AI-produced images cannot solve diversity issues in dermatology databases

Image databases of skin conditions are notoriously biased towards lighter skin tones. Rather than waiting for the slow process of collecting images of conditions such as cancer or inflammation on darker skin, one group wants to fill in the gaps using artificial intelligence. He’s working on an AI program to generate synthetic images of diseases on darker skin – and is using those images for a tool that could help diagnose skin cancer.

“Having real images of darker skin is the ultimate solution,” says Eman Rezk, a machine learning expert at McMaster University in Canada who is working on the project. “Until we have that data, we need to find a way to close the gap.”

But other experts working in the field worry that using synthetic images could introduce their own problems. The focus should be on adding more diverse real images to existing databases, says Roxana Daneshjou, a clinical dermatology researcher at Stanford University. “Creating synthetic data seems like an easier path than working hard to create a diverse dataset,” she says.

There are dozens of efforts to use AI in dermatology. Researchers are creating tools that can scan images of skin rashes and moles to determine the most likely type of problem. Dermatologists can then use the results to help them make diagnoses. But most of the tools are built on image databases that don’t include many examples of conditions on darker skin or don’t have good information about the range of skin tones they have. include. This makes it difficult for bands to be sure that a tool will be as accurate on darker skin.

That’s why Rezk and the team turned to computer graphics. The project has four main phases. The team has already analyzed available image sets to understand just how underrepresented darker skin tones are. He also developed an AI program that used images of skin conditions on lighter skin to produce images of those conditions on dark skin and validated the images the model gave them. “Thanks to advances in AI and deep learning, we were able to use available white scan images to generate high-quality synthetic images with different skin tones,” Rezk explains.

Next, the team will combine the synthetic images of darker skin with real images of lighter skin to create a program that can detect skin cancer. They will constantly check image databases to find new real images of skin conditions on darker skin tones that they can add to the future model, Rezk says.

The team isn’t the first to create images of synthetic skin – a group that included Google Health researchers published a paper in 2019 outlining a method to generate them, and it could create images of different skin tones . (Google is interested in AI in dermatology and last spring announced a tool that can identify skin conditions.)

Rezk says the synthetic images are a stopgap until there are more real images of conditions on darker skin available. Daneshjou, however, is concerned about using synthetic images, even as a temporary fix. Research teams should carefully check whether the AI-generated images would have any usual quirks that people couldn’t see with the naked eye. This type of quirk could theoretically skew the results of an AI program. The only way to confirm that synthetic images work as well as real images in a model would be to compare them with real images – which are rare. “Then comes back to, well, why not just try to get some more real footage?” she says.

If a diagnostic model is based on synthetic images from one group and real images from another — even temporarily — that’s a problem, Daneshjou says. This could lead to the model working differently on different skin tones.

Relying on synthetic data could also make people less likely to push for real and diverse images, she says. “If you’re going to do this, are you actually going to keep doing the work? she says. “Actually, I’d like to see more people working to get real data that’s diverse, rather than trying to do this workaround.”


Not all news on the site expresses the point of view of the site, but we transmit this news automatically and translate it through programmatic technology on the site and not from a human editor.
Back to top button