A strategic research alliance between UMBC and the University of Maryland, Baltimore (UMB) has selected four new interdisciplinary projects, each a fresh take on a complex challenge.
The Accelerated Translational Incubator Pilot (ATIP) Program funding this work brings together UMBC’s strengths in areas like cybersecurity, data science, artificial intelligence, statistics, and the social sciences with UMB expertise in medicine, pharmacy, nursing, and dentistry. The two universities have worked together on various shared graduate and research programs over the last decade. This partnership with UMB’s Institute for Clinical and Translational Research (ICTR) was launched in August 2019, and faculty have been quick to jump on the opportunity to pursue novel collaborative research.
“The UMB-ICTR is thrilled to partner with UMBC,” says ICTR Director Stephen N. Davis, chair of medicine at UMB and physician-in-chief at the University of Maryland Medical Center. “The ATIP grant program supports top quality clinical and community-based collaborative projects. The grant applications from UMBC are particularly innovative and exciting, scoring very highly in the review process.”
“All of the ATIP proposals are subject to a very rigorous internal peer review process,” says Karl V. Steiner, vice president for research at UMBC. “Their success is a strong indicator of the quality of the intellectual contributions of our faculty at UMBC.”
Meaningful research relationships
“A key goal for this strategic alliance is to develop meaningful partnerships among researchers at both institutions and to establish teams with complementary expertise,” Steiner explains.
Since early 2020, UMBC faculty have secured a total of 12 ATIP awards. This most recent round of UMBC grant recipients includes Lujie Karen Chen, information systems; Lira Yoon, psychology; Chein-I Chang, computer science and electrical engineering (CSEE); Yi Huang, mathematics and statistics; and Tinoosh Mohsenin, CSEE. UMB collaborators include Kelly Doran, School of Nursing; Mathangi Gopalakrishnan, School of Pharmacy; and Michael Domanski and Mohammad Sajadi, both of the School of Medicine.
The four new awards granted in this round focus on a broad range of topics: using machine learning algorithms for transfusion risk assessment, evaluating the effects of serum lipid levels on the progression of renal dysfunction, using a multimodal sensory machine learning framework to diagnose COVID-19, and examining how to predict and manage stress in healthcare workers who work in long-term care facilities.
These types of sustained partnerships help researchers gain insight into new fields and work with experts with whom they might not have otherwise collaborated, to generate novel findings.
Measuring stress to manage stress
So, what does interdisciplinary research actually look like?
For the long-term care workers pilot project, Chen, Yoon, and Doran combine data science, clinical psychology, and nursing. From May 2021 through April 2022, they will examine workers’ experiences of job stress in long-term care facilities using both physiological and qualitative measures.
The challenges posed by COVID-19 have sharpened the team’s interest in supporting longer-term care workers through research.
“Their work has always been important, but especially given the pandemic, the importance of their work has really been highlighted,” says Yoon. “It’s a demanding job with high stress, and that’s reflected by high job turnover, which is not good for the clients at long-term care facilities.”
The goal of the project is to better identify stress triggers within moments of them happening. This could enable workers to address their stress before it gets to a higher level where it can negatively impact their health and work.
Stress monitoring in action
To identify stress on a more moment-to-moment basis, the team plans to use a combination of surveys and sensors to track exact times when workers become stressed in their workday.
Workers will wear the sensors throughout their shifts to measure heart rate and electrodermal activity, and will fill out surveys five to seven times per day. They will also complete end-of-day interviews to identify additional stressors and other critical information not captured by the surveys.
Information from the surveys and interviews will help the researchers to decipher data obtained from the sensors to understand stress triggers.
“Eventually, we want to be able to just use sensor data to be able to tell whether long-term care workers are about to experience stress or not,” Yoon says. With this knowledge, researchers can then design interventions to improve the work environment, Chen explains, to reduce stress triggers and stress experiences.
Once the pilot project is complete, the team hopes to undertake a larger scale study to collect data from multiple long-term healthcare sites.
This work wouldn’t be possible without all three researchers bringing together their diverse expertise and perspectives. As Doran says, “The collaboration provides a support system for us to build relationships and kind of cross-train each other and build off each other’s ideas.”
“One of the reasons I love doing collaborative research,” she shares, “is because you’re expanding your networks and research to make meaningful differences.”
Article written by Allison Matyus for UMBC News.
Tags: CAHSS, CNMS, COEIT, CSEE, IS, MathStat, Psychology, Research
