Teaching computers to find cancers in the human body

Applying recent advances in deep learning to detect cancers in mammographic images.

Breast cancer is one of the most common forms of cancers in women with about two million new cases every year. In 2015, about half a million women died from the disease despite extensive screenings of the population and much research into better treatments. Deep learning and AI can potentially help doctors to analyse images from mammography to detect the cancer at an early state, which is one of the best ways to increase the survivability of the patient.

A first step towards this goal is to teach the computer to recognise different types of structures and landmarks in images, which then can be used to classification (per pixel) of tissues to find tumours. The use of AI can in this case help medical doctors by decreasing their workload via marking and sorting images that require further analysis by an expert.

Image is used under Creative Commons with credits to Nevit Dilmen.