DAX Copilot: ambient AI scribe may help reduce surgical resident clinical documentation burden

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Clinical Documentation

Background:

• High clinical documentation burden impacts surgical residents, potentially leading to burnout and affecting patient care.
• Ambient AI tools offer a potential solution to alleviate documentation workload.

Purpose of the Study:

• To critically assess the accuracy, usefulness, and readiness of DAX-Copilot for surgery residents in an inpatient setting.

Main Methods:

• DAX-Copilot, a HIPAA-compliant AI scribe, was tested in 25 simulated inpatient scenarios.
• Simulated patient interactions used multiple voices in a representative setting.
• Three authors scored generated notes using the Modified Patient Documentation Quality Instrument (PDQI-9).

Main Results:

• DAX-Copilot accurately converted speech to notes and was easy to use.
• Generated notes scored an average of 46.91/50, excelling in accuracy, thoroughness, and comprehensibility.
• The tool required minor edits for surgical note style, scoring 4/5 for organization and usefulness.

Conclusions:

• Ambient AI, like DAX-Copilot, shows potential to revolutionize clinical documentation for surgical residents.
• Further research is needed for diverse patient encounters and optimizing institutional proficiency.