Researchers: AI in healthcare presents ethics concerns

Listen
Text
  • Small
  • Medium
  • Large

Stanford (Calif.) University researchers are concerned about the ethical implications of using machine learning tools to make healthcare decisions for patients, according to a perspective piece published March 15 in the New England Journal of Medicine and reported by Medical Xpress.

Artificial intelligence is promising in healthcare. For example, it can be used to gain insights from health data to improve diagnostics and ultimately inform better decisions for patient care. However, physicians and scientists must be wary of the ethical risks posed by incorporating these tools into their decision-making processes.

"Because of the many potential benefits, there's a strong desire in society to have these tools piloted and implemented into healthcare," lead author, Danton Char, MD, assistant professor of anesthesiology, perioperative and pain medicine at Stanford, told Medical Xpress. "But we have begun to notice, from implementations in non-health care areas, that there can be ethical problems with algorithmic learning when it's deployed at a large scale."

Here are four concerns the authors raised.

1. Algorithms created with potentially biased data could result in algorithms tainted with the same biases, which would compromise the clinical recommendations they make. Algorithms can also be designed to skew results.

2. Physicians must understand how algorithms are created, be able to assess the data used to create the statistical models that predict outcomes, know how the models function and be careful not to overly depend on them.

3. Certain data gathered — such as patient health information, diagnostics and outcomes — contribute to the "collective knowledge" of information healthcare systems' gather, without considering individual clinical experiences or the human aspect of care delivery.

4. Machine learning-based clinical guidance could be considered a third-party "actor" in the physician-patient relationship, which could complicate patients' expectations.

Click here to download the paper.

More articles on artificial intelligence:
Google AI scans eye images to predict cardiovascular problems
KLAS: 8 popular IT vendors ranked by AI imaging capabilities
49% of organizations can't deploy AI projects because of data challenges, survey finds

Copyright © 2021 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.

 

Featured Whitepapers

Featured Webinars