5 data issues limiting AI's potential in healthcare

Technology is increasingly interloping with efforts to make healthcare more efficient. Still, experts are finding it difficult to apply artificial intelligence to medical treatment, according to a Feb. 24 article published in The Wall Street Journal.

The potential sale of IBM Watson Health has further highlighted this challenge. IBM has said its mission is to create massive data sets to assist physicians in treating difficult health issues like cancer.

Experts say it can be difficult to apply AI in medical care, especially for complex diseases like cancer, because of data issues.

Here are five such data issues that limit AI's potential in healthcare, which experts shared with WSJ:

  1. Having access to broad patient data may not be effective for specific patient care.

  2. Gaps in knowledge on complex diseases, incomplete medical histories and treatment outcomes may skew results.

  3. Patient data is recorded in many different formats, across different systems owned by healthcare providers and insurers, among other organizations.

  4. AI does not anticipate what patients want in their treatment. Patients can turn down certain treatments or vaccines, actions AI may not be able to compute.

  5. The stakes are high for AI technology to function perfectly. If the AI model is incorrect, it may put the patient's health at risk.


To read the full article, click here.

Copyright © 2024 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.

 

Featured Whitepapers

Featured Webinars