Related Experiment Video
Updated: Dec 19, 2025
07:51
Hydra, a Computer-Based Platform for Aiding Clinicians in Cardiovascular Analysis and Diagnosis
Published on: September 26, 2018
7.9K
A collaborative online AI engine for CT-based COVID-19 diagnosis.
Yongchao Xu1,2, Liya Ma1, Fan Yang3
1Department of Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China.
Medrxiv : the Preprint Server for Health Sciences
|June 9, 2020
Summary
Federated learning improved artificial intelligence (AI) models for diagnosing COVID-19 from chest CT scans. The Unified CT-COVID AI Diagnostic Initiative (UCADI) enhanced model generalization across institutions without sharing patient data.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Infectious Disease Diagnostics
Background:
- Chest computed tomography (CT) plays a role in diagnosing COVID-19.
- Developing robust AI for CT diagnosis is hindered by data deficiency, isolation, and heterogeneity.
- Lack of AI model generalization prevents clinical application.
Conclusions:
- Federated learning, via the UCADI framework, effectively addresses AI model generalization issues in COVID-19 CT diagnosis.
- The UCADI initiative facilitates the development of a robust, globally validated AI tool for COVID-19 detection.
- This approach supports global health initiatives by enabling knowledge sharing and the fight against COVID-19.
