AI technology Beats Human at Detecting Cervical Cancer in NIH Study

A computer algorithm, developed by National Institutes of Health (NIH) researchers, was able to successfully identify changes in the cervix, pointing at cancer cells better than a human expert looking at the same images through a microscope.

Researchers have successfully developed an algorithm with the help of artificial intelligence (AI) that can accurately identify precancerous changes in the cervix. This technology is very beneficial for areas that are experiencing a lack of medical professionals.

Insight on the Study

The research team, led by investigators from NIH, developed a computer algorithm that can successfully analyze digital images of a female’s cervix, which usually requires medical attention. This artificial intelligence approach, called the automated visual evaluation, has the technology and potential to screen for cervical cancer, a major benefit for settings with low resources and services. This AI algorithm allows for automated evaluation by health care workers with minimal training.

This new method is especially valuable for communities that do not have enough resources to support expensive equipment. Such settings currently use a method known as visual inspection with acetic acid (VIA) in which dilute acetic acid is applied to the cervix to examine the cervix with the naked eye. This method is of low cost and is usually convenient.  However, it lacks the ability to be sure off high accuracy and therefore requires improvement. This new study paves a way to make changes to the previous method and improve existing procedures for cervical cancer screenings.

This method was developed using complex data sets that allow the algorithm to recognize patterns in medical images, such as identifying cervical cancer cells. Created collaboratively by the National Cancer Institute and Global Good, and Intellectual Ventures fund helping to make Humanitarian Impact, this approach uses images collected during routine cervical cancer screenings in order to identify changes in the precancerous cells.   

The algorithm was created by using more than 60,000 cervical images which were collected in the 1990s during a cervical cancer screening study in Costa Rica. This prospective study allowed the researchers to gain complete information since the photos were digital and allowed for a deep learning algorithm to identify cervical cancers.

The study was verified with the National Library of Medicine, a center of information innovation that functions as the world’s largest biomedical library and produces electronic information on a variety of topics. Verifying the findings showed that the algorithm did in fact outperform human experts.

The findings from the study raise further curiosity and speculation about computers and their ability to replace human information. However, such algorithms could be a huge step forward for public health if these technologies are placed in low income communities.

About the National Cancer Institute (NCI)

The National Cancer Institute (NCI) is the federal government’s agency for training and cancer research. They have a strong mission to lead and support cancer research across nations to help people improve their lifestyles and promote healthier lives. The NCI leads the National Cancer Program to reduce the prevalence of cancer and improve the lives of patients with cancer with the help of research projects and the development of new interventions.

Mark Schiffman, M.D., M.P.H. of the National Cancer Institute (NCI) Division of Cancer Epidemiology and Genetics said in a release, “Our findings show that a deep learning algorithm can use images collected during routine cervical cancer screening to identify precancerous changes that, if left untreated, may develop into cancer… In fact, the computer analysis of the images was better at identifying precancer than a human expert reviewer of Pap tests under the microscope (cytology)”.

NCI has made a significant impact; their investments have helped to reduce rates of new cancer cases and deaths due to cancer to decline in the last few decades in America. By using cancer detection techniques, diagnosis and patient care have allowed individuals to live longer and healthier lives.

Artificial Intelligence (AI) in the Healthcare Industry

Artificial intelligence in healthcare is one of the most rapidly rising industries to help transform healthcare and improve many lives. With many diseases and illnesses present in the world today, AI makes it easier to solve problems that would otherwise take years to solve. The implementation of AI in healthcare institutions helps to improve outpatient care and advance society in the tech world.

Even though this is a big success in the healthcare industry, critics point out that AI could make things worse for low-income areas of the country that have less access to digital devices. There are many companies and organizations that continue research to use technology in the healthcare field to help advance society to the next step.

This AI approach to identifying cervical precancer could revolutionize the way cancers are diagnosed and identified. The automated visual evaluation is easy to perform and health care workers are able to use cellular devices for cervical screenings and treatments during patient visits.