Cough Classification of Unknown Emerging Respiratory Disease with Federated Learning
Abstract
Artificial intelligence offers great potential to address the need for rapid diagnostic testing in pandemic scenarios. Concerns about security and privacy, however, complicate the collection of large representative medical data. Federated Learning (FL), a machine learning paradigm, enables distributed training by exchanging gradient information between the server and edge devices without data access. To this end, we propose an FL-based approach to solve a multiclass classification problem in a hypothetical pandemic scenario, where we learn to distinguish the cough of an unknown emerging disease from existing ones. Our proposed federated cough classifier algorithm achieves 45% Matthews correlation coefficient (MCC) on COVID-19 and 69% overall MCC classification performance, when nine COVID-19 patients accumulate a total of 2535 cough samples and edge devices send their gradient information to the server model 77 times. Our experiments show that our proposed approach is able to learn to classify the cough of an unknown disease in a privacy-preserving manner.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.