How can AI sentiment analysis apply to complex medical diagnoses?

(Press-News.org) Taking a page from market research tactics, UC San Francisco experts are studying whether artificial intelligence (AI) can improve diagnosis of a complex liver condition by using the clinical notes of multiple providers.   

Their recent study, published in Gastro Hep Advances, focused on hepatorenal syndrome (HRS), a complex condition associated with liver disease that is often difficult to diagnose during hospitalization. The researchers sought to learn if large language models could analyze the clinical notes of multiple physicians and other providers to improve diagnostic accuracy and streamline patient care. 

“The concept is inspired by sentiment analysis technology commonly used with reviews in online shopping platforms, where AI summarizes collective opinions,” said Jin Ge, MD, MBA, UCSF assistant professor of medicine and gastroenterologist, who led the study. “We utilized this approach to determine if collective sentiment could predict an HRS diagnosis.” 

The study compared traditional diagnostic methods based on clinical variables, such as lab results, with an AI-enhanced model that incorporated sentiment analysis derived from clinical notes. Incorporating AI-generated sentiment scores significantly improved predictive accuracy for HRS diagnosis upon patient discharge. 

The technology offers clarity in situations where conflicting recommendations among health care professionals may arise, providing a unified summary of the care team’s consensus for clinicians and patients alike. While still in the research phase, this application has the potential to transform decision-making in hospitals and enhance patient outcomes. 

“Using the ‘wisdom of the crowd’ doesn’t just predict outcomes, it offers a directional insight into what the clinical care team collectively thinks about a patient’s condition,” said Ge. “For cases with mixed opinions or uncertainty, AI-generated summaries could help align care decisions and expedite treatment plans.” 

The study has not yet been implemented in clinical practice but could pave the way for future trials. The researchers aim to evaluate how this information might influence real-world decision-making and patient care. 

Authors: Please see the study for a full list of authors. 

Funding: National Institute of Diabetes and Digestive and Kidney Diseases, UCSF Liver Center Grant, National Center for Advancing Translational Sciences, and National Institute on Aging. 

Disclosures: 

Dr. Jennifer C. Lai receives research support from Lipocene and Vir Biotechnologies; receives an education grant from Nestle Nutrition Sciences; serves on an advisory board for Novo Nordisk; and consults for Genfit, Third Rock Ventures, and Boehringer Ingelheim. 

Dr. Giuseppe Cullaro consults for Ocelot Bio and Retro Biosciences. 

Dr. Jin Ge receives research support from Merck and Co; and consults for Astellas Pharmaceuticals/Iota Biosciences. 

All other authors have no disclosures to report. 

END