Artificial intelligence (AI) is a main topic of conversation in the healthcare and life sciences industries, with specialties like oncology quickly taking advantage of what AI tools have to offer.
While there are still many challenges ahead, including lingering issues around ethics, equity, and trustworthiness, there’s no question that AI has the potential to dramatically change the way humans interact throughout the care process.
At the Clinical Pathways Congress + Cancer Care Business Exchange event held in Boston September 6-8, 2024, a panel of AI and oncology experts convened to share their perspectives on how AI is already having an impact on oncology as well as their predictions for how next-generation models will help support better outcomes and lower costs in the future.
Here are their top five areas where AI is making a difference in the way researchers, oncologists, and patients engage in cancer care.
Radiology and pathology have always been on the leading edge of AI adoption, and these specialties are continuing to benefit from advances in generative AI (genAI) and other models, said Douglas Flora, MD, Executive Medical Director at St. Elizabeth Healthcare and Editor-in-Chief of the AI in Precision Oncology journal.
“AI is an exceptional tool for helping to read a mammogram, MRI, CT scan, or digital slides for pathologists,” he said. “We’re leaning heavily into AI tools for our imaging departments, and we have already detected dozens of incidental cancers on scans within the last year or two, allowing those patients to get earlier treatment. It’s a very solid use case that’s already bearing fruit.”
Dr. Flora added that many of the manuscripts flooding his editor’s inbox are focused on getting predictive around patient responses to emerging therapy types.
“Predicting responses to immunotherapy would be a holy grail in many regards,” he said. “We could make better decisions up front, or pivot more quickly, if we had a better idea of who is likely to do well on these therapies. It’s time-consuming and very expensive to go down the wrong path for too long. Using AI to start off on the right foot would bring benefits to patients and cancer centers alike.”
AI tools are highly effective at sifting through massive volumes of multi-source data to detect patterns that are generally hidden from human eyes. This makes them ideal for picking out red flags that might raise an individual’s risk of developing a cancer or experiencing a complication during treatment.
“Many cancer centers are participating in value-based care models, which makes it even more important to be able to proactively address clinical issues that might otherwise lead to an ED visit or hospitalization,” explained Jeff Hunnicutt, CEO of the Highlands Oncology Group, based in Arkansas. “AI tools are becoming incredibly helpful for equipping nurse navigators with a targeted task list so they can reach out proactively to higher risk individuals and hopefully prevent negative outcomes.”
Ambient listening systems, which bring automated documentation generation to the exam room, are making waves across the industry. Both Dr. Flora and Hunnicutt have experience with these products, and believe that such AI-powered, voice-enabled documentation solutions can improve patient care and provider experiences.
“It’s become remarkably helpful for building a clinical note that is very close to complete,” said Hunnicutt. “It’s had a huge benefit on the clinical flow and the ability to really focus on the patient while you’re in the room with them.”
“It significantly reduces the time required for a provider to review and complete documentation, which helps them see more patients, focus on those patients better, and get home on time at the end of the day. I’m hopeful we’ll see much more of these types of tools to streamline workflows.”
A very long, complex process of R&D precedes the treatment that happens in the clinic, and there is a huge amount of scientific data to comb through before expert organizations can recommend optimized treatment pathways and clinical guidelines for frontline providers.
AI is playing a central role in aggregating, curating, and surfacing data of interest so researchers can zero in on promising approaches, said Will Shapiro, VP of Data Science & AI at Flatiron Health.
“My team currently combs through billions of patient documents to extract relevant variables to better understand treatment patterns and support recommendations. AI has made this feasible to do at scale; we’ve been able to go from learning from the experience of 400,000 patients to learning from 4 million patients, with a high degree of accuracy.”
“Having AI help us understand all that detail and rich clinical history. Paired with genomic data, clinical data at scale, is game-changing for research, developing new therapies, guideline development, and understanding the impact of care on specific populations. It’s the gateway to truly personalized medicine for everyone, and we couldn’t be moving as fast towards that goal if we didn’t have AI.”
To learn more about how AI-enabled solutions can support your cancer center, contact Flatiron Health.