Google’s AI Can Spot Lung Cancer Months Before Doctors

Google claims it has created an AI that’s capable of detecting lung cancer a year before any human doctor.

During its developer conference in Mountain View, California, the Silicon Valley tech giant explained how its new artificial intelligence could potentially increase the survival chances of patients. And, frankly, it was quite impressive.

According to Google’s health researcher, Lily Peng, five out of six radiologists would miss subtle lung lesions on a CT scan. But, with the deep learning model, health professionals would not only identify the lesion, but they could reach a diagnosis one year ahead.

In a statement to the press, Ms. Peng said:

“That year could translate into an increased survival rate of 40 percent.”

Google’s Artificial Intelligence In Healthcare

For a while now, Alphabet’s life sciences company, Verily, has been working with Google’s Artificial Intelligence researchers.

In an attempt to improve the health of people with diabetes, Verily created a contact lens that doesn’t require a blood test to monitor blood glucose levels. However, the scientist couldn’t make the lens compatible with human tears. So, they canceled the project.

According to Ms. Peng, Google hopes to make new diagnostic tools accessible to as many people as possible. So, not only would the tech giant work with hospitals in the future, but Verily may soon reach an agreement with the National Health Service (NHS) in England.

Last year, the company conducted a pilot study in which it analyzed anonymized patient data and tried to predict their potential chronic conditions.

Aside from using AI as a diagnostic tool, Google is also working on a smart assistant for people with disabilities. The AI can learn to understand their facial expressions and use it to perform functions – such as turning the light off and communicate with others.

Speaking to the press, Project Manager, Julie Cattiau said:

“Our AI algorithms currently aim to accommodate individuals who speak English and have impairments typically associated with ALS, but we believe that our research can be applied to larger groups of people and to different speech impairments.”

Alongside improved speech recognition, Google is also training personalized AI algorithms to recognize gestures and sound and take actions.

For example, it can send text messages or generate a spoken command to Google Home.

“This may be particularly helpful to people who are severely disabled and cannot speak,” says Cattiau.