Related Experiment Video
Updated: Jun 12, 2026
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
Published on: July 11, 2025
Prompt-directed ambient artificial intelligence for automated multidisciplinary tumor board documentation
R Philips1, A Mahajan2, L Estephan3
1Department of Surgery, Section of Otolaryngology-Head & Neck Surgery, University of Chicago, Chicago, IL, USA; Department of Otolaryngology-Head & Neck Surgery, Thomas Jefferson University Hospital, Philadelphia, PA, US.
Prompt-directed ambient artificial intelligence (AI) significantly improved head and neck cancer tumor board documentation quality and reduced administrative burden compared to manual methods. This AI approach enhances medical summaries without increasing discussion times.
Area of Science:
- Oncology
- Medical Informatics
- Artificial Intelligence in Healthcare
Background:
- Multidisciplinary tumor board documentation is complex.
- Ambient artificial intelligence (AI) summarizes voice data using generative methods.
- Strategic prompting can enhance AI model output for medical dictation.
Purpose of the Study:
- To evaluate prompt-directed ambient AI for automated tumor board documentation.
- To compare ambient AI documentation with manual documentation processes.
Main Methods:
- A prospective quality improvement study was conducted in 2025.
- Verbal prompting was used during head and neck cancer tumor board discussions.
- Documentation quality was assessed using a modified MODe rubric across five domains.
Main Results:
- Prompt-directed ambient AI improved tumor characteristics and total scores versus unprompted AI.
- Ambient AI significantly improved total and discussion domain scores compared to manual documentation.
- Ambient AI reduced administrative time per patient by 0.001.
Conclusions:
- Directed prompting enhances ambient AI summaries for tumor boards.
- Ambient AI improves documentation quality and reduces administrative burden versus manual methods.
