AI can use wireless signals to reduce errors in self-administered medication

Listen

Errors in self-administered medication can be mitigated by a new artificial intelligence tool that can monitor wireless signals in a patient's home, according to a March 18 study published in Nature Medicine.

Researchers from the Massachusetts Institute of Technology in Cambridge developed a contactless and unobtrusive AI tool that can automatically detect medication self-administration errors.

MSA errors are common in those with chronic diseases, where up to 50 percent of patients don't take their medication as prescribed. These errors increase when tools like inhalers or insulin tools are required to administer the medicine, the study said.

Six things to know about the AI tool:

  1. The AI tool is contactless and passive, and there is no additional burden to the patient or healthcare provider.

  2. To use the tool, a sensor is mounted to the wall in a patient's home and transmits a low-powered wireless signal. The tool is then able to analyze the reflections of the signals using AI techniques.

  3. Since the human body is 60 percent water, the tool reflects the surrounding radio signals and adjusts them with the person's movements.

  4. The AI tool examines the wireless reflections to determine if the patient followed the appropriate steps of using the medication devices and generates an alert if the patient failed to do it properly.

  5. There were 107 individuals examined with the AI tool. For MSA events that use insulin pens, patients missed a common step in 150 events and failed to adhere to the duration requirement in 155.

  6. For MSA events with inhaler devices, 149 events missed a required step and 168 failed to adhere to duration requirements.

For images on how the AI tool works or to read the full study, click here.

More articles on artificial intelligence: 
Sharecare develops machine learning model to predict allergy flare-ups
FDA authorizes 1st machine learning COVID-19 screening tool

Machine learning tool can predict severe illness, death from COVID-19 in patients

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

 

Featured Whitepapers

Featured Webinars