Kenya: Using AI for Cancer Screening

Cary Chaplin, David Bell and Celina Shocken, in an article posted on IEEE Spectrum, present a promising new cervical cancer screening campaign in Africa, by Global Good in partnership with MobileODT and Pink Ribbon Red Ribbon, that relies on AI and visual images to detect cervical cancer. The AI system determines whether signs of cancer are present based on photos taken using the "EVA Scope" (Enhanced Visual Assessment Scope), which is a clip-on attachment for a smartphone. This turns the smartphone into a device similar to a colposcope, the tool gynecologists use to view a magnified image of a woman’s cervix. With an associated phone app, the screeners can analyze the image, show it to the patient, and store the data in the cloud. The authors detail the development of the analytic tool that is used in the system, describing in detail the underlying technology (Convolutional Neural Networks, or CNN) and how the AI system must be trained to recognize signs of cancer - a fascinating story. The system, which could potentially be adapted to perform other types of health screening in the field, is being launched in rural areas in Kenya, and additional trials will begin in Ethiopia during 2017. 

July 2017