AI agent calling cath lab patients to help them prepare for treatment
A pilot program at Mount Sinai Hospital in New York is showing that artificial intelligence (AI) can take on time-consuming administrative work in the cath lab to help nurses stay focused on their patients.
The program uses an AI virtual agent to handle preprocedural phone calls to patients scheduled for cardiac catheterization procedures. The agent, named Sofiya, conducts the same preparatory calls that nurses traditionally do, covering appointment logistics such as timing, location, fasting instructions, drug allergies, diabetes status and what to bring to the hospital.
“We have a volume of 16,000 plus per year, so we have two to three registered nurses just calling patients over and over ... It's mind-numbing work. What we're proud of is, we were able to deploy and clinically use an AI agent," explained Andriy Vengrenyuk, MS, senior data analyst for interventional cardiology at the Icahn School of Medicine at Mount Sinai. He spoke about the pilot program during a session at TCT 2025 in San Francisco and then spoke to Cardiology Business for an exclusive video interview.
The technology is not a replacement for clinicians, but a way to ensure nurses can work at the top of their license. Nurses make these patient calls because, inevitably, clinical questions come up as the cath lab visits approach. According to Vengrenyuk, that is simply the human side of medicine.
Sofiya is programmed to respond to patient questions and hold a conversation when needed. The agent share specific about the procedure, for example, and describe what the patient should expect.
Some patients may view the AI in a negative way and want to speak to a human operator instead. For this reason, Vengrenyuk said they built in a very low threshold for calls to be transferred to a human nurse.
Human oversight of the AI reaching out to patients
Because the system is used in a clinical setting, safeguards were built in to prevent errors or misinformation by using human oversight.
“That is very important because of the stage we are at with AI, especially when it has a clinical impact. By protocol, every single one of the calls done through Sofiya is reviewed by those nurses.”
Deployment of Sofiya required an extensive internal review, passing through about a dozen hospital committees and becoming the first project to go before Mount Sinai’s AI review board. New governance policies and patient-facing disclosures about AI were also required.
Vengrenyuk said his team would like to expand the program to include phone calls placed after procedures. He also thinks it could provide value if used in other departments.
AI helps address growing staffing shortages
The AI pilot program is trying to addresses growing cardiology staffing shortages and clinician burnout. Vengrenyuk said burnout concerns exploded during the COVID-19 pandemic and have only intensified since. Administrative burden is a major contributor to burnout throughout all healthcare specialties, including cardiology.
“Administrative work takes up something like 50-60% of clinician time," Vengrenyuk said.
Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: [email protected]