Preoperative planning is ripe for AI to assist

Surgical care encompasses more than the procedure itself, spanning preoperative planning to postoperative follow-up—all areas where artificial intelligence holds potential.

“In surgery, the scope is quite broad,” said Thomas Fuchs, who this month joined the Icahn School of Medicine at Mount Sinai as dean of AI and human health, as well as co-director of the Hasso Plattner Institute for Digital Health at Mount Sinai. “If you start at the beginning, before you even go into surgery, the right diagnosis is key.”

AI could support pathologists in detecting diseases and radiologists in preoperative planning, among other specialties that participate in the episode of care.

Fuchs is a co-founder of Paige.AI, a spinoff of Memorial Sloan Kettering Cancer Center that uses AI to help diagnose cancer patients. Last year, the startup was granted breakthrough-device designation by the Food and Drug Administration.

The Cleveland Clinic’s epilepsy center uses AI algorithms—some developed externally and some developed by the clinic—to review MRI scans of the brain and identify where seizures are coming from, so that surgeons know which area to remove in surgery. The center started putting together that set of algorithms about a decade ago.

Around 25% of patients with difficult-to-treat seizures and who are planning to have brain surgery still present as having “normal” MRIs, said Dr. Imad Najm, the epilepsy center’s director.

With AI, the center is able to identify lesions in about half of those patients’ medical images.