UC San Diego researchers have developed an artificial intelligence system to detect the likelihood of skin cancer in all ethnicities.
KNX News’ Emily Valdez reported how the AI model uses genetics along with things like income, age, and lifestyle factors to detect the likelihood of skin cancer. For people with European ancestry, the model is 90% accurate. For people with non-European ancestry, it’s 81% accurate.
“One of the reasons why this is really important is because people with a non-European ancestry, they're more difficult to diagnose skin cancer in,” Dr. Kelly Frazer, the study author, said.
She explained how the AI tool can determine who will need a full body scanned by a dermatologist.
Frazer pointed out people with darker skin tones tend to get cancers in less obvious places like the bottom of their feet or in their mouth. In these cases, skin cancer is often detected later and has poorer outcomes.
Want to get caught up on what's happening in SoCal every weekday afternoon? Click to follow The L.A. Local wherever you get podcasts.
“Using the genetics of individuals in combination with their clinical histories is how these tools are likely going to work well,” she said.
The AI system is not publicly available yet, but Frazer said it will be in the future.
Follow KNX News 97.1 FM