A simple way to screen for cervical cancer with a smartphone

2019

Challenge

Cervical cancer is a leading cause of cancer death among women worldwide, with over 250,000 deaths per year. It is caused by persistent infection of the cervix with carcinogenic types of the human papilloma virus (HPV). Although cervical cancer is largely preventable through vaccination and screening, the health-care infrastructure required for successful prevention programs is out of reach for low- and middle-income countries where 90 percent of cervical cancer deaths occur.

Advance

An IRP research team led by senior investigator Mark Schiffman, M.D., M.P.H., in collaboration with investigators from the non-profit organization Global Good, developed a computer algorithm that can analyze digital images of a woman’s cervix and accurately identify precancerous changes that require medical attention. The team then used the technology to analyze cervical images and detect abnormalities in the first large, long-term prospective study to use that machine learning-based approach, called automated visual evaluation. Prior studies that had applied machine learning to cervical cancer screening had also produced promising results, but they were too small and short-lived to recommend adoption of the technology.

Impact

This artificial intelligence approach can be performed with minimal training using a smartphone enabled with a camera, making it ideal for countries with sparse healthcare resources where cervical cancer is a leading cause of illness and death among women. Thus, it has the potential to revolutionize cervical cancer screening and dramatically reduce deaths from the disease.

Publications

Hu L, Bell D, Antani S, Xue Z, Yu K, Horning MP, Gachuhi N, Wilson B, Jaiswal MS, Befano B, Long LR, Herrero R, Einstein MH, Burk RD, Demarco M, Gage JC, Rodriguez AC, Wentzensen N, Schiffman M. (2019). An observational study of deep learning and automated evaluation of cervical images for cancer screening. J Natl Cancer Inst. Sep 1;111(9):923-932.

Related Research in Action