The October 2022 issue of IEEE Spectrum is here!

Close bar

AI Cardiologist Aces Its First Medical Exam

A neural network outperforms human cardiologists in a task involving heart scans

3 min read
Photo-illustration of an AI computer analyzing heart scans.
Photo-illustration: Rima Arnaout

Rima Arnaout wants to be clear: The AI she created to analyze heart scans, which easily outperformed human experts on its task, is not ready to replace cardiologists. 

It was a limited task, she notes, just the first step in what a cardiologist does when evaluating an echocardiogram (the image produced by bouncing sound waves off the heart). “The best technique is still inside the head of the trained echocardiographer,” she says.

But with experimental artificial intelligence systems making such rapid progress in the medical realm, particularly on tasks involving medical images, Arnaout does see the potential for big changes in her profession. And when her 10-year-old cousin expressed the desire to be a radiologist when she grows up, Arnaout had some clear advice: “I told her that she should learn to code,” she says with a laugh. 

When both the AI and expert cardiologists were asked to classify the images, the AI achieved an accuracy of 92 percent. The humans got only 79 percent correct.

Arnaout, an assistant professor and practicing cardiologist at UC San Francisco, is keeping up with the times through her research in computational medicine; she published this new study in the journal Digital Medicine

In the study, Arnaout and her colleagues used deep learning, specifically something called a convolutional neural network, to train an AI system that can classify echocardiograms according to the type of view shown.

This classification is a cardiologist’s first step when examining an image of the heart. Because the heart is such a complex structure—it’s an asymmetrical organ with four chambers, four valves, and blood constantly flowing in and out through several vessels—echocardiographers take videos from many different positions. When the doctors are ready to analyze those videos, they first have to figure out which view they’re looking at and which anatomical features they can see. 

Typically the cardiologist would look at a relatively high-resolution video of the echocardiogram, showing a shifting image captured as the imaging tool was moved around the patient’s chest. But the AI had a much harder task. It was given still images taken from video clips, and the images had been shrunk to just 60 by 80 pixels each. 

Six echocardiogram images showing different views of the heart.

Image: A Madani et al.
The AI had to sort heart scan images into categories based on which view of the heart they presented.

When both the AI and expert cardiologists were asked to sort these tiny black-and-white images into 15 categories of views, the AI achieved an accuracy of 92 percent. The humans got only 79 percent correct. “These were excellent echocardiographers,” Arnaout says, “but it’s a hard task. We’re not used to seeing the images shrunken down and out of context.”

The AI only performed this first step in the analysis of a heart image and the making of a diagnosis. A human cardiologist looks at many of these scans to examine more than 20 structures within the heart, then synthesizes that information to arrive at a conclusion.

Arnaout is now working on a new version of the technology that can take the next steps to identify different diseases and heart problems. “A human echocardiographer can look at any heart, no matter what the defect, and figure out what’s going on,” Arnaout says. “I’m interested in building a platform that can do that.”

Even if she accomplishes her goal, though, she doesn’t think human cardiologists will be put out of their jobs. “As cardiologists, we read the images and then go see the patient,” she says. “So we’re both reading images and practicing medicine. I don’t think that second piece will be taken over so quickly.”

The Conversation (0)

Restoring Hearing With Beams of Light

Gene therapy and optoelectronics could radically upgrade hearing for millions of people

13 min read
A computer graphic shows a gray structure that’s curled like a snail’s shell. A big purple line runs through it. Many clusters of smaller red lines are scattered throughout the curled structure.

Human hearing depends on the cochlea, a snail-shaped structure in the inner ear. A new kind of cochlear implant for people with disabling hearing loss would use beams of light to stimulate the cochlear nerve.

Lakshay Khurana and Daniel Keppeler
Blue

There’s a popular misconception that cochlear implants restore natural hearing. In fact, these marvels of engineering give people a new kind of “electric hearing” that they must learn how to use.

Natural hearing results from vibrations hitting tiny structures called hair cells within the cochlea in the inner ear. A cochlear implant bypasses the damaged or dysfunctional parts of the ear and uses electrodes to directly stimulate the cochlear nerve, which sends signals to the brain. When my hearing-impaired patients have their cochlear implants turned on for the first time, they often report that voices sound flat and robotic and that background noises blur together and drown out voices. Although users can have many sessions with technicians to “tune” and adjust their implants’ settings to make sounds more pleasant and helpful, there’s a limit to what can be achieved with today’s technology.

Keep Reading ↓Show less
{"imageShortcodeIds":[]}