Dopamine: Differentially Private Federated Learning on Medical Data
This addresses privacy concerns for medical data sharing across hospitals, enabling more effective training of diagnostic models, though it is incremental as it builds on existing federated learning and differential privacy techniques.
The paper tackles the problem of training deep neural networks on distributed medical data while preserving patient privacy, by proposing Dopamine, a system that combines federated learning with differentially-private stochastic gradient descent and secure aggregation to achieve a better trade-off between privacy guarantees and accuracy. Results on a diabetic retinopathy task show that Dopamine provides a differential privacy guarantee close to centralized training and achieves better classification accuracy than federated learning with parallel differential privacy.
While rich medical datasets are hosted in hospitals distributed across the world, concerns on patients' privacy is a barrier against using such data to train deep neural networks (DNNs) for medical diagnostics. We propose Dopamine, a system to train DNNs on distributed datasets, which employs federated learning (FL) with differentially-private stochastic gradient descent (DPSGD), and, in combination with secure aggregation, can establish a better trade-off between differential privacy (DP) guarantee and DNN's accuracy than other approaches. Results on a diabetic retinopathy~(DR) task show that Dopamine provides a DP guarantee close to the centralized training counterpart, while achieving a better classification accuracy than FL with parallel DP where DPSGD is applied without coordination. Code is available at https://github.com/ipc-lab/private-ml-for-health.