David Papp Blog

Now Your Phone Can Help Detect Early Skin Cancer

“Watson,” the machine that defeated Jeopardy Champions Ken Jennings and Brad Rutter in 2011, has evolved.  It’s part of IBM’s research and development in “Augmented Intelligence” or cognitive computer technology.  And the new Watson is helping people recognize early forms of skin cancer and save lives.

In 2014, IBM’s Watson Health Initiative began work with the American Cancer Society to help cancer patients and families. Later that year, Watson’s cognitive technology joined forces with the Sloan Kettering Research Centre. The computer was trained to recognize specific features and patterns indicating cancerous skin growths.

Now a recent study is using IBM technology and photos taken with Dermascopes that attach to smartphones for early skin cancer detection.  These lightweight cameras designed to get the best images for skin evaluation are already in use by doctors.  For less than $500, Dermalite offers a model that clips right onto your smartphone or tablet.  It boasts a 15-mm lens and high-powered LED lighting.

Doctors submitted photos of suspicious skin.  Each was evaluated by a computer trained to recognize melanoma with 900 images from the International Skin Imaging Collaboration.  Results show the chances of there being a melanoma.

Early testing showed the computer achieving an accuracy of 76%, while a team of 8 expert dermatologists only fared 70.5%.  Nearly 5 million people are treated every year in the United States for skin cancer.  Melanoma is responsible for 9.000 deaths every year.  With early detection a critical factor in survival rates, that 7% may have significant potential to save lives.

IBM is not alone in its quest to use AI in medicine. In 2014, Google acquired Deep Mind. Remember when IBM’s Deep Blue beat world champion Garry Kasparov in chess?  20 years later, Deep Mind’s AlphaGo has mastered Go.  The program defeated South Korean Go champion Lee Se-dol three times this past March.  And last summer, the company announced a partnership with Moorfields Eye Hospital in London.  They will teach computers to diagnose eye disease using a database of more than a million eye scans.

It’s the perfect marriage of science and humanity.  In the words of Guru Banavar, Chief Science Officer of Cognitive Computing and Vice President at IBM Research:

“It is not unreasonable to expect that within [the world’s] rapidly growing body of digital information lie the much-needed clues for professionals to solve the major societal challenges of our time from defeating cancer … to managing the complexity of the global economy. And we at IBM believe that AI, or ‘cognitive,’ systems are the tools that will help us accomplish these ambitious goals.”