Zebra capitalizes on Israel’s strong computer vision industry and automates medical diagnoses computed by looking at x-rays, MRIs, and other scans
Shefayim-based Zebra Medical Vision roped in $12 million for its computer vision auto-diagnosing solution.
The new investment was led by Intermountain Healthcare with participation from existing funders Khosla Ventures and Marc Benioff. The round brings their total funding to $20 million following an $8 million round in April 2015.
Zebra capitalizes on Israel’s strong computer vision industry and automates medical diagnoses computed by looking at x-rays, MRIs, and other scans. The company already claims hundreds of thousands of cases have been scanned with algorithms specifically focused on liver health, lung health, cardiovascular analysis, and bone density. They currently claim over 1,100 customers.
Their press release described the market as favorable right now as it looks to integrate more machine learning and big data into diagnostics. Paradoxically, more efficient diagnostics are necessary to reach previously isolated areas of the world with a limited number of medical professionals in niche fields.
“In many places in the world, people don’t have access to doctors and in these places where there is no sufficient radiology diagnosis ability, the nurses and general physicians will be able to get value from the technology,” Zebra CEO and C0-Founder Elad Benjamin told Geektime. “In developed counties, much of the value will come from detecting things that radiologies cannot see with their own eyes or to help them prioritize cases and increase their throughput.”
We asked if this new melding of algorithms had been running long enough to detect common patterns among diagnoses that machines were picking up, specifically the sorts of positive tests that the computer vision software was churning out that weren’t visible to the naked eye.
“Yes. Our ability to research on massive data set with multiple modalities [has] already yielded [a] few algorithms that find things that otherwise would be missed. For example, we can now extract bone density score from a regular existing CT scan (previously this was done only by a bone density scanner),” Benjamin said.
They are targeting large hospitals and accountable care providers as customers, saying each marketed algorithm has already run several hundred thousand scans without providing a specific number. They hope to eclipse 1 million scans by the end of 2016. They are still working to refine their tech with dozens of new algorithms. Benjamin said he couldn’t provide an exact rate for speed of diagnosis, saying each algorithm and test worked differently.
“We work with our partners to find the best way to create the most impact and value on people’s health,” Benjamin told Geektime. “There are for example 150 million scans that are already stored on the Dell cloud and with our partnership with Dell we will be able to offer these 1,000 hospitals insights about what’s going on with their existing population. It’s called ‘population health management’ and it’s one of the key topics in accountable care.”
Based in Kibbutz Shefayim, the company was founded in 2014 by Eyal Gura, Eyal Toledano, and Elad Benjamin.