Study Finds that AI Is Twice as Good at Detecting Lung Cancer as Doctors

Study Finds that AI Is Twice as Good at Detecting Lung Cancer as Doctors
SHARE

According to new research, artificial intelligence is more than twice as effective as radiologists at screening people for lung cancer.

Machine learning-based software greatly improved the detection of lung nodules on chest X-rays in a real-world environment.

The method is more sensitive and less prone to overlook situations that require swift medical attention. That being said, it may increase patients’ chances of surviving this aggressive cancer type.

One of the study’s authors, Dr. Jin Mo Goo, from Seoul National University Hospital, South Korea, states that “Detecting lung nodules is one of the crucial tasks in chest X rays. Many studies have suggested AI based computer aided detection software can improve radiologists’ performance, but it isn’t widely used.”

Due to the absence of early signs, lung cancer is much too frequently discovered after it has progressed too much, so this new promising diagnosis technique is great news!

10,476 individuals who had chest X-rays at a screening facility between June of 2020 and December of 2021 were included in this new study.

“As our trial was conducted with a rather pragmatic approach, almost all participants were included, which is a real clinical setting,’ Dr. Goo went on to say.

The research also accounted for sex, age, as well as their history of lung cancer.

Furthermore, 11 percent of them were also either current or former smokers.

Participants were split into two groups, AI and non-AI, at random and equally.

One of three radiologists used AI to evaluate the X-rays of the first group, but the radiologists who read the images for the second group did not use it.

AI identified lung nodules that needed to be treated in 0.59 percent of the instances, as opposed to 0.25 percent of the cases where radiologists just looked at the photos with no assistance from artificial intelligence.

In total, two percent of participants had nodules identified on their lungs.

Dr. Goo mentions that “Our study provided strong evidence AI could truly help in interpreting chest radiographies. This will contribute to diagnosing chest diseases, especially lung cancer, which is more effectively at an earlier stage.”

The findings, which were published in Radiology, also revealed that AI is consistent across populations—even for those with bad or recently operated on lungs.

Previous studies have demonstrated that AI can aid in the prediction of a variety of cancers, including breast cancer, as well as gastrointestinal disorders, Alzheimer’s, dementia and heart attacks.


SHARE

Katherine is just getting her start as a journalist. She attended a technical school while still in high school where she learned a variety of skills, from photography to nutrition. Her enthusiasm for both natural and human sciences is real so she particularly enjoys covering topics on medicine and the environment.

Post Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.