Artificial intelligence in medicine is on the rise

Don’t be surprised if a clinician spends a few minutes with a stethoscope and an iPad the next time you’re at a primary-care checkup in Naples.

It’s the next generation of digital stethoscopes able to detect the faintest of sounds of possible irregularities in your heart. Combined with an artificial intelligence application, an immediate analysis is done if there is a heart murmur or something else that needs attention.

The NCH Physician Group in early September is launching a three-month pilot with the technology developed by Eko Health, a California-based digital health company. The system was released for use in March.

It’s just one example of how NCH Healthcare System is exploring and using artificial intelligence, similar to how Lee Health in Lee County has been using and expanding its use for quite some time. Physicians Regional Healthcare System in Collier is using it for stroke care.

Artificial intelligence applications are already used behind the scenes in hospitals to automate workforce tasks, improve patient flow, for operating room scheduling and elsewhere to improve efficiency.

More:AI can help doctors with breast cancer risk predictions, a new study found

Other uses involve medical imaging for more accurate evaluations of cancer and diseases which have decreased errors, improved timeliness in diagnoses while reducing physician workloads, according to the National Institutes of Health.

To help the healthcare industry navigate the growth of artificial intelligence tools in medicine, a much-anticipated blueprint with specific recommendations was released in April by the Coalition for Health AI.  Coalition members come from academic centers, hospitals, artificial intelligence practitioners and elsewhere.

What’s happening in Southwest Florida

NCH has selected five physicians to test out the new Eko stethoscopes in concert with the company’s artificial intelligence software, said Dr. Bryan Murphey, president of the physician group and ambulatory chief medical officer at NCH.

More:Lee Health examines pros and cons of changing to private, nonprofit system

In time, all primary-care physicians in the group will be using them.

The new stethoscopes amplify the sounds in the heart for better detection of heart murmurs or more advanced valve disease, Murphey said.

The findings are uploaded into a software platform and an analysis is available within minutes. The physician can tell patients right away what’s going on.

Heart disease is the leading cause of death in the U.S. and was responsible for one in five deaths in 2021, according to the U.S.  Centers for Disease Control and Prevention. The disease costs more than $240 billion in 2019 in healthcare services and lost productivity.

Eric Davidson, president and chief operating officer of Eko, said the software platform detects 90% of structural murmurs which is more than double the 43% found without it.

More:NCH Heart Institute gets preliminary approval from Naples design board

NCH is one of five hospitals in the U.S. piloting the stethoscopes with the artificial intelligence database. A South Florida private concierge physician practice was also used as a test site, Davidson said.

What about strokes?

At Physicians Regional in Collier, the radiology department has been using artificial intelligence to process some CT scans for patients since 2021, said Sarah Strzalka, director of stroke and quality.

The artificial intelligence helps clinicians identify large vessel occlusions that need a mechanical thrombectomy to remove the blood clot in the brain, she said.

Physicians Regional plans to adopt other diagnostic imaging capabilities with artificial intelligence for identifying different categories of stroke, she said.

“Our stroke team loves utilizing AI for imaging because it is much faster and better images than the traditional imaging processing,” she said.

Since Physicians Regional started using the artificial intelligence platform, it has been used with 3,770 patients and 155 large blockages have been identified, she said.

NCH has also been using an artificial intelligence tool for more than a year during colonoscopies which helps the physician identify polyps that sometimes are not easy to detect in the folds of the colon, Murphey said.

“So it is back-up technology,” Murphey said. “It is standard for our (group).”

Improved operations; patient outcomes

Lee Health has utilized artificial intelligence to improve internal operations for efficiency and in clinical care applications with imaging for mammography, stroke care, and heart disease, according to Jonathan Witenko, system director of virtual health and telemedicine.

The hospital system also uses artificial intelligence applications to predict patient risks in multiple ways, such as for readmissions, for surgical site infections, for falls and more.

Projects in the works or near future will be in the area of precision medicine which take into account patients’ individual characteristics for disease treatment.

Another application will help with patient movement in the hospital for their care from the time of admission to discharge, he said.

The goal is to make sure everything is getting done efficiently, including when housekeeping comes in to get the room ready for the next patient.

“It is continuously evolving,” Witenko said.

Improving efficiency in operating room scheduling is another way artificial intelligence is being used to determine when the different staff support need to arrive and when the patient should be brought in, Witenko said.

Another area is for prior authorization with insurers for diagnostics and procedures, said Dr. William Carracino, Lee Health's vice president and chief digital health executive. Through artificial intelligence automation, the prior authorization process can be done any time of day or night.

For patients, that means improved access to services they need, Carracino said.

What do consumers think?

Hospitals which have struggled from financial and manpower losses during the COVID-19 pandemic could save upwards of $360 billion annually over the next five years via artificial intelligence uses, according to an analysis by the National Bureau of Economic Research released in January.

“These opportunities could also lead to non-financial benefits such as improved healthcare quality, increased access, better patient experience, and greater clinician satisfaction,” according to the study.

Consumers are mixed on how they see artificial intelligence being used for their medical care, according to a Pew Research Center survey released in February that involved 11,000 adults.

Six in 10 respondents, or 60%, said they would feel uncomfortable if their own healthcare provider uses artificial intelligence to diagnose a disease and recommend treatment, the results show.

Roughly 38%, or four in 10 adults, said they think it will reduce mistakes while 33% said it would lead to worse outcomes; 27% said it would not make much difference.

The security of patient health records is an issue as 37% of the respondents said using artificial intelligence in medicine would make the security of their records worse while 22% said it would improve security.