Skip to content

AutoDISCERN: Using AI to assist patients in assessing health web page quality

Researchers Involved

MD, PhD Michael Krauthammer

research areas

AI
Health information
Patient

timeframe

2016 - 2021

Background

Patients increasingly turn to search engines and online content before, or in place of, talking with a health professional. However, the quality of online health information is not regulated. To address this, the NHS developed the DISCERN criteria to evaluate the quality of online health information. An AI version of the tool could run in the background and alert patients when they are viewing low quality information.

Project goals

Develop an AI system to evaluate the quality of online health information according to the DISCERN criteria.

Methodology

We built an automated implementation of the DISCERN instrument using Natural Language Processing & Machine Learning models.

Results

The AI models achieved accuracy scores of 81%. In comparison, human raters achieve a manual rating accuracy of 94%. Our research suggests that it is feasible to automate online health information quality assessment, which is an important step towards empowering patients to become informed partners in the healthcare process.

Publication

Kinkead L, Allam A, Krauthammer M. AutoDiscern: rating the quality of online health information with hierarchical encoder attention-based neural networks. BMC Med Inform Decis Mak. 2020 Jun 9;20(1):104. doi: 10.1186/s12911-020-01131-z