Methodist Mansfield Hospital successfully uses AI tools to support medical staff

Methodist Mansfield Hospital is now at the forefront of medical technology, having recently implemented artificial intelligence (AI) in its new Neuro Intensive Care Unit.

Methodist Mansfield’s AI technology, developed by brain imaging software company RAPID-AI, was introduced at the hospital late last year and then carried over to its neurological care unit, which opened in February.

Implementing this tool can save staff a lot of time and protect brain tissue in the process.

“This speed will cut our arrival time by five to 10 minutes,” Juan Fresquez, president of Mansfield Methodist Church, told the Arlington City Council. “Just like in a heart attack, when blood flow is restricted, the heart muscle begins to weaken. The same is true with the brain, so every minute counts.”

Methodist Mansfield Hospital is using RAPID-AI technology by running CT scans through software programmed to flag potential abnormalities. If the AI ​​finds anything abnormal, it notifies the medical team. Neuroradiologists then determine if further treatment is needed.

Fresquez then recounted the story of Kristin Booth, 45, who was taken to hospital after waking up unable to speak or move her right arm.

According to the hospital’s magazine, Shine, the mother of five underwent treatment to unclog a large blood vessel, and just two days after the discovery and treatment, she was cheering on her son at a basketball game.

Left untreated, blood clots like Booth’s can cause permanent impairment of walking and speech, as well as paralysis and numbness in the face, arms and legs, but thanks to the AI ​​software’s rapid detection, staff were able to prevent these lasting effects.

But, Fresquez later told KERA, AI technology is not entirely new to medicine.

“Artificial intelligence has been around for quite some time,” Fresquez said, “I’ve been in healthcare since the mid-’80s and had the opportunity to witness the early rise of artificial intelligence. It really started with computer-aided detection (CAD) and image processing, which is just an early version of what we know as artificial intelligence today.”

The use of AI has become a hotly debated topic in recent years. A world in which AI takes over traditional human roles has been consistently depicted in countless science fiction media, such as the Will Smith film “I, Robot” and Disney/Pixar’s “Wall-E.” However, advances in generative AI tools such as ChatGPT and Dall-E have raised concerns like never before.

Actors and writers unions SAG-AFTRA and WGA took steps to block studios from using AI to create content during a historic strike last year.

But “frightening” is not a word that neurointensive care unit manager Rachel Cochrane would use. She argues that AI cannot do what trained doctors and radiologists can do because the technology is not always perfect.

“It’s always going to be a human doctor that diagnoses the patient, and (AI) is not going to replace that human,” Cochran said. “Healthcare is very holistic, and there are a lot of different things that have to be taken into account when you’re thinking about treating people.”

Jambul Viswanatha, vice chancellor for minority health and health disparities at the UNT Health Sciences Center, shares Cochran’s belief that AI will never fully replace humans in the health care industry.

“At best, AI tools will supplement rather than replace experts,” Viswanath says, “surgical outcomes will be better and operations will probably be completed much faster than they are now. But I certainly don’t see robots taking over surgeons’ jobs.”

Viswanatha’s team is currently working on several national projects on artificial intelligence machine learning to “advance health equity and researcher diversity.”

Part of that mission involves ensuring AI tools are trained on diverse datasets, and Vishwanatha says a potential drawback with AI lies in what the tools are learning from.

Collecting data from a single demographic group based on race, ethnicity or gender and then inputting it into a learning system can give doctors and health professionals biased information.

“I think there’s a huge opportunity to leverage AI learning models in surgery,” Viswanatha says, “but I’m also very concerned about how this will impact different communities. Not all communities have access to the surgical care they need, so if a surgical AI model is developed based on existing surgical data, it could impact some communities that aren’t represented in that data.”

Fresquez agrees, citing issues that plagued marginalized communities in the ’80s.

“We need a database that covers all nationalities and genders,” Fresquez says. “If you look back at the early days of imaging, computer-aided detection (CAD) was mainly used in mammography, and it was impossible to detect because there were clear differences in the incidence of cancer among the different nationalities and ethnic groups that were evaluated.”

These cancer databases have since been updated to better serve Black and brown communities: when hospitals first collected the data, only wealthy white women had access to treatment.

Fresquez said Methodist Mansfield is committed to providing diverse results in these databases.

“The more data you collect, the better your results,” Fresquez says. “The size of the database reduces the difference in standard deviations, so the more data points you have, the sharper the data and resulting reports will be.”

RAPID-AI is currently used in over 2,000 healthcare facilities in over 100 countries, helping them collect a wide variety of data.

KERA contacted RAPID-AI about its data collection methods but had not received a response at the time of writing.

Currently, Methodist Mansfield only uses AI in its neurointensive care unit, but Fresquez said the hospital is open to leveraging AI tools in the future.

“We don’t want to wait too long to make a difference,” Fresquez said. “I never thought this technology would be available. I think we’re fortunate enough to be the first ones in the health care system to really make it happen. I’m very optimistic and excited about the future.”

Emmanuel Rivas Valenzuela is KERA’s SPJ News Intern for Summer 2024. Have a tip? Email Emmanuel Rivas Valenzuela at erivas@kera.org.

KERA News is made possible by the generosity of our members. If you find this reporting useful, please consider making a tax-deductible donation today. Thank you.