United StatesFinance and insurance2019 – ongoingCDPHP
Medical-record extraction and automated HEDIS reporting
CDPHP produced 2 HEDIS reports daily after machine learning replaced manual medical-data preparation
Problem
CDPHP manually extracted and organized medical records. Preparing a HEDIS health-plan performance report took 4 to 5 days.
Approach
CDPHP's architecture team built a pipeline with support from AWS Professional Services. Amazon Textract extracted document text, Comprehend Medical identified medical information, and SageMaker supported downstream machine learning. The automated system fed recurring HEDIS reporting.
Results
Takeaway
Medical-text extraction supported recurring reporting that had occupied data scientists almost exclusively.
Categories
- Industry
- Finance and insurance
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
CDPHP automated medical-information extraction and produced 2 HEDIS reports daily, replacing a manual process requiring 4 to 5 days per report. The page carries no date; the Internet Archive first saved it on April 12, 2022.
Full details are in the original sources.