Renowned Hospital Explores AI to Transform Patient Care

Preparing for a patient consultation often forces Dr. Alexander Ryu, an internal medicine physician at the Mayo Clinic, to sift through dozens, sometimes hundreds, of pages of medical history.

Many individuals seek out the prestigious clinic for third or fourth opinions, bringing along massive, disorganized files from other healthcare providers. Now, a novel artificial intelligence application is assisting medical professionals in reviewing these records more efficiently. It generates pertinent patient summaries, arranges files chronologically, and significantly enhances searchability.

Dr. Ryu noted that this application, named Record Time, saves him between five and 30 minutes of prep work per appointment, varying by case complexity. This recovered time can then be dedicated to direct patient interaction. Furthermore, Record Time acts as a safeguard against overlooking crucial details hidden deep within a file—information that could alter diagnostic or treatment plans.

“We receive a huge volume of these records, tens of millions of pages every year, and we needed a way to find important information in that,” explained Ryu, who also functions as the vice chair of innovation for the Mayo Clinic’s Department of Medicine.

Healthcare integrations are widely viewed as some of the most promising areas for AI advancement. Tech giants such as Google, OpenAI, and Anthropic have introduced health-focused chatbot functionalities, and tens of millions of people are currently utilizing AI to answer medical queries. Additionally, Silicon Valley executives regularly make bold promises about AI curing cancer and various other ailments. However, these claims frequently resemble marketing rhetoric, especially since the major tech firms primarily target broader consumer and enterprise markets.

Nevertheless, Record Time represents just one method by which the Mayo Clinic, recognized globally as a premier hospital system, is leveraging AI with the ambition of enhancing patient outcomes and saving lives. The institution is collaborating with companies including Microsoft and Scale AI to harness its vast repositories of patient data and clinical research for AI development.

According to Dr. Matthew Callstrom, a radiologist and the medical director of the Mayo Clinic’s generative AI initiative, approximately 150 AI models are currently active within the hospital network.

Integrating AI into healthcare is a contentious issue, prompting significant concerns regarding diagnostic accuracy and the protection of patient privacy.

Earlier this month, Traci Tamiko Eto, the former Director of Research Operations at the Mayo Clinic, filed a lawsuit against the organization. She claims she faced retaliation after voicing concerns over the privacy and oversight protocols of certain Mayo AI systems.

Andrea Kalmanovitz, a spokesperson for the Mayo Clinic, declined to comment on the active lawsuit but emphasized the hospital’s commitment. She stated they are “committed to the responsible development and deployment of AI, with privacy, security, transparency and compliance embedded throughout our processes.”

“Our research and clinical innovation are conducted in accordance with applicable laws and regulations and we remain steadfast in upholding the trust patients place in us and respecting their privacy,” Kalmanovitz added in her statement.

‘Potentially life changing’

The profound utility of AI in the medical field stems from its exceptional capability to detect patterns within massive datasets, observed Jason Droege, CEO of Scale AI, the company that partnered with Mayo Clinic on Record Time.

“AI can step in and do a lot of the tedious work that very specialized doctors or medical professionals do to speed up that process — get to more accurate diagnoses, faster so you can treat more people,” Droege explained. “This is an industry where a lot of what doctors are doing, and nurses and others, is pattern recognition.”

Callstrom shared with CNN that his belief in AI's capabilities was solidified in 2016 when he witnessed how the technology could help radiologists pinpoint subtle, preliminary indicators of cancer in medical scans.

Currently, the Mayo Clinic is conducting a clinical trial to determine if AI can accurately identify patients who are at risk for, or who possess, early-stage pancreatic cancer. The hospital has said this implementation could potentially diagnose the disease years before conventional methods. Presently, pancreatic cancer is rarely detected before it has spread regionally or metastasized, at which point the five-year survival rate is a mere 9%, Callstrom noted.

The hospital has also effectively deployed AI to scrutinize patient heart rhythms, successfully predicting the potential development of atrial fibrillation—a condition linked to blood clots and strokes.

“For those patients where you actually find it and identify it, it’s potentially life changing,” Callstrom remarked.

Balancing speed and trust

In developing these tools, the Mayo Clinic pairs technology specialists with medical practitioners to identify the most pressing clinical challenges. A significant aspect of Callstrom’s role is verifying that the clinic’s AI solutions are both accurate and trusted by medical staff and patients alike.

He explained that AI tools undergo a rigorous evaluation akin to a clinical trial. Initially, a tool is tested on a small patient cohort under strict physician supervision. After measuring its performance, the testing phase is gradually broadened to a larger demographic.

Even after widespread implementation, the Mayo Clinic continuously monitors the tool's efficacy.

“On the physician side … we’re skeptical a lot,” Callstrom admitted. “We give them the option to try (a new AI tool) and if they like it, they use it. If they don’t want to use it, they don’t have to. And the best measure of how well we’re doing is the adoption rate.”

Callstrom frequently fields questions from hospital employees regarding the impact of AI on their employment. He maintains that, thus far, jobs are not disappearing; rather, they are evolving.

For instance, the nursing staff collaborated on the creation of an AI system that actively listens and transcribes notes during patient consultations. This innovation could potentially halve the hour or more nurses currently spend daily documenting these visits.

“What it’s doing is letting them spend more time talking to patients,” Callstrom highlighted.

Droege from Scale AI believes the healthcare sector is merely scratching the surface of AI’s potential. However, he warns against prioritizing rapid adoption over precision.

“These predictions where everything is going to be fixed in a year or two, I think that’s wildly ambitious,” he cautioned. “Quality of care is the bar, and then speed … in healthcare, you want to get it right, as fast as possible.”