By Ashwini Sakharkar 23 Oct, 2024

Collected at: https://www.techexplorist.com/first-wearable-ai-powered-camera-detects-medication-errors/91456/

A team of researchers from the University of Washington School of Medicine/UW Medicine has developed the first wearable camera system that, with the help of artificial intelligence, detects potential errors in medication delivery.

In a recent test, the system demonstrated exceptional proficiency in identifying potential errors in medication delivery within busy clinical environments. The AI achieved an impressive 99.6% sensitivity and 98.8% specificity in detecting vial-swap errors.

Dr. Kelly Michaelsen, co-lead author and assistant professor of anesthesiology and pain medicine at the University of Washington School of Medicine, emphasized the potential of this AI-powered camera system as a crucial safeguard, particularly in high-pressure medical settings such as operating rooms, intensive-care units, and emergency-medicine environments.

“The thought of being able to help patients in real-time or to prevent a medication error before it happens is very powerful,” she said. “One can hope for a 100% performance, but even humans cannot achieve that. In a survey of more than 100 anesthesia providers, the majority desired the system to be more than 95% accurate, which is a goal we achieved.”

Drug administration errors pose a significant risk in anesthesia and intensive care, with an estimated 5% to 10% of all drugs given being associated with errors. Injectable medication-related adverse events affect approximately 1.2 million patients annually, incurring a cost of $5.1 billion.

Syringe and vial-swap errors commonly occur during intravenous injections, accounting for about 20% of mistakes. Safety measures, such as a barcode system, are in place to prevent such errors, but practitioners may overlook these steps during high-stress situations.

The researchers set out to develop a cutting-edge deep-learning model, in combination with a GoPro camera, capable of accurately identifying the contents of cylindrical vials and syringes. This technology would then provide a timely warning before the medication is administered to the patient.

The training process for the model was extensive, spanning several months. It involved gathering high-resolution 4K video footage of 418 drug draws performed by 13 anesthesiology providers in diverse operating room environments. The videos captured healthcare professionals handling vials and syringes containing specific medications. Subsequently, these video segments were meticulously annotated to train the model to recognize the contents and containers.

It’s important to note that the video system does not rely on reading the text on each vial. Instead, it utilizes advanced visual cues such as the size and shape of the vials and syringes, the color of the vial caps, and the size of the label print.

The computational model was specifically trained to focus on medications in the foreground while ignoring vials and syringes in the background. This means that AI can accurately detect the specific syringe being used by a healthcare provider, improving safety and efficiency in healthcare practices. The potential of AI and deep learning to enhance various healthcare processes is just beginning to be explored.

The study also involved researchers from Carnegie Mellon University, Makerere University in Uganda, and the Toyota Research Institute, which built and tested the system.

Journal reference:

  1. Justin Chan, Solomon Nsumba, Mitchell Wortsman, Achal Dave, Ludwig Schmidt, Shyamnath Gollakota & Kelly Michaelsen. Detecting clinical medication errors with AI enabled wearable cameras. npj Digital Medicine, 2024; DOI: 10.1038/s41746-024-01295-2

Leave a Reply

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

0 0 votes
Article Rating
Subscribe
Notify of
guest
0 Comments
Inline Feedbacks
View all comments