Can AI detect skin cancer? FDA authorizes use of device to help doctors identify suspicious moles.
Can AI be used to treat cancer patients? New study aims to find out
01:41
The Food and Drug Administration has authorized the first artificial intelligence-powered medical device to help doctors detect the most common forms of skin cancer in patients.
The technology, from Miami-based medical device maker DermaSensor, is used to further evaluate lesions that doctors have already flagged as suspicious and is not meant to be used as a screening tool, according to the FDA.
More specifically, the non-invasive, handheld device uses AI-powered spectroscopy tech to assess cellular and below-the-skin’s-surface characteristics of lesions on patients. The device, also called DermaSensor, provides real-time results based on an AI algorithm that is trained on data related to more than 4,000 malignant and benign lesions, according to the company. It then delivers a “spectral similarity score” to known cases in order to complement a physician’s own assessment of a mole or lesion.
statement announcing the FDA clearance.
In addition to helping spot melanoma, which is the most deadly form of skin cancer, the device can also assess moles for basal cell carcinoma and squamous cell carcinoma.
One in five Americans will have developed a form of skin cancer by the age of 70, according to the American Academy of Dermatology, which puts the cumulative cost of treatment in the U.S. at more than $8 billion. Most skin cancers are curable if detected early.
In approving the DermaSensor device, the FDA is requiring that the company conduct additional validation testing in patients from broadly representative demographic groups, including those who are at lower risk of skin cancer.
Megan Cerullo
Source: cbsnews.com