| Doctors might soon be able to detect skin cancer simply by waving a handheld device over a suspicious-looking skin lesion thanks to the work of a UH graduate student.
Eric Stotzer, a computer science Ph.D. student, is developing software that could help primary care doctors more quickly and accurately diagnose melanomas, the fastest growing type of skin cancer in the U.S.
The pocket-size device would capture a digital image of the skin and analyze its size, texture, color and blood vessel characteristics. Using a vast database of skin lesion images, the software will be programmed to recognize cancerous features and provide a numerical probability that a lesion is malignant.
Stotzer, a part-time student, also earned his master’s degree from UH and works at Texas Instruments, where he is developing a set of general software tools that can be used in a variety of handheld medical imaging devices.
Stotzer is part of a UH research team led by George Zouridakis, associate professor of computer science. The group has filed a patent for the device and could have a working prototype ready in a year. |
