Artificial Intelligence (AI) may be prepared for the elimination of cervical cancer, according to a study that showed that computer algorithms can better detect pre-history lesions than trained professionals or conventional screening tests.According to the World Health Organization, cervical cancer is the fourth most common cancer in women.In recent years, there has been great progress in the detection and vaccination to prevent the spread of human papillomavirus, which causes most cases of cancer of this type, but these discoveries have mostly benefited women in rich countries. .
In 2012, 266,000 women died of cervical cancer in the world, 90% of them in low- and middle-income countries."Cervical cancer is now a disease of poverty, low resources," says study leader Mark Schiffman, a doctor in the Epidemiology and Cancer Genetics Department at the National Cancer Institute near Washington, looking for a cancer drug. 35 years.
"We are trying to find ways that are extremely cheap, very simple, but very precise, that we can attack cervical cancer and also a little later with a simple technique based on mobile phones or something like that." He said.
Schiffman was part of a team that built an algorithm from a file that is over 60,000 cervix images collected in Costa Rica.The images were shot with a speculum, a low light and a camera, without the need for advanced images.The study started in the nineties and involved more than 9,400 women who were accompanied by up to 18 years of age.The AI technique, called an automated visual assessment, detected precancerous cells with 91% accuracy, according to a report published in the Journal of the National Cancer Institute.
For comparison, a review by a specialist doctor found 69% of previous cancers, while conventional laboratory tests, such as Papanicolaou, found 71%.In women aged 25 to 49 years with the greatest risk of cervical cancer, the AI algorithm was even more accurate since it found 97.7% of pre-cancer cells."He worked much better than people who saw these same images. He certainly did a lot better than me," Schiffman said.The goal is to introduce technology in the next three to five years, to enroll more patients into clinical trials around the world and ultimately make them easily accessible everywhere.
More news about MSN:
Everyday elements that can cause cancer. An internal star gallery