Smart Eyewear in the Operating Room
This project is funded by NIH and aims to improve patient safety and medical record fidelity. It is a collaboration between the Departments of Anesthesiology and the UW Mobile Systems Lab.
Check out our publication in npj Digital Medicine: https://www.nature.com/articles/s41746-024-01295-2
Our lab has developed a pioneering wearable camera system designed to automatically detect potential clinical medication errors before drug delivery. Drug-related errors are a significant cause of preventable patient harm in clinical settings. This system utilizes deep learning algorithms to detect and classify drug labels on syringes and vials during drug preparation in real-world operating rooms.We collected a unique, large-scale video dataset from head-mounted cameras (4K footage) across 13 anesthesiology providers, two hospitals, and 17 operating rooms over 55 days, The dataset includes footage of actual medication preparation events and is augmented with vial swap errors performed in a controlled environment. All videos are de-identified, with protected health information cropped or blurred.
The system was evaluated on 418 drug draw events and achieved remarkable accuracy in detecting vial swap errors:
Sensitivity: 99.6% (95% CI 98.8 to 100.0%)
Specificity: 98.8% (95% CI 97.1 to 100.0%)
These results highlight the potential for our wearable camera system to provide a crucial secondary check, offering a chance to intervene prior to a potential medical error. Future efforts will focus on building mechanisms to provide real-time feedback to clinicians, measuring syringe volume to calculate medication dose, and integrating with electronic medical record systems for automatic drug information recording.
New Developments in Neuromuscular Blockade
This project through GaviaLabs aims to improve understanding of neuromuscular blockade in patients undergoing surgery requiring complete immobilization.
In addition to testing all commercially available products, we have built a novel mechanomygraphy device so this 'gold standard' assessment technique can also be used in studies.