Artificial intelligence is more accurate than doctors in the diagnosis of cervical cancer

A study has shown that artificial intelligence can be more accurate than traditional doctors and screening in the early diagnosis of cervical cancer.

This type of cancer is the fourth most prevalent in the world and, every year, 570,000 women, according to the World Health Organization.

In rich countries, advances in diagnostic and vaccination campaigns for cervical papillomavirus have helped reduce the number of deaths, but this is not the case in poor countries, where 90% of deaths worldwide were caused by the disease.

"Cervical cancer has become a poverty-related disease," said Dr. Mark Schiffman, a physician at the Washington National Cancer Institute and co-author of the study.

"We are trying to find simpler and more accurate ways to avoid infection through vaccines or simple techniques, for example using a smartphone."

The doctor works with a scientific team that has been able to design an information program based on 60,000 images of the cervix. Based on these images, the program can directly identify the cells near the cancer cells. It showed an efficiency of 91%.

In comparison, the doctor is accurate in his work at 69%, and the traditional exam at 57%.

"The program is even more accurate than the doctors who take pictures," said Mark Schiffman.

After completing the necessary development, the doctor hopes that this program will be widely used and accessible to all. It does not require any training, just take pictures and let it analyze the result.