NURS04 Ambient clinical intelligence for procedural sedation: consciousness level assessments

Overview

Procedural sedation involves the administration of sedative and/or analgesic medications that allow patients to tolerate painful or uncomfortable diagnostic or therapeutic procedures without general anesthesia. It is commonly used for radiology, cardiac catheterisation and endoscopy procedures. The number, complexity, and diversity of procedures performed with procedural sedation is growing.

Clinicians perform regular assessments of the adequacy of procedural sedation to inform their decision-making around sedation titration and also for documentation of the care provided. Additionally, regular assessment of consciousness is required during procedures performed with procedural sedation but is poorly documented.

Using artificial intelligence to augment clinical documentation systems, commonly termed 'ambient clinical intelligence', is considered one of the major near-term advances expected to improve the efficiency of healthcare. In this project we will use computer vision and natural language processing to devise a system that automates documentation of consciousness level assessment in real-time, allowing clinicians to maintain their focus on the provision of care rather than peripheral tasks.

Research engagement

This work will involve the student annotating videos of patients who are undergoing procedural  sedation with labels rating level of consciousness using the Observer's Assessment of Alertness Sedation scale.

Research activities

The student will be supervised by Aaron Conway and will primarily use annotation software to assign labels to segments of videos of procedures where a determination of consciousness can be made.  The student will also contribute to undertaking basic descriptive analysis of the annotations in preparation for subsequent stages of the project.

Outcomes

The outcome of this work will be a dataset of annotations of consciousness level assessment. This data will be used to train machine learning models to predict consciouness level assessments with the view towards creating an automated system that can be incorporated into practice to facilitate ambient clinical intelligence.

Skills and experience

An ideal candidate for this project is interested in machine learning or artificial intelligence in healthcare and nursing and may have had some exposure to procedural sedation, but that is not a requirement.

Start date

1 November, 2024

End date

21 February, 2025

Location

QUT Kelvin Grove or remote

Keywords

• nursing
• artificial intelligence
• machine learning
• sedation

Contact

aaron.conway@qut.edu.au