Google unveiled an artificial intelligence (AI) system that demonstrated an amazing ability to detect lung cancer in early testing and even outperformed radiologists.
The tech giant trained the algorithm on 42,000 patient CT scans taken during a National Institutes of Health clinical trial and managed to outperform six radiologists in determining whether patients had cancer, according to a study published in Nature Medicine.
The AI was able to detect 5 percent more cancers and cut false positives – that is, when cancer is suspected but a nodule is actually innocuous – by 11 percent from reviewing a single scan, the study says.
“These people have a technology that will improve the precision of screening tremendously,” Dr. Otis Brawley, a professor of oncology and epidemiology at Johns Hopkins University who was not involved with the study, told STAT.
Brawley told the medical news outlet he is generally a skeptic of lung cancer screening, but emphasized that Google’s performance in reducing false positives was a significant step forward.
“It’s going to prevent more bad things from happening to people who are being screened,” added Brawley.
Artificial intelligence has long been posited as a way to potentially improve screenings and allow doctors to pinpoint a range of malignant tumors with far more accuracy.
However, Google’s system will need to undergo more rigorous testing before it could be put into medical practice. Since the study was limited to patients who had already been treated, it is not known whether the system would result in better outcomes when used on new patients.
Lung cancer is the leading cause of cancer death among men and women. There are about 228,000 new cases of lung cancer and 142,000 deaths from it each year, according to the American Cancer Society.