AI in medicine needs to be carefully deployed to counter bias – and not entrench it

AI in medicine has the potential to solve challenging problems but must be deployed carefully to counter biases, not entrench them.

The problems: AI algorithms in healthcare are trained using large data sets that often reflect long-standing racial inequities in healthcare delivery.
* Clinicians provide different care to white patients and patients of color, while people of color are often underrepresented in training data sets.
* A 2019 study found that an algorithm used to predict healthcare needs for over 100 million people was biased against Black patients.

Efforts to combat bias: Hospitals and researchers are joining forces to build national coalitions and share best practices to combat bias.
* The Biden administration has recently released proposals to design guardrails for the emerging technology.
* The Office for Civil Rights at HHS has proposed updated regulations that prohibit discrimination in clinical algorithms.

Challenges and concerns: Experts have concerns about the clarity and public availability of guidelines, as too restrictive regulations may cause physicians to avoid AI tools altogether.
* Hospitals with fewer resources may struggle to stay on the right side of the law without clear guidance.
* Addressing underlying racial inequities in healthcare remains essential.

View original article on NPR

This summary was created by an AI system. The use of this summary is subject to our Terms of Service.

Contact us about this post

Posted

in

by

Tags:

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *