You’re reading the web edition of STAT Health Tech, our guide to how tech is transforming the life sciences. Sign up to get this newsletter delivered in your inbox every Tuesday and Thursday.
What to watch: Google’s attempts to monetize AI in health
The search giant is taking a stab at monetizing its artificial intelligence for health care markets through a new deal with iCAD, a cancer therapy company that plans to license Google Health’s breast cancer detection and risk prediction models and incorporate them into its clinical practice. Using Google’s AI Model — and its cloud infrastructure, which I’ll share more about later — could boost adoption of iCAD’s broader cancer detection services, an executive said in a press release.
These types of partnerships are key to growing Google Health’s business, Greg Corrado, Google’s head of Health AI, said in another release.
As we reported in 2020, a Google Health study published in Nature suggested that the company’s breast cancer screening tool could detect cancer more accurately than clinicians. But it’s still not clear how widely health systems will use the technology, nor how easy it is to incorporate it into their clinical workflows. (If you have thoughts or anecdotes, give us a shout!)
Speaking of Google Cloud…
…My colleague Mario Aguilar was watching closely as Google pitched its cloud business to hospitals at the HLTH industry conference in Las Vegas. The tech giant is positioning its cloud business as a low-effort, almost turn-key service for health systems wary of moving their data into the cloud, Mario writes. It’s even starting to sell “accelerators”: off-the-shelf packages of data models and dashboards for common use cases in health.
And the company may have a leg up on other consumer-focused tech companies eager to break into health — especially Amazon, whose expansion has garnered concern from traditional health systems, Mario writes. Read more from Mario here.
NIH’s plan to crowdsource Covid-19 test results
The National Institutes of Health recently soft-launched an online hub where people can self-report the results of at-home Covid-19 tests, which often aren’t reported elsewhere and are something of a blindspot for researchers and public health departments. MakeMyTestCount.org asks people to report a positive or negative result and test type along with some basic information like sex and whether they’re experiencing symptoms. But how the agency plans to grapple with potential hitches, like fabricated test results, is still unclear.
Andrew Weitz, who co-leads the effort at NIH, told me that while the site is protected against distributed denial of service attacks as well as multiple repeated attempts to submit results, “maybe a broader question is how do we know that all the results that people are submitting are actually true, and the answer is we don’t,” he said. He said the effort is a bit of an experiment, and that even if it’s somewhat messy, it could still give researchers and public health departments more insight into at-home testing than they have now.
“We’re not trying to invade anyone’s privacy or track people,” he said. “I think it’s important for people to understand that before this all existed if they went to a laboratory and got a Covid test or went to their doctors’ office and got a Covid test that information is already being reported…we’re simply asking people who are willing to share that information themselves to do so voluntarily.”
Yet another telehealth merger rumor
Telehealth behemoth Amwell is reportedly in talks to acquire virtual mental health therapy company Talkspace, according to Israeli publication Calcalist, which reported that the acquisition would be worth about $200 million or $1.50 per share. Neither Talkspace nor Amwell would confirm the report. As Mario’s been writing, mental health tech companies are facing increasing headwinds as sales slow — and Talkspace in particular has been warning investors and analysts since the summer that potential customers are deferring their decision-making in the face of economic uncertainty.
More thoughts on FTC’s commercial data use inquiry
The comment period on the Federal Trade Commission’s inquiry into commercial surveillance and data security closed last week, and we’ve heard from a few more of you about worst-case scenarios if consumer health data use remains largely unregulated. The Legal Action Center, for instance, points out that the use of apps for conditions like substance use disorder, or location data indicating visits to harm reduction or addiction treatment centers, could betray a person’s health conditions and lead to discrimination and other harms.
And if you have several hours to kill, you might check out the Electronic Privacy Information Center’s 230-page comment, which notes that while a consumer “might reasonably assume that everything she searches for on the healthcare search site will be protected and limited in the same way as her interactions with her doctor would be,” some websites in fact only comply with stricter data requirements when they’re working with a health care provider.
“In reality, the healthcare search site discloses information about her device, her geolocation data, her contact information, her demographics, and her searches for healthcare providers to ‘ad networks’ and ‘analytics partners.’ She can then be targeted and profiled by these unknown third-party businesses, which is an out-of-context secondary use of her sensitive personal information.”
What we’re reading
- How air quality sensors could keep workers healthy, New York Times
- Telehealth companies are lobbying to extend pandemic-era rules, WSJ
- Who’s the next health care M&A target? Fierce Healthcare