a pathologist in a lab coat stands at a computer monitor reviewing data using AI to analyze blood smears.

Scopio Labs developed a system that uses to analyze blood smears to help pathologists diagnose leukemia, anemia, heart failure, lymphoma, autoimmune disorders and malaria.

'We're on the Precipice'

AI powers tools that help pathologists spend less time counting cells and use their refined skills to make complicated diagnoses.

Laura Gates Dec 17, 2024

FOR OVER A hundred years, pathologists have worked with their faces planted in microscopes, meticulously counting cells and examining biological samples on glass slides.

Now, pathologists are shifting their gaze to computer screens — and the assistance being provided by artificial intelligence. What the future holds: quicker diagnoses, better predictions about patient response to treatments, and more accurate estimates of their overall prognosis.

“We’re early in the digital pathology space,” said Michael Feldman, MD, PhD, chair of the Department of Pathology and Laboratory Medicine and the Manwaring Professor of Pathology and Laboratory Medicine at the Indiana University School of Medicine. “We’re on the precipice.”

Until recently, AI was primarily used in research. Now it’s making its way to the clinic. Feldman has been at the forefront, first envisioning the potential for AI about 25 years ago while working alongside Anant Madabhushi, PhD, a global leader in using AI to better understand and treat diseases.

Recruited from the University of Pennsylvania in February 2023, Feldman is now teaming up with AI leaders from around IU School of Medicine. He says the school’s strengths in data science and analytics, along with its access to data on patients around the state, are valuable assets. “There’s a huge opportunity to move this science forward,” he said.

Feldman was part of a team of scientists that validated a new digital imaging system that uses AI to analyze blood smears. Developed by Israel-based Scopio Labs, the system employs a new approach for tests used to diagnose leukemia, anemia, heart failure, lymphoma, autoimmune disorders and malaria. Its big advantage is speed: Scopio estimates its technology reduces analysis turnaround time by 60 percent.

IU Health became the first health system in the nation to install a Scopio machine in the summer of 2023. It was stationed at IU Health’s centralized pathology lab in Indianapolis. Within a month, the system proved its value.

A severely ill child crashed in the emergency room. The presence of a high percentage of immature cells in the patient’s blood prompted a pathological review. The doctor on hand needed answers — quickly.

Manual reviews involve packaging samples, calling a courier, and shipping everything to the central lab for expert examination under the microscope. They can take up to five hours, depending on travel time. Scopio fast-tracked the process. In about 10 minutes, it enabled a remote diagnosis: leukemia.

Micheal Feldman Headshot

“Before I retire, I think we’ll be all digital. There are already some pathologists who practice without a scope, in this country and around the world.”

Micheal Feldman, MD, PhD

Such capabilities are important for IU Health, a 16-hospital, statewide system with some locations an hour or more from the central pathology lab. Scopio not only saves time but also can increase competency, said Jessica Isaac, executive director of the pathology laboratory at IU Health.

“It’s kind of like ‘phone-a-friend,’” she said. “If I’m working alone in the middle of the night in Frankfort and I see something weird on a slide, I would be able to send the link and tag cells on the slide so when the pathologist opens the file, it goes directly to that cell with my notes. It simulates that shoulder-to-shoulder environment where you can get a second opinion.”

IU Health soon plans to install Scopio systems in all its regional pathology labs.

Scopio was recently FDA approved to read bone marrow samples to help hematopathologists diagnose blood cancers and other diseases more rapidly.

“To read bone marrow, somebody has to sit and count 500 cells,” Feldman said. “For a technician with a fairly skilled set of eyes and hands, it can take 15 to 20 minutes,” he said. Scopio can do it in seconds. “That kind of efficiency is a good thing.”

Beyond the clinical benefits, AI can accelerate scientific discoveries. If patient data is added to a big-data system, machine learning can make connections leading to better models for prediction and treatment of diseases, Feldman said.

AI can also be a powerful tool for teaching. If a pathology resident is struggling with a concept, the resident might request slide archives to look at under the microscope. Those slides would need to be pulled from central storage and physically delivered.

“If we were thoughtful about this, we would have digital libraries of slides where the resident could look at 50 cases and really learn the topic self-service style,” Feldman said.

As the field transitions to digital, current pathology trainees must learn both old ways and new techniques. But change is coming. “Before I retire, I think we’ll be all digital,” Feldman said. “There are already some pathologists who practice without a scope, in this country and around  the world.”

Donor support can help the Department of Pathology attract talented researchers and back bold research. Learn more about its priorities and make a gift today.

Tags: Artificial Intelligence, Pathology