GermanyHealthcare and life sciences2024 – ongoingFreiburg University Medical Center
AI drafting of ophthalmology clinical letters
Freiburg used AI for clinical letters and reported 93.1% usability in a model evaluation
Problem
Clinical letters required accurate German medical terminology and a consistent structure. Preparing this documentation occupied clinicians' time.
Approach
Freiburg researchers trained language models on ophthalmology records and had medical staff assess generated letters. BLOOM-CLP-German produced the strongest results. The ophthalmology clinic also confirmed routine use of an AI letter-writing tool.
Results
Takeaway
The 93.1% figure measured documents judged usable after minor corrections in an evaluation. It did not measure routine production accuracy or time saved.
Categories
- Size
- Large enterprise
- Industry
- Healthcare and life sciences
- Business function
- Internal work and knowledge management
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
The hospital reported 93.1% usability after minor corrections for its best evaluated model. It separately confirmed routine AI-assisted letter drafting in ophthalmology.
Full details are in the original sources.
