Google wants to leverage its artificial intelligence technology to detect cancer sooner.
The company's technology could catch early signs missed by trained oncologists, Google Product Manager Dr. Lily Peng said at the company'son Tuesday.
Lung cancer causes more deaths than any other cancer, according to the World Health Organization. It's also among the most common types of cancer, with more than 2 million cases.
"We know that when cases are diagnosed early, patients have a higher chance of survival," Peng said. "But unfortunately, over 80% of lung cancers are not caught early."
Early stage cancer can be hard to see on a CT scan, and patients with late stage cancer often show subtle signs on early scans.
Using lung cancer scans from the National Cancer Institute and Northwestern University, Google trained a neural network to detect malignancy at a level equal to or above the abilities of a trained radiologist, Peng said.
In one case, a patient with no symptoms and no history of cancer had a CT scan that was interpreted as normal. A year later, another scan picked up late stage cancer. Google used its AI system to review the initial scan, and the model was able to detect the early signs a year before the patient was diagnosed.
Peng said for patients like this, early detection could translate to an increased survival rate of 40%.
"Clearly, this is a promising but early result," Peng said. "We're very much looking forward to partnering with the medical community to use technology like this to help improve outcomes for patients."