Artificial Intelligence Vision Software Identifies Forms Of Skin Cancer

Researchers from Stanford University have discovered an algorithm that can accurately identify skin cancer. Modifying an existing Google image recognition algorithm, the team has made and artificial intelligence system that can detect different cancers and benign lesions, drawing from a library of over 130,000 dermatology photos representing over 2,000 diseases. In its early test stage, performance was at a 91% accuracy rating compared to a hypothetically flawless system.

Diagnosing skin cancer begins with a visual examination. A dermatologist usually looks at the suspicious lesion with both the naked eye and with the aid of a dermatoscope, which is a handheld microscope that provides low-level magnification of the skin. The researchers at Stanford believe that with refined and rigorous resting, their AI system will be able to provide identification services at the same or even heightened accuracy level as a dermatologist.

Although this algorithm currently exists on a computer, the team would like to make it smart phone compatible in the near future, offering reliable skin diagnoses from the comfort of your home. Vision testing and scanning are becoming more predominant in the medical industry with the advance of technology. From life sciences to electronics, Eclipse Automation offers progressive vision capabilities to assist in managing and improving the accuracy, efficiency and quality of all your manufacturing solutions.

Source: Engadget