A new cognitive visual-computing technology presents an effective method for physicians to diagnose skin cancer and melanoma earlier. IBM is collaborating with Memorial Sloan Kettering (New York City) to develop and research the potential uses of this system, which can analyze dermatological images of skin lesions and help identify abnormalities. The technology learns by analyzing medical images and recognizing specific patterns. “Skin cancer is a major public health problem,” says Allan Halpern, M.D., chief of dermatology service at Memorial Sloan Kettering. “Treatment options exist, with the best outcomes attained through early detection. Accurately distinguishing the earliest cancers from concerning benign lesions can be very challenging, even for dermatologists, so having the aid of analytics that can recognize medical images and detect small variations over time could vastly improve patient prognoses.” Preliminary research has revealed excellent accuracy in analyzing dermoscopy images.
Additionally, the U.S. Food and Drug Administration (FDA) approved a new drug ahead of schedule to treat patients with later stages of melanoma when tumors have spread to other areas of the body—unresectable or metastatic melanoma. The Bristol-Myers Squibb drug Opdivo works by inhibiting the PD-1 protein on cells, which obstructs the immune system from attacking melanomic tumors. The approval comes three months earlier than expected following positive results from a clinical trial. Of the 120 participants, 32 percent had their tumors shrink, and the effect lasted for more than six months in one-third of that group. A separate study found the drug safe for use, with only rash, itching, cough, and respiratory infections as common side effects. Opdivo represents a significant improvement in safety and efficacy of treatment options available for the rarely curable types of melanomas.—Jennifer Nied