How Hands-On AI Training Accelerates Research
Summary
- The Miller School of Medicine’s Dr. Lina Shehadeh launched a series of AI-assisted Python coding and research skills workshops.
- The workshops have now logged 12 total sessions, with 169 unique participants, 293 total participants and a range of roles for attendees, including faculty members, postdocs, multiple graduate and medical students and staff members.
- Sessions are deliberately hands-on and paced to accommodate everyone, from first-time coders to seasoned researchers brushing up on AI-enabled tools.
When Lina Shehadeh, Ph.D., walks into a room, she brings a kind of quiet energy: steady, optimistic and unmistakably forward-looking. A research professor in the University of Miami Miller School of Medicine’s Division of Cardiovascular Medicine, she has become one of the school’s most sought-after voices on artificial intelligence. In the past year, she launched a series of AI-assisted Python coding and research skills workshops that have attracted a cross-section of the academic community: faculty, postdocs, staff, graduate students and even senior leadership.
“We’re seeing a real hunger for practical AI skills,” Dr. Shehadeh said. “People aren’t asking whether they should use AI anymore. They’re asking how to use it responsibly, how to validate it and how to make it accelerate their work instead of complicating it.”
Her assessment isn’t hypothetical. The workshops have now logged 12 total sessions, with 169 unique participants and 293 total participants. The range of roles surprises even her. In almost every class, attendees included faculty members, postdocs, multiple graduate and medical students and staff members.
“It’s one of my favorite parts,” she said. “When a senior faculty member and a first-year Ph.D. student are troubleshooting code together, you realize you’re watching a cultural shift in real time.”
A Research Demand the Field Can’t Ignore
Dr. Shehadeh’s workshops are landing at a moment of rapid change in biomedical research. At the 2025 Society of Vascular and Interventional Neurology (SVIN) conference in Orlando, she presented on AI-assisted research skills as part of a panel on publishing in the age of AI. The response, she says, was eye-opening.
“These residents and fellows weren’t testing the waters,” she recalled. “They wanted to know how to automate ASPECTS scoring, how to predict thrombectomy outcomes, how to shrink their literature review timelines from weeks to days. There’s an urgency now.”
During the session, she shared examples of using tools like Claude and ChatGPT to cut data analysis time, generate publication-ready visualizations in minutes and draft methods or statistical sections with proper validation.
“The message was simple,” she said. “AI isn’t replacing scientific expertise. It’s amplifying it, if you keep rigorous validation at every step.”
Real Impact in the Lab
The workshops also serve as a catalyst for breakthroughs across the Miller School’s research landscape.
Wayne Balkan, Ph.D., a research associate professor in the Miller School’s Department of Medicine, recently shared a striking example. After one of his colleagues attended the session, the lab revisited clinical trial data using new analysis techniques learned in the workshop. The fresh approach uncovered trends that had previously gone unnoticed.
“It was one of those moments where you think, ‘What else is hiding in our data?’” Dr. Shehadeh said. “That’s exactly why we do this.”
The impact has not gone unnoticed by senior leadership. Joshua Hare, M.D., professor of medicine and director of the Interdisciplinary Stem Cell Institute at the Miller School, emphasized how crucial the workshops are for lowering the barrier to entry.
“This will take off like wildfire,” Dr. Hare said. “Everybody knows how important this is, but I think people are intimidated and don’t know how to do it or how to get started. These seminars are really demystifying what you have to do and helping people get that start, which is absolutely critical.”
“I’m so excited to be part of these workshops,” said Colleen Atkins, Ph.D., an associate professor of neurological surgery and associate dean for graduate studies at the Miller School. “AI has seemed a little scary to me. I’ve started to use it, but i haven’t been able to really incorporate it into my research and into the analysis in my laboratory, as well as our graduate education data. And this workshop has already given me the skills to do that.”
Stories like these, Dr. Shehadeh noted, show the program’s larger goal: making advanced analytic capabilities accessible to every researcher, regardless of coding background.
“Like Jumping Out of a Plane with a Parachute”
One of the workshop’s most popular analogies came not from Dr. Shehadeh, but from participants themselves.
“They tell me the workshops feel like jumping out of a plane, with a parachute,” she said with a laugh. “There’s that moment of fear when you run your first AI-assisted script and then, suddenly, everything feels possible.”
Those possibilities range widely: generating heat maps and regression plots with a few lines of code, building outcome prediction models and transforming multi-tab spreadsheets into interpretable visuals in minutes.
The structure of the workshops helps. Sessions are deliberately hands-on and paced to accommodate everyone, from first-time coders to seasoned researchers brushing up on AI-enabled tools. Former trainees now assist with new cohorts, creating what she calls “a self-sustaining ecosystem of learning.”
“You learn the material, and then you teach the next group,” she said. “That’s how you build long-term capability.”
Empowering the Next Generation of Research
Dr. Shehadeh believes the future of academic medicine depends on democratizing technical fluency.
“Every day you spend manually cleaning data or writing boilerplate code is a day you could spend on clinical insights or experimental design,” she said. “AI is the tool that lets us reclaim that time.”
Her vision is not of automation for its own sake, but of elevating human expertise by removing friction, accelerating discovery and sharpening scientific rigor through better validation.
“Start small,” she said. “Validate everything. And don’t be afraid to explore. Some of the best discoveries hide in the places we haven’t been able to look yet because the work was too time-consuming. Not anymore.”
Tags: AI, artificial intelligence, Department of Medical Education, Division of Cardiovascular Medicine, Dr. Lina Shehadeh, medical education, technology