Skip to content

Smart Eyewear to Automatically Detect Drug Delivery Events

The Use of a Wearable Data Input Device (Google Glass, Microsoft HoloLens or GoPro Like Camera Mounted to Eyewear) to Collect Drug Delivery Event Data

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
ClinicalTrials.gov
Registry ID
NCT05184660
Enrollment
20
Registered
2022-01-11
Start date
2021-01-04
Completion date
2026-12-01
Last updated
2024-05-08

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Smart Eyewear

Keywords

automatic detection, smart eyewear, drug delivery, drug administration, patient safety, anesthetic drug, anesthesia

Brief summary

This project is designed as a prospective observational study of medication preparation and delivery using a novel wearable data input device to automate detection of drug delivery events in the anesthesia workspace at University of Washington Medical Center.

Detailed description

Detailed Summary: The goal of this study is to devise a high-fidelity system that automatically collects data about anesthesia care, specifically drug preparation and delivery to the patient. We use a novel wearable data input device that collects computer processed images (video only, no audio) to determine details of the preparation and delivery of medications in the operating room. The wearable data input device is used by an anesthesia provider. The device collects computer processed images and in the anesthesia workspace, which will be used to develop machine learning algorithms that can automatically recognize drug administration events in order to automate medical record keeping in this environment. Procedure: The principle investigator or a study coordinator will review the electronic operating room schedule to screen for potential data collection. The operating room team will be notified of the study the day prior to and day of data collection. The consented anesthesia provider will wear a point of view device to record drug preparation and drug delivery events in the operating room. Video recordings will be edited to remove Protected Health Information (PHI). Edited footage will then be annotated by volunteers on a computer vision learning tool website. The computer processed images will be then used to develop machine learning algorithms that can automatically recognize drug administration events. Participants: The initial recruitment goal for the study is 10 nurse anesthetists recording 5 days each in the operating room. Nurse anesthetists are recruited as participants through word of mouth. If determined after data analysis that additional data needs to collected, additional anesthesia providers will be consented.

Interventions

None listed

Sponsors

University of Washington
Lead SponsorOTHER

Study design

Observational model
OTHER
Time perspective
PROSPECTIVE

Eligibility

Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes

Inclusion criteria

* Anesthesia providers (attending anesthesiologist or certified registered nurse anesthetist) who work in an operating room at University of Washington Montlake Medical Center

Exclusion criteria

* Medical students and anesthesiology residents

Design outcomes

Primary

MeasureTime frameDescription
Operating room footage8-10 hoursOperating room footage of drug preparation and drug delivery events recorded by anesthesia provider using a point-of-view device.

Secondary

MeasureTime frameDescription
Computer vision toolkit8-10 hoursA computer vision toolkit which can identify syringes and medications in the hand of a provider in near real time.

Countries

United States

Contacts

Primary ContactKelly Michaelsen, MD, PhD
kellyem@uw.edu7169085411
Backup ContactSharon T Nguyen, BS
sharon17@uw.edu2066188673

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026