A Machine Learning-Based Privacy-Preserving Model for COVID-19 Patient using Differential Privacy

Abstract

Machine learning models are often trained on the dataset, which contains sensitive information. Private Aggregation of Teacher Ensembles is a helpful framework that can be used along with Differential privacy to preserve users' privacy. The teacher models are trained on a disjoint dataset, and the aggregated knowledge of the teacher model is then used to train the student model. The proposed method used the teacher models technique to detect whether a person's chest CT-Scan is affected by COVID-19 or not while preserving the user's privacy and compared the proposed privacy-preserving model with a standard CNN model to compare their accuracy.

Ähnliche Arbeiten