences across ~140 species and are currently curating data from 3+ years. When looking at the floral expression in our forest from the last two dry summers, one native species (Wild Ginger) has disappeared, while many non-native and invasive plants were able to thrive. With our data set we can quantitatively look at entire plant families and monitor ecological change. A secondary benefit of this endeavor was to improve the educational capacity of our classes. Students were fully immersed in the forest and floral details. Photographs from students also helped create a repository for plant images of floral phenology stages which strongly aids in identification and higher level data quality. Our project helped undergraduates contribute scientifically while learning about their world. These contributions can help inform decisions on issues such as mitigating climate change or ecosystem preservation. Future directions include expansion of this to city parks and popular hiking trails to broaden our quantitative understanding of our ecosystems.

Aging is an important problem in biomedical research. Given the increased risk of death with age, techniques to delay aging hold substantial promise for human well-being. The premise of my research is that rapamycin, a drug commonly used in transplant patients which is hypothesized to have development-slowing effects, can slow development of fruit flies. One finding from a previous study, however, is that there is enormous genetic variation within the Drosophila population (akin to the variation between dog breeds) which results in varying sensitivity to rapamycin. Earlier measurements indicated that strains of fruit flies that were sensitive to rapamycin had higher levels of histamine, a molecule that participates in metabolism, when on the drug. We set out to determine whether histamine could effectively extend sensitivity to the drug to a wider range of genotypes within Drosophila. Thus, I hypothesize that introducing histamine to the food that fruit fly strains that are genetically resistant to rapamycin are consuming will increase sensitivity. To test this hypothesis, I added solutions with different concentrations of histamine to food with or without rapamycin. I then placed eggs from a resistant strain onto the food to observe the time it takes from egg to pupa. Because I hypothesize that increasing histamine levels will make a resistant strain of Drosophila sensitive to rapamycin, the strain should become sensitive and consequently show an increase in development time compared to the conditions without histamine. Each test condition of a histamine solution and rapamycin has a corresponding control with no rapamycin. Hence, if the histamine-treated conditions show slower pupation times than the histamine-free controls, the hypothesis is validated. If validated, this work could help researchers understand ways to provide the benefits of rapamycin to individuals who might otherwise be genetically resistant to its impact in both aging and medical contexts.

Recent studies and news sources have brought attention to the possible health risks associated with cooking on a gas stove due to the release of high concentrations of pollutants, such as NO₂, particulate matter (PM), and other chemical pollutants. But currently, there is little information on how individuals can monitor their own indoor air. Several low-cost air quality sensors are available, but little data to show if these work for home environments. Low-cost sensors are important because they can be more accessible to more people due to their affordability. Two brands of low-cost air quality sensors were used in this study to measure PM and various chemical pollutants: the TSI AirAssure and the Vaisala AQT530. We conducted several kitchen experiments that involved cooking on the stove or the oven while monitoring pollutant levels on these sensors. I found that pollutant levels rose significantly when cooking on both appliances. One experiment revealed that within minutes of cooking in the oven, NO₂ levels went above the recommended EPA standard of 100 ppb per hour. I also did a comparison experiment between the Vaisala AQT530 and a high-quality sensor from the Puget Sound Clean Air Agency at the downtown air quality station (10^th and Weller). Results showed moderate correlations between the sensors for NO, NO₂, and CO concentrations and a strong correlation for PM concentrations. We will continue to evaluate the sensors by doing a direct comparison of these sensors with standard regulatory methods in a test kitchen. The goal is to derive correction equations that can be applied to the data received from these low-cost sensors. My findings in the kitchen and sensor comparison experiments will provide valuable information on when pollutant levels impose health risks and information on the level of accuracy of the data retrieved from these devices.

The drug rapamycin has been shown to extend lifespan in model organisms ranging from mice to fruit flies, but little is known about how genetic variation affects the response to rapamycin. It is possible that some individuals might not respond to the drug, or may even respond negatively. To study the impact of genetic variation on rapamycin sensitivity, the Promislow lab has been using the fruit fly, Drosophila melanogaster, to test the ability of rapamycin to slow early development in different strains. They found a range of responses, from strains that were completely resistant to those that were highly sensitive. To better understand why resistant and sensitive strains differ, the Promislow lab analyzed their metabolome profiles. The metabolome consists of diverse small molecules, metabolites, that are fundamental to sustaining life in cells, and the relative abundance of metabolites is referred to as a metabolome profile. By comparing metabolome profiles, the Promislow lab found that the metabolome of sensitive larvae treated with rapamycin was similar to those of larvae starved of nutrients. One possible explanation for this observation is that sensitive larvae on rapamycin-treated food eat less food than resistant larvae. To test my hypothesis, I am measuring how much food they ingest, using dyed food, and scoring how much dye each larva ingests over time. If sensitive larvae eat less rapamycin-treated food than resistant larvae, then we can investigate why rapamycin affects feeding. If no significant difference is found between scores for sensitive and resistant larvae, then we can investigate why rapamycin affects early development in sensitive strains when their food consumption is not limited. By determining if sensitivity to rapamycin is accompanied by differences in feeding, then we could potentially manipulate feeding to sensitize flies, and possibly humans, to this beneficial drug.

The period around birth is when neonates are at the highest risk of neurological injury or death. A common neonatal neurological injury is hypoxic-ischemic encephalopathy (HIE), which occurs after the brain does not receive enough oxygen or blood flow. There is a large disparity in the severity and long-term neurodevelopmental outcomes of HIE between high-income countries (HICs) and low-and-middle income countries (LMICs). In HICs, HIE occurs in 1-4 neonates per 1,000 births. In LMICs, the instance of HIE is at least 2-3 times higher. Furthermore, cases of HIE seen in LMICs suggest a different type of injury - a more prolonged intermittent injury resulting in white matter injury - compared to HIE in high-income countries that is more acute and affects the deep grey matter. Therapeutic hypothermia (TH) has been the standard of care for HIE in HICs; however, TH is not an effective treatment for HIE in LMICs. Thus, the creation of alternative and accessible therapies for HIE in LMICs is crucial. This study will seek to model HIE as seen in LMICs through an in vitro ferret model that may be used to pilot therapies before applying them to in vivo models. Organotypic brain slices from postnatal day (P) 21 ferrets, equivalent to a term neonate, will be cultured and randomized to receive increasing intervals of oxygen glucose deprivation (OGD), with and without serum deprivation. Serum deprivation is defined as culturing in 2.5% serum as opposed to the standard 5% to mimic certain aspects of malnutrition that may be more common in LMICs. Cell death and white matter injury will be assessed 24 hours after OGD. We hypothesize that slices with more rounds of intermittent OGD and serum deprivation will display relatively more cell death and white matter injury, thus serving as a model of HIE in LMICs.