Picture Perfect – Researchers and Librarians Partner to Improve Medical Imaging
by Abby Winterburn, Student Communications Assistant
The impact an image can have is powerful. Whether historical, sentimental or in some cases, medical. When it comes to improving medical imagery, mainly CT scans, the recent work between researchers at the University of Wisconsin-Madison and its librarians showcases a whole new picture of the power of partnership.
In 2021, post-doctoral researcher Jevin Lortie and his team, including Grace Gage, Adam J. Kuchina, Ben Rush, Tim Szcykutowiczbegan, and Steven Heymsfield, began researching the ability of CT scans to view muscle health. His lab focused on imaging techniques and types, like MRI, X-rays, and ultrasounds. The objective was to determine what CT scanner variables might bias our muscle measures.
“We looked at variables that affected scans, like whether the brand or adjustment around the energy to the scanner’s x-ray tube impacted muscle measures and sent them out of whack,” commented Lortie.
Lortie noted the project was a review to develop a complex article search. This meant the researchers needed to complete an article search that would surface any articles on existing research in their topic area, then screen out the articles they did not want. They would then synthesize the remaining data and compile it into organized tables.
Lortie discovered he needed support managing and reviewing the research during this project. Enter the University’s librarians. The librarians assisted Lortie in determining what kind of review to conduct, given their topic area. They developed syntax for searches for each website, analyzed search terms to determine how narrow/wide they were and helped with any search framing adjustments.
“We needed our searches to hit the topics we wanted without retrieving too many articles,” said Lortie. “Some of the first searches we conducted were getting 10,000 or more hits, which isn’t manageable,” he added. “Due to their expertise, the librarians could get this number to a reasonable amount, like 1,000, by refining our search terms and still hitting all the articles we needed in our search.”
The librarians also helped refine search terms. They taught Lortie and his team to use technology like Covidence screening tools, article management software, and Medical Subject Headings Terms for searches. This advanced search method uses “a group of words categorized under one keyword,” Lortie explained. The librarians also provided context about the research process in Lortie’s methods section of his final paper.
“We ended up with two searches that pulled in 1,495 articles, and only 18 were included in our study,” commented Lortie when asked to expand on the librarian’s help. “The librarians helped make this a manageable task and reduce the number of irrelevant articles.”
This partnership benefited Lortie, his team, his research, and the Librarians.
Barbara Sisolak, a Science and Engineering Librarian at the University, helped with this project. She noted that even with the extensive expertise librarians have in assisting researchers through the research lifecycle; they learn ways to streamline and improve their skills.
“We learned quite a lot about working on a project of this nature,” she mentioned in an interview. “We tried out new tools Jevin’s team, and we could use to facilitate the research process.”
Sisolak noted she and her fellow librarians, Jessica Newman and Angel Tang, all had a place at the table during the duration of Lortie’s research. Angel Tang directed the librarian team by organizing notes, scheduling meetings, and maintaining regular communication with the researchers. It became a genuine team effort.
“We valued what this research would have on patient care, and we knew we had to help. Research projects of this magnitude require careful collaboration. In this case, our expertise as librarians provided tremendous support not otherwise available,” said Sisolak. “Librarians and researchers create a powerful partnership, and librarians provide invaluable expertise to the process,” she said.
Systematic reviews are time-consuming, so the librarians and project team spent significant time connecting to understand the research question. The librarians would then advise what they could assist in building professional and detailed search strategies to accommodate multiple databases retrieving significant amounts of scientific literature.
“The initial search needed to be sensitive, detailed, replicable, and as comprehensive as possible to find all potentially relevant studies,” explained Sisolak. “We [librarians] had to find the literature that Jevin’s team did not know existed.”
Lortie and his team plan to use the data collected throughout the project to determine how each CT variable affects muscle measures.
“We will generate recommendations for future research that involves CT and muscle, such as which variables are essential to consider and which are less so, as well as which variables are key to document in the methods sections articles,” explained Lortie.
As Jevin Lortie reflected on his research and plans for the future, he reminisced on the critical nature of support the librarians offered.
“I did not realize how helpful they would be,” said Jevin. “I recommended in a seminar to my department that people get a hold of them [the Librarians] when doing reviews or even just writing intros for research-based papers.”
As for the librarian’s thoughts on assisting with research efforts? They say it sums up what makes their work so valuable – and enjoyable.
“The foundation of any literature review is retrieving the appropriate literature. That’s a complex job, but we have the skills to accomplish this efficiently and effectively,” explained Sisolak. “We love to teach and share this knowledge with our campus research community and are always delighted to get involved.”