United StatesHealthcare and life sciences2018 – ongoingUC San Diego Health
Pneumonia detection in clinical X-rays
UC San Diego Health's pneumonia imaging model affected clinical decisions 20% of the time
Problem
Pneumonia could be difficult to detect in X-rays. Researchers needed to put their detection model into clinicians' existing workflow.
Approach
Researchers developed a machine-learning model that marked suspected pneumonia with colored overlays. UC San Diego Health connected it to clinical imaging and patient records through a HIPAA-compliant AWS environment.
Results
Takeaway
Clinicians received the model's findings within patient records, where they could use them in diagnosis and treatment.
Categories
- Size
- Large enterprise
- Industry
- Healthcare and life sciences
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer visionPredictive analytics and ML
- Adoption
- Built in-house
- Goal
- Quality and risk
- Evidence
- Vendor case study
Sources
Source summaries
UC San Diego Health integrated pneumonia detection into clinical imaging workflows. AWS cited a published evaluation reporting an effect on clinical decision-making 20% of the time. The page carries no date; the Internet Archive first saved it on March 18, 2022.
Full details are in the original sources.