Read Original Story

Facial-recognition software for diagnosing genetic disorders: STAT leaves too many questions unanswered


2 Star



Facial-recognition software finds a new use: diagnosing genetic disorders

Our Review Summary

This is a story that has a lot of good reporting, but by focusing on one company in the facial-recognition field, it leaves too many questions unanswered. The possibilities with facial recognition software are profound, and yet we don’t know from this story whether they truly could be used effectively for identifying and treating disease or whether they will be relegated mainly to sideshow status.

It’s also difficult to understand the “market” for this startup if most patients in the U.S. have access to genetic testing, which is needed for an actual diagnosis.


Why This Matters

For developing countries or areas without access to genetic testing, facial recognition software could help providers narrow the list of disorders. Still, outside the most common genetic disorders (such as Down’s syndrome), it seems unlikely that this software will be useful in the absence of confirmatory testing.


Does the story adequately discuss the costs of the intervention?

Not Satisfactory

There are no prices in this story. Cost is critical, particularly as this is proposed as helpful in resource-poor settings.

Does the story adequately quantify the benefits of the treatment/test/product/procedure?

Not Satisfactory

The story quantifies benefits for a related bit of science but not for the core product being described in the piece. For that, it says, “Face2Gene declined to provide similar numbers for their technology. ‘Since Face2Gene is a search and reference informational tool, the terms sensitivity and specificity are difficult to apply to our output,’ Gelbman cautioned.”

If we don’t know anything about sensitivity and specificity, we’re not sure how this can play any role in helping with diagnosis, even in narrowing the list of possible/probable disorders. The story needed to explore this problem more.

Does the story adequately explain/quantify the harms of the intervention?

Not Satisfactory

There is no explanation of harms here, but with any screening test (or “reference informational tool” as Face2Gene calls itself), there is a risk of overdiagnosis or incorrect diagnosis, without confirmatory genetic testing. That could be devastating to parents of affected children for example. Even though the company says that the software should not be used as a diagnostic tool, the allure to do so “off label” is there.

Does the story seem to grasp the quality of the evidence?

Not Satisfactory

Because the story jumps from descriptions of what Face2Gene does to a project from the NIH and then back to the company, it gives the impression that this is strong science with a proven track record. Also, the first anecdote talks about one of the NIH scientists having a “superpower” that allows him to diagnose disorders by looking at faces. It’s a metaphor, of course, but it underscores the overall impression that this emerging technology is on very solid footing. Yet, other than one set of numbers from one study, we are given no evidence of that footing.

Does the story commit disease-mongering?


There is no disease mongering in this piece.

Does the story use independent sources and identify conflicts of interest?

Not Satisfactory

It’s not exactly clear where the NIH work and the company’s work begin and end. It talks about NIH-funded researchers gathering photos and uploading those to Face2Gene. That should be made more clear.

An outside expert to comment on whether this is useful would have been helpful.

Does the story compare the new approach with existing alternatives?


The story makes clear that there is physician experience, as well as genetic tests, as alternatives. Because the software-makers insist this new tool is not to be used for diagnoses, we’re still not sure what the benefit of this new technology would be, and why it’s needed, though.

Does the story establish the availability of the treatment/test/product/procedure?

Not Satisfactory

It’s not clear if this technology is available to anyone or if it is entirely in the research phase. This sounds developmental, but it was not clear. On visiting their website, it appears that it is already available for use.

Does the story establish the true novelty of the approach?


It’s clearly established that’s what “new” here is using facial recognition software to diagnose genetic disorders.

Does the story appear to rely solely or largely on a news release?


The story does not rely on a news release.

Total Score: 4 of 10 Satisfactory


Please note, comments are no longer published through this website. All previously made comments are still archived and available for viewing through select posts.