National Healthcare Quality and Disparities Report
Latest available findings on quality of and access to health care
Data
- Data Infographics
- Data Visualizations
- Data Tools
- Data Innovations
- All-Payer Claims Database
- Healthcare Cost and Utilization Project (HCUP)
- Medical Expenditure Panel Survey (MEPS)
- AHRQ Quality Indicator Tools for Data Analytics
- State Snapshots
- United States Health Information Knowledgebase (USHIK)
- Data Sources Available from AHRQ
Search All Research Studies
AHRQ Research Studies Date
AHRQ Research Studies
Sign up: AHRQ Research Studies Email updates
Research Studies is a compilation of published research articles funded by AHRQ or authored by AHRQ researchers.
Results
1 to 2 of 2 Research Studies DisplayedPeng L, Luo G, Walker A
Evaluation of federated learning variations for COVID-19 diagnosis using chest radiographs from 42 US and European hospitals.
The goals of this study were to compare a single-site, COVID-19 computer diagnosis system that used the Federated Averaging (FedAvg) algorithm with 3-client Federated learning (FL) models, and to evaluate the performance of the four FL variations. Researchers leveraged a FL healthcare collaborative that included data from five US and European healthcare systems encompassing 42 hospitals. They concluded that FedAvg could significantly improve generalization of the model in comparison with other personalization FL algorithms--FedProx, FedBN, and FedAMP--but at the cost of poor internal validity.
AHRQ-funded; HS026379.
Citation: Peng L, Luo G, Walker A .
Evaluation of federated learning variations for COVID-19 diagnosis using chest radiographs from 42 US and European hospitals.
J Am Med Inform Assoc 2022 Dec 13;30(1):54-63. doi: 10.1093/jamia/ocac188..
Keywords: COVID-19, Diagnostic Safety and Quality, Imaging, Hospitals
Sun J, Peng L, Li T
Performance of a chest radiograph AI diagnostic tool for COVID-19: a prospective observational study.
The purpose of this observational study was to evaluate the real-time performance of an interpretable artificial intelligence (AI) model to detect COVID-19 on chest radiographs. The researchers utilized 95,363 chest radiographs for model training, external validation, and real-time validation. There were 5,335 real-time predictions and a COVID-19 prevalence of 4.8%. The study found that participants positive for COVID-19 had higher COVID-19 diagnostic scores than participants negative for COVID-19. Real-time model performance remained the same during the 19 weeks of implementation. Model sensitivity was higher in men than in women, but model specificity was higher in women. Sensitivity was higher for Asian and Black participants compared with White participants. The COVID-19 AI diagnostic system had worse accuracy compared with radiologist predictions. The researchers concluded that AI tools underperform when compared with radiologist results.
AHRQ-funded; HS026379.
Citation: Sun J, Peng L, Li T .
Performance of a chest radiograph AI diagnostic tool for COVID-19: a prospective observational study.
Radiol Artif Intell 2022 Jul;4(4):e210217. doi: 10.1148/ryai.210217..
Keywords: COVID-19, Imaging, Diagnostic Safety and Quality, Health Information Technology (HIT)