In 2018, lung cancer accounted for approximately 25 percent of all cancer deaths – making it the leading cause of cancer-related deaths in both men and women in the U.S. Early detection is crucial to prevent the tumor from spreading and can decrease mortality by 14 to 20 percent among high-risk populations. Here’s where Google enters the picture. Thanks to a new study from a team of researchers at the company, the solution may actually point to using the capabilities of artificial intelligence.
When it comes to screening for lung cancer, most radiologists opt to use computed tomography (CT) scans, however, nodules can either result in false positives or escape detection. Errors like these can delay diagnosis until it becomes too late to treat the patient. Using a deep learning algorithm, Google’s artificial intelligence system was able to see through the cancer’s disguises. According to the study, the AI was able to outperform six radiologists in determining whether patients had lung cancer.
"Not only can we better diagnose someone with cancer, we can also say if someone doesn't have cancer, potentially saving them from an invasive, costly, and risky lung biopsy," said Dr. Mozziyar Etemadi, co-author of the study.
In the Google-funded study, CT images were taken from about 15,000 patients from a National Institutes of Health (NIH) study in 2002. After analyzing a single scan, AI was able to detect lung cancer five percent more often than experts. The technology was also 11 percent more likely to decrease the rate of false positives. The results highlight the possibility for artificial intelligence to increase the chances for early detection of lung cancer – creating an opportunity for life-saving treatment before the tumor reaches advanced stages.
Why This Matters –
As Google continues to conduct further trials and research, the AI system will be made available through their Cloud Healthcare API. The technology could make lung cancer screenings more reliable across the world. Considering other research that has already been done for using machine learning to identify breast cancer and skin cancer in patients, Google’s AI continues to showcase the possibilities of AI in healthcare.
But, for those who fear that this tech could potentially eliminate the need for radiologists, Google engineers have provided reassurance that the AI system is only designed to improve the ability to detect nodules and determine if they’re dangerous. If results of future studies continue to support these findings, we can expect to see more machine learning in the detection and treatment of life-threatening diseases.