The Surprising Ways AI Can Transform Surgery

(SeaPRwire) –   In a landmark event last summer, an artificial intelligence system observed 17 hours of surgical footage and subsequently performed a gallbladder removal entirely on its own. This achievement marked a significant milestone, demonstrating AI’s capacity to function within the operating theater beyond a mere supportive role. However, this feat also overshadowed the more profound potential of AI to aid surgeons, an impact that extends far beyond the act of cutting tissue.

The surgical journey for a patient typically starts long before the first incision and continues well after they leave the operating room. It may be triggered by a persistent knee pain, an accidental fall, or concerning results from a radiology scan. Patients are then confronted with a series of critical decisions, from opting for surgery to selecting the most suitable surgeon and surgical approach. The process persists long after wound closure, involving physical therapy, scar management, and the often slow and frustrating path to recovery. It is across this extensive timeline—filled with intricate discussions and choices—that AI holds the key to revolutionizing surgical care.

AI is already reshaping patient care in numerous other medical fields. Over 80% of physicians in the United States report using AI tools as a knowledge co-pilot to verify diagnoses, determine medication dosages, draft clinical notes, analyze medical images, and interpret lab results. New applications are emerging to help clinicians answer specific questions, handle administrative tasks automatically, and synthesize complex information.

Surgery, by contrast, introduces a distinct set of challenges.

Each year, over 4 million individuals die within 30 days of a surgical procedure, establishing postoperative mortality as a major global cause of death. Complications can develop in the weeks leading up to an operation, where factors like a patient’s blood sugar levels, nutritional status, and anxiety influence how their body will handle the stress of surgery. Other issues can arise during the extended recovery phase, when skipped therapy sessions or ignored aftercare instructions impede healing. An AI system capable of supporting both patients and medical staff throughout this entire continuum would be significantly more impactful for surgeons than any autonomous robot in the OR.

We are now witnessing the initial steps of AI addressing these issues. Advanced algorithms can pinpoint which patients face the greatest risk of complications before they ever enter the operating room. Another AI application tracks the walking speed of hip or knee replacement patients during their recovery and notifies their physician if progress is lagging. With over 2 million robotic-assisted surgeries already conducted annually in the U.S., AI-driven simulators are also revolutionizing the training of surgical residents.

The potential applications for AI extend to many other areas. A highly desired tool would be an AI system that assists patients in choosing the optimal surgeon and procedure for their specific needs, offering impartial advice on the advantages and disadvantages at each decision point. Conversational AI tools could alleviate the fear associated with surgery by delivering emotional support and tailored guidance before and after the operation. Surgery is a long-term, physically demanding, and frequently irreversible process. Its success relies on a team including caregivers, physical therapists, case managers, home-health aides, and the patient’s own dedication—each element representing an opportunity for AI support that remains largely untapped.

Nevertheless, progress must be approached with caution. A 2024 study revealed that AI systems suggested different diagnostic tests for patients based on their race and gender. In a surgical context, such bias influences more than just diagnosis. It affects who is even considered a candidate for surgery, the type of procedure recommended, the level of postoperative care provided, and whether a patient recovers with adequate support or alone.

Other critical questions remain. If an AI program makes an error, who is held accountable? AI systems already have the capability to assess a surgeon’s skill and performance during robotic operations—should this performance data be publicly available? Furthermore, as AI assumes more surgical tasks, could the essential skills gained only through hands-on experience be lost by future generations of surgeons?

Developers and healthcare institutions creating new surgical AI technologies should avoid focusing exclusively on the most visible applications, such as a robotic arm performing a procedure autonomously. Work of equal importance occurs in the days and weeks preceding a patient’s surgery and the months following it. Preventing just one-third of current surgical complications would save millions of lives and tens of billions of dollars annually.

The next major advancement in surgical AI may not generate the same headlines as a robot removing a gallbladder. Instead, it might manifest as a patient identifying a complication before it escalates into an emergency, or successfully completing physical therapy because an app detected they were falling behind. When applied to the 300 million surgeries performed worldwide each year, this is where the true potential to save lives exists.