STAT

Opinion: Machine learning for clinical decision-making: pay attention to what you don’t see

Don't take results from machine learning algorithms at face value. Ask what information isn't available. What subgroups haven't been prioritized? Who is on the research team?

Even as machine learning and artificial intelligence are drawing substantial attention in health care, overzealousness for these technologies has created an environment in which other critical aspects of the research are often overlooked.

There’s no question that the increasing availability of large data sources and off-the-shelf machine learning tools offer tremendous resources to researchers. Yet a lack of understanding about the limitations of both the data and the algorithms can lead to erroneous or unsupported conclusions.

Given that machine learning in the health domain can have a direct impact on people’s lives, broad claims emerging from this kind of research should not be embraced in the data and analyses.

You’re reading a preview, subscribe to read more.

More from STAT

STAT2 min read
STAT+: Pharmalittle: We’re Reading About Amylyx Pulling Its ALS Drug, GLP-1 Drugs For Parkinson’s, And More
Amylyx Pharmaceuticals will take its ALS drug Relyvrio off the market in the U.S. and Canada, ending a multi-year saga for patients with the rare neurodegenerative disease.
STAT2 min read
STAT+: Pharmalittle: We’re Reading About An OptumRx Contract, 340B Dispute Resolution, And More
Cardinal Health announced its pharmaceutical distribution contracts with UnitedHealth's OptumRx unit will not be renewed after they expire in June.
STAT1 min read
USDA Faulted For Disclosing Scant Information About Outbreaks Of H5N1 Avian Flu In Cattle
With 28 herds in eight states infected with H5N1 bird flu, scientists are calling on the U.S. to release more data to help them assess the risk.

Related