Generative Adversarial Networks (GANs) are one of the well-known models to generate synthetic data including images, especially for research communities that cannot use original sensitive datasets because they are not publicly accessible. One of the main challenges in this area is to preserve the privacy of individuals who participate in the training of the GAN models. To address this challenge, we introduce a Differentially Private Conditional GAN (DP-CGAN) training framework based on a new clipping and perturbation strategy, which improves the performance of the model while preserving privacy of the training dataset. DP-CGAN generates both synthetic data and corresponding labels and leverages the recently introduced Rényi differential privacy accountant to track the spent privacy budget. The experimental results show that DP-CGAN can generate visually and empirically promising results on the MNIST dataset with a single-digit epsilon parameter in differential privacy.

ABSTRACTGenome-wide association studies (GWAS) have been widely used to unravel connections between genetic variants and diseases. Larger sample sizes in GWAS can lead to discovering more associations and more accurate genetic predictors. However, sharing and combining distributed genomic data to increase the sample size is often challenging or even impossible due to privacy concerns and privacy protection laws such as the GDPR. While meta-analysis has been established as an effective approach to combine summary statistics of several GWAS, its accuracy can be attenuated in the presence of cross-study heterogeneity. Here, we present sPLINK (safe PLINK), a user-friendly tool, which performs federated GWAS on distributed datasets while preserving the privacy of data and the accuracy of the results. sPLINK neither exchanges raw data nor does it rely on summary statistics. Instead, it performs model training in a federated manner, communicating only model parameters between cohorts and a central server. We verify that the federated results from sPLINK are the same as those from aggregated analyses conducted with PLINK. We demonstrate that sPLINK is robust against heterogeneous data (phenotype and confounding factors) distributions across cohorts while existing meta-analysis tools considerably lose accuracy in such scenarios. We also show that sPLINK achieves practical runtime, in order of minutes or hours, and acceptable network bandwidth consumption for chi-square and linear/logistic regression tests. Federated analysis with sPLINK, thus, has the potential to replace meta-analysis as the gold standard for collaborative GWAS. The user-friendly, readily usable sPLINK tool is available at https://exbio.wzw.tum.de/splink.

Meta-analysis has been established as an effective approach to combining summary statistics of several genome-wide association studies (GWAS). However, the accuracy of meta-analysis can be attenuated in the presence of cross-study heterogeneity. We present sPLINK, a hybrid federated and user-friendly tool, which performs privacy-aware GWAS on distributed datasets while preserving the accuracy of the results. sPLINK is robust against heterogeneous distributions of data across cohorts while meta-analysis considerably loses accuracy in such scenarios. sPLINK achieves practical runtime and acceptable network usage for chi-square and linear/logistic regression tests. sPLINK is available at https://exbio.wzw.tum.de/splink.

Background Artificial intelligence (AI) has been successfully applied in numerous scientific domains. In biomedicine, AI has already shown tremendous potential, e.g., in the interpretation of next-generation sequencing data and in the design of clinical decision support systems. Objectives However, training an AI model on sensitive data raises concerns about the privacy of individual participants. For example, summary statistics of a genome-wide association study can be used to determine the presence or absence of an individual in a given dataset. This considerable privacy risk has led to restrictions in accessing genomic and other biomedical data, which is detrimental for collaborative research and impedes scientific progress. Hence, there has been a substantial effort to develop AI methods that can learn from sensitive data while protecting individuals' privacy. Method This paper provides a structured overview of recent advances in privacy-preserving AI techniques in biomedicine. It places the most important state-of-the-art approaches within a unified taxonomy and discusses their strengths, limitations, and open problems. Conclusion As the most promising direction, we suggest combining federated machine learning as a more scalable approach with other additional privacy-preserving techniques. This would allow to merge the advantages to provide privacy guarantees in a distributed way for biomedical applications. Nonetheless, more research is necessary as hybrid approaches pose new challenges such as additional network or computation overhead.

No abstract