AI-powered Clinical Note-Taking
La Trobe University
The Challenge
Clinicians, especially doctors and nurses, spend a considerable amount of time on administrative tasks like typing patient notes, writing referral letters, updating electronic health records (EHRs). This clerical work detracts from direct patient care, contributes to high rates of professional burnout, and can compromise the accuracy and detail of documentation when notes are rushed. The traditional methods of note-taking often fail to capture the detail and nuance of a clinical conversation, and the time spent typing can create a barrier between the clinician and the patient.
The Partnership
This project involves a collaborative effort between researchers and healthcare professionals in a simulated clinical setting. The core of the project is the partnership with the users – the doctors and nurses who will test the software. The project also implicitly involves a partnership with the software developers – Lyrebird, Heidi, and I-scribe – as their products are the subject of the study.
The Solution
The project looks at the adoption of AI-powered virtual scribe technology to support clinical note-taking.
These tools use “ambient listening” to capture and transcribe clinical conversation in real-time that leverages AI and machine learning to distinguish between relevant clinical information, background chatter and automatically generates a structured clinical note, summary or referral letter. This eliminates manual typing, allowing clinicians to maintain eye contact and focus on the patient.
Three AI scribes being studied are particularly relevant as they are all Australian-based and/or compliant with Australian Privacy Principles, providing a localised solution to the documentation challenge.
Recognition & Impact
- By providing a clear, evidence-based roadmap, the project’s recommendations can inform policy and best practices for the selection and implementation of AI transcription technology.
- The project findings can help clinicians feel more confident in adopting these new technologies and can help reduce administrative burden, potentially mitigating burnout.
- More detailed and accurate notes, generated in real-time, can also lead to better communication improve quality of care and the patient experience
- The project’s recommendations on implementation and optimisation will enable organisations to make informed decisions that can lead to substantial time and cost savings without sacrificing quality or patient privacy.
Project Highlights
- Comparative Analysis: This project provides direct, side-by-side comparison of three specific AI tools that will offer practical, actionable data.
- Mixed-Methods Approach: The use of both quantitative and qualitative methods to provide insights into not just the technical performance, but also the human experience of using the software.
- Workflow Integration: A key focus of the study is to understand how these tools fit into or disrupt existing clinical workflows.
Contact Person & Details
- Prof. James Boyd, Chair of Digital Health and Innovation, School of Psychology and Public Health, La Trobe University james.boyd@latrobe.edu.au
- Jeff Jones, Director, Innovation Central Melbourne jeffrey.jones@latrobe.edu.au