Two researchers from the University of Cincinnati (UC) are set to present their work on using artificial intelligence to improve care and research for Parkinson's disease at a first-of-its-kind conference in Washington, D.C., later this month.

Joaquin Vizcarra, MD, a research assistant professor in UC's College of Medicine, and Nathaly Chinchihualpa, MD, a neurology research fellow at the same institution, will speak at the AI for Parkinson's conference on November 19. The event, organized by the American Parkinson Disease Association, brings together experts from across the country to discuss emerging applications of AI in the field.

According to UC, the researchers are part of the university's Clinical Intelligence Lab (CIL), which is exploring several ways AI could address longstanding challenges in Parkinson's care. One major issue is access: in 2019, fewer than 10% of Parkinson's patients saw a specialist, and those who did waited an average of 90 days for an appointment. With the global number of Parkinson's diagnoses projected to rise by 112% between 2021 and 2050, according to researchers cited by UC, the gap is expected to widen.

Vizcarra told UC that the lab is investigating how large language models (LLMs) and agentic workflows could help non-specialist clinicians provide better care. "We're exploring how to empower primary care physicians, general neurologists or potentially advanced practice providers to have a 'specialist in their pocket' with appropriate benchmarks in different levels of care," he said.

On the research side, the team is using LLMs to extract and organize information from clinical notes in electronic health records. Vizcarra noted that routine clinical records often contain more frequent data on symptoms and treatment than annual research visits, but the information is unstructured and difficult to analyze at scale. "In one study, we analyzed more than 25,000 clinical notes from approximately 2,800 patients in over 10-plus years of follow-up," he said.

Chinchihualpa added that the lab is working with the Cincinnati Cohort Biomarker Program, a large prospective study of participants with movement and cognitive disorders led by Alberto Espay, MD. The AI tools can complement data collected at research visits by pulling additional detail from clinical narratives.

LLMs may also help identify early signals of disease progression that clinicians might otherwise miss, according to Vizcarra. "There's a lot of valuable information buried in clinical narratives, and these technologies allow us to extract and organize those bits and pieces into structured data over time," he said. "When you process it with these technologies, maybe you're starting to get signals that happened earlier that you otherwise would have not recognized."

The team is also working to address several challenges, including data privacy. "We are working very hard making sure that we have good governance in terms of data, brokerage, and in keeping each patient's privacy protected," Vizcarra said. Other obstacles include the need for strong computing power to scale the work beyond a single clinic, and expanding the dataset to include the 90% of Parkinson's patients who do not see specialists and are therefore underrepresented in current clinical data.

At the conference, Vizcarra will present an overview of current AI tools used in Parkinson's clinics and discuss near-future technologies for patient care. Chinchihualpa will give a lightning presentation describing a framework that uses LLMs to integrate clinical notes with data from the Cincinnati Cohort Biomarker Program, turning them into granular research data on how the disease evolves in each patient. She will also receive a Rising Star Award from the conference organizers.

For university students interested in careers in medicine, data science, or neurology, the work highlights a growing intersection between clinical care and artificial intelligence. The researchers' focus on using AI to extend specialist expertise to underserved populations also points to potential opportunities in health equity and digital health innovation.