Skin cancer is a prevalent problem in today’s society, and it is usually first detected with irregular looking moles or legions on the skin. But scientists have now created an “automated dermatologist” that can detect skin cancer as accurately, or better, than a doctor by just looking at a photograph of the problem area.

The research that led scientists and dermatologists to create the technology has been published in the journal Nature. 

In order to create this impressive AI system, computer scientists at Stanford University “trained” the program to determine whether or not a mole was cancerous by creating a database of over 13,000 photos of skin cancer for it to analyze.

The algorithm was then able to use the information to decide whether a mole looked benign or malignant.

After the computer was programmed, it was tested against 21 working dermatologists to determine its accuracy in diagnosing skin cancer.

Researchers found the computer matched or exceeded the performance of the doctors.

The team hopes that one day service users will be able to user their smart phones to have parts of their skin checked if they are unable to see a doctor or one is not available in their area.

A future app could also save money on preliminary doctor’s visits, as well as save time for both the patient and doctor if the mole is determined to be harmless by the app.

Researchers on the study spoke about why they decided to develop the technology:

“In the United States, there are only 10,000 dermatologists. But there are 360 million Americans. There are 5.4 million new cases of skin cancer in the US every year. One in 5 Americans will be diagnosed with skin cancer in their lifetime.

Although melanomas represent fewer than 5% of all skin cancers in the United States, they account for some 75% of all skin-cancer-related deaths, and responsible for over 10,000 deaths annually in the United States alone. Early detection is critical, as the estimated 5-year survival rate for melanoma drops from over 99% if detected in its earliest stages to about 14% if detected in its latest stages.”

In the future, perhaps this technology will lead to further developments that allow for virtual diagnoses.


Related Articles