The integration of cutting-edge technologies continues to reshape the way medical professionals approach diagnostics and treatment. Among these innovations, augmented intelligence (AI) has emerged as a game-changer in the field of pathology as well as other medical specialties.
The use of augmented intelligence can help improve patient outcomes, but it won’t replace the physician, according to College of American Pathologists (CAP) news briefing.
“We really need to be thinking of this as a tool that a physician uses. Just like pathologists use microscopes, we’re now incorporating this additional tool into our workflows,” said AMA member Emily E. Volk, MD, president of the College of the American Pathologists and vice president of pathology and clinical laboratories for the University of Louisville Health System.
Here are some ways AI is being used to revolutionize disease diagnosis and improve patient care as discussed during the news briefing.
Tailor diagnosis to the individual
Machine learning—a subgroup of augmented intelligence—and deep learning can be used in pathology to analyze complex topics such as the molecular basis of cancer.
“The bottom line is pathologists need to make the most accurate and correct diagnosis they possibly can,” said AMA member Eric Glassy, MD, chair of the CAP Information Technology Leadership Committee and a community pathologist in Southern California. “With the assistance of artificial intelligence, image analysis, computational algorithms, I can be much more accurate and provide clinicians a much better picture of what treatment is best to start a patient on their journey.”
Improve efficacy of treatment
Augmented intelligence “can touch many aspects of the drug development process and pipeline,” said Patricia Raciti, MD, member of the CAP Digital and Computational Pathology Committee, and scientific director within the molecular pathology, oncology translational research division at Janssen Research & Development, LLC.
AI can be used “to characterize and quantify the immune cells in and around the tumor and potentially gain a lot of biological insights into the disease mechanisms themselves for the efficacy of treatments that we’re developing,” Dr. Raciti said. “It's really become a bedrock for us to be able to provide the most effective therapeutics, especially in the world of cancer treatment.”
Guide the decision-making process
“The pathologist has access to these marvelous technologies, but ultimately it is the pathologist as the physician practicing anatomic pathology and laboratory medicine making the diagnosis,” said Dr. Volk. “These instruments are not making the diagnosis, that is physician work. These are just tools that augment that ability.”
“It can replace some mundane tasks. It can replace some activities that a pathologist finds more challenging or grueling, and it can allow us the time to properly analyze and think through the diagnosis,” Dr. Glassy said. “But at the end of the day, the pathologist is the one who makes the decision.”
Bridge gaps in care for rural areas
Augmented intelligence inherently learns from expert level data and that “expertise is now a piece of software that can be deployed in resource restricted countries, rural areas, underserved areas,” said Matthew G. Hanna, MD, vice chair of CAP Artificial Intelligence Committee and director of digital pathology informatics at Memorial Sloan Kettering Cancer Center.
“And all of these learnings now can be used for those pathologists who already have a shortage and also be a virtual consultant,” Dr. Hanna said. “This democratization of pathology expertise is really one of the big promises and advantages to deploying these sorts of software across the world.”
“It’s easy to forget the role that pathologists can play, even from the laboratory, in contributing to bridging gaps in health equity and this is one of those opportunities,” Dr. Volk said.
Support with ChatGPT
“There’s truth and misinformation within that body of knowledge that ChatGPT has access to, so it’s only as good as what it’s been trained on,” Dr. Glassy said. “But there are new ChatGPT applications which are customized to only securely look at medical data or … at the information in a patient’s medical record.”
“When you apply that software technology to localized sites or it’s trained only on journals or articles that are peer reviewed, then you’ve got a wonderful opportunity to engage pathologists and improve patient care,” Dr. Glassy said.
