The Israeli startup Zebra Medical Vision, which uses artificial intelligence to read medical scans in order to automatically detect anomalies, has partnered with multinational tech giant Google to provide its sophisticated algorithms on Google Cloud, where hospitals and medical professionals can access the service for $1 per scan.
Zebra Medical Vision, founded in 2014, developed 11 algorithms, each of which is trained to identify the visual symptoms for a different disease, by using thousands of medical scans of diseases that had already been diagnosed and labeled. The algorithms, when fed a new medical scan, can recognize visual cues that indicate whether the disease is present.
Get our weekly highlights directly in your inbox!Sign up
Zebra is currently able to detect diseases such as breast cancer, vertebral fractures, osteoporosis, fatty liver, aneurysms, and brain bleeds. Its software is installed on-site in hospitals around the world, but the cost for storing the images can be prohibitive.
Zebra says it’s one of the reasons it made its service available on Google Cloud.
Eyal Gura, co-founder and chairman of Zebra Medical Vision, says the company recognized “two strong trends coming to place — one is the urgent need for more scalable and secure storage for hospitals due to the overwhelming amounts of digital data. The other urgent need is how to analyze this data quickly and accurately to bring value to doctors and patients.”
“The new collaboration we just announced with Google Cloud will enable hospitals and patients to realize value from both trends,” he told CNBC on Wednesday.
Zebra announced the service, which it calls AI1 or All-in-1, on its website late last month before the partnership with Google was reached. CEO Elad Benjamin wrote that the impetus behind the move was the company’s vision “of affordable, accessible imaging technology for everyone.”
“We truly believe the introduction of AI1 and this approach will encourage the faster global adoption of advanced tools that can improve health and save more lives,” he wrote.
In a September interview with NoCamels, Gura said Zebra Medical Vision’s algorithms were more accurate than expert radiologists. He said that the company’s breast cancer algorithm, which was developed and trained using 350,000 mammograms of diagnosed breast cancer, detects breast cancer with a 92 percent accuracy, in comparison to the 87 percent accuracy rate of radiologists.
The company indicated it was working with 50 hospitals around the world, which use the algorithms to prevent future diseases in patients, analyze sections of the population at risk of certain diseases, and assist radiologists by providing information about scans.
Zebra is also experimenting with a consumer product called Profound, which allows patients to upload a medical imaging scan and receive an automated analysis within an hour.
Earlier this year, Zebra announced a partnership with Telerad Tech, which will see its technology implemented in 150 hospitals and healthcare organizations in 20 developing countries. Gura said it was the first scaled implementation of artificial intelligence for medical imaging diagnosis in these countries, and is a step towards Zebra’s vision of saving thousands of lives in developing countries.
Gura told NoCamels that “machine and deep learning will help the finite amount of radiologists we have currently to be 10 times better, and analyze 10 times more reports per hour.” In addition, “machines will do all the radiology work, and the radiologists will just supervise them. They will be more like commercial flight pilots.”
The company’s goal is to continue expanding and improving healthcare capabilities in all areas that involve visuals, from dermatology to cytology and digital pathology.
Founded three years ago by Gura, Benjamin, and Eyal Toledano, the startup has since raised $20 million in two rounds led by Khosla Ventures and Intermountain Healthcare. Zebra’s 30 employees are based in Israel, the US, and the UK.
Earlier this year, Zebra received the 2017 Atlas Award given to one top Israeli startup by the Ayn Rand Center, and at the 2017 Innovation Summit, CB Insights named it among the top 50 most promising AI startups in the world.