What is the future of AI in healthcare?
- Laurie D. Babcock
- April 4, 2024
- Learning
- Contributors: Rendy Ramos, Video | James Greco, Photography
5 – minute read
From students interacting with holographic patients to researchers using pattern recognition to detect cancer cells digitally, artificial intelligence (AI) is being explored at Florida Gulf Coast University as a tool in the healthcare setting.
FGCU360 sat down with two faculty members working on AI applications in Marieb College of Health & Human Services and U.A. Whitaker College of Engineering.
Paging Dr. Hologram
Princess Leia projected from R2D2 in “Star Wars: A New Hope.” The shark from the fictional “Jaws 19” bearing down on Marty McFly in “Back to the Future Part II.” The idea behind the holodeck on the TV series “Star Trek: The Next Generation.”
These are examples of holograms, lifelike 3D images made of nothing but light creating the illusion of solid objects. The technology has been around for 60 years, but its use in healthcare education is new at FGCU. Marieb College of Health & Human Services recently acquired seven units that simulate holograms from Proto Hologram.
“Access to this technology is exciting for students as it helps to enhance immersive learning experiences,” says Traci Grove, simulation education coordinator and a nursing instructor in Marieb College.
The college has one EPIC Hologram, about 4.5 feet wide and a little over 7 feet tall, that can deliver a life-size human holographic-style image. Using 2D images or video, the unit creates a 3D-like experience. The other six units in Marieb College are tabletop models, making them easy to transport from classroom to classroom.
All seven units have the same capabilities — each has a smart camera enabling two-way communication and projects 4k images and video.
Eventually, the units will be housed in a new center planned for the college’s next academic building.
Joe Buhain, recently hired as director of interprofessional simulation and emerging technology, will oversee the new center and help students learn by simulating various healthcare situations. With the help of advanced technology, they can experience realistic scenarios and develop skills needed to provide better patient care.
“I think the recent phenomenon of hologram technology really stems from advancements in its accessibility and integration. Over the past few years, improvements in hardware, software and networking have made this technology more feasible — especially in the academic and healthcare realm,” says Grove.
Now that the hardware is in, Marieb College will introduce the software to students this semester.
Called Dr. Hologram, it’s a telepresence healthcare technology that allows students to interact with patients in a safe environment.
“We have the ability to record different cases, depending on our objectives, and either present the case as a pre-recorded hologram to the student or live beam standardized patients, or even guest speakers, into the classroom,” says Grove.
A patient, represented as a holographic-style image, is beamed to any of the seven units. Students hear a pre-recorded presentation or have a real-time interaction. The patient responds to questions, provides information about symptoms and engages with students just as if he or she were physically present.
“As a healthcare simulationist, my goal is to create realistic environments to increase the ability of the learner suspending disbelief,” says Grove.
Interacting with virtual patients like they’re really in the room will, hopefully, enhance their overall learning experience and bridge the empathy gap.
It worked for Princess Leia and her plea for help.
AI takes aim at cancer
Most of us have been asked to prove we are not bots by clicking on bicycles, hills or buses in a picture array before accessing a secure online account. According to one biomedical engineer and FGCU professor in computing and software engineering, that same pattern recognition idea is being used by artificial intelligence to find cancer.
“Unfortunately, most of us will experience cancer within our close circle eventually,” says Pelin Ciris of U.A. Whitaker College of Engineering. “But generally, the outlook for any cancer improves if detected — and treated — early.” She applies the principles and problem-solving techniques of engineering to biology and medicine, including cancer detection research.
Over 100 types of cancers exist, affecting nearly every part of the body. “Cancer detection is a very complex problem being attacked from many perspectives. AI can help in nearly every stage,” Ciris says.
“First of all, AI doesn’t know that it is detecting cancer. AI is essentially very good at recognizing patterns, and gets better when it’s trained on more examples of the patterns it is asked to identify or distinguish from one another,” she says.
Biomedical engineers train algorithms on data, including electronic health records and diagnostic tests (like MRI, X-rays and colonoscopies), to distinguish between cancers and healthy tissues.
According to the National Institutes of Health, machine learning (ML) — a subset of AI that enables computers to learn from training data — “has been highly effective at predicting various types of cancer, including breast, brain, lung, liver and prostate cancer. In fact, AI and ML have demonstrated greater accuracy in predicting cancer than clinicians.”
FGCU engineering students are learning these techniques in theory but also in practice from faculty like Ciris. She teaches a course on processing digital images, and her colleague Baharul Islam teaches another course on analyzing them.
“The techniques we’re teaching are widely applicable, and our students are being exposed to many areas from astronomy — think of the rover on Mars — to manufacturing, like automatic defect identification,” Ciris says.
“Although medical imaging is indeed my favorite still,” she adds.