Can image recognition technologies diagnose skin cancer?

Researchers from Stanford University have trained a computer to detect skin cancer moles and lesions just as precise as a dermatologist would.

The study, published in Nature, has data suggesting that a cell phone app could be able to help patients diagnose skin cancer on their own. With skin cancer being one of the most common cancers in the United States, the team from Stanford developed this image recognition technology to bring dermatology expertise to areas where a dermatologist is not available.

Dermatologists will usually look at a mole and if they find abnormalities, they have the patient follow up with biopsies and tests. With this new automated dermatologist, patients may now be able to get a diagnosis much more quickly. 

Find out more on how the 'automated dermatologist' works:

 

Jodelle joined TriMed Media Group in 2016 as a senior writer, focusing on content for Radiology Business and Health Imaging. After receiving her master's from DePaul University, she worked as a news reporter and communications specialist.

Around the web

Back in September, the FDA approved GE HealthCare’s new PET radiotracer, flurpiridaz F-18, for patients with known or suspected CAD. It is seen by many in the industry as a major step forward in patient care. 

After three years of intermittent shortages of nuclear imaging tracer technetium-99m pyrophosphate, there are no signs of the shortage abating.

GE HealthCare said the price of iodine contrast increased by more than 200% between 2017 to 2023. Will new Chinese tariffs drive costs even higher?