AI flags heart failure patients at risk of functional decline

Michael Walter | October 20, 2026 | Cardiovascular Business | Heart Failure

Advanced artificial intelligence (AI) models can anticipate when older heart failure patients may start to experience functional decline, according to new findings published in the Journal of Cardiac Failure.[1]

Researchers explored data from more than 6,000 patients 65 years old or older who were hospitalized for heart failure and prescribed rehabilitation. The median age of the cohort was 82 years old. An advanced machine learning model was then developed capable of evaluating a patient’s one-year risk of experiencing functional maintenance, functional decline or death. Functional decline was specifically defined as anyone who transitions to a housebound/bedridden lifestyle due to their health.

Overall, 32.5% of patients experienced functional decline and more than 14% of patients died within one year of being discharged from the hospital. The team’s advanced AI model, which used 77 different factors to make its assessments, was able to identify patients at risk of functional maintenance, functional decline or death with an improved accuracy compared to the Kihon Checklist screening tool.

By using advanced AI in this way, the study’s authors believe care teams can have a better understanding of a patient’s long-term health. This has multiple benefits, including the ability to discuss topics such as rehabilitation with patients as well as their loved ones.

“Instead of asking only whether a patient will survive, we need to ask whether they will be able to maintain the life they value after discharge,” lead author Kanji Yamada, PT, PhD, a researcher with Juntendo University in Japan, said in a statement. “A practical risk estimate could help clinicians and patients discuss rehabilitation needs earlier and make better use of limited resources.”

Click here to read the full study.