Abstract: Respective local parameters parametrizing a base function are determined to provide at least one local trained function for each of multiple client systems by training the respective local trained function using machine learning with multiple training datasets on the respective client system, wherein at least some of the training datasets are specific to the respective client system. A respective local plaintext dataset including the local parameters of the respective local trained function is encrypted to generate a respective local encrypted dataset on the respective client system. The local encrypted datasets are transmitted to an aggregating system, and the global encrypted dataset is calculated from the local encrypted datasets using the calculation algorithm.

Abstract: A computer-implemented method for providing an encrypted global model trained by federated learning, comprises providing an initial global model to a first client; training the provided initial global model based on first training data, to generate a first client trained model; determining first client data based on the first client trained model; homomorphically encrypting the first client data based on a first encryption key, wherein the homomorphical encryption of the first client data is based on a matrix multiplication of the first client data and the first encryption key, to generate encrypted first client data; sending the encrypted first client data to a central aggregator; determining the encrypted global model based on the encrypted first client data, wherein determining the encrypted global model comprises aggregating the encrypted first client data with encrypted second client data; and providing the encrypted global model.

Abstract: A medical data management system is for managing medical data. In an embodiment, the system includes a medical data gateway, including a processor connectable to a plurality of input devices, the medical data gateway being connected to a local network. Further, the processor is configured to carry out: collecting medical data, associated to a non-pseudonymized patient identifier and a data source identifier, from an input device; pseudonymizing at least the non-pseudonymized patient identifier of the medical data; exporting the pseudonymized medical data to a remote storage, the remote storage being part of a remote network external to the local network; allowing access to the non-pseudonymized patient identifier to a local application, running in the local network, and/or refusing access to the non-pseudonymized patient identifier to a remote application, running outside of the local network. A corresponding medical data management method is for managing medical data.

Abstract: A gateway and a method are provided for securely storing (and/or securely retrieving) medical data the method for storing comprising at least steps of: obtaining, in a secure environment, medical data which include patient property data as well as patient identifier data wherein the patient identifier data indicate at least one patient to which the patient property data correspond; generating, in the secure environment de identified medical data by replacing the patient identifier data in the medical data with non-patient-identifying coded identifiers; generating, in the secure environment, a re-identifying database indicating correspondences between the non-patient-identifying coded identifiers and the patient identifier data; generating n encrypted re-identifying database by applying, in the secure environment, at least one symmetric and/or asymmetric encryption method to the re-identifying database; storing the encrypted re-identifying database and the de-identified medical data on a cloud storage outsi

Abstract: A computer-implemented method is for providing processed data. In an embodiment, the method includes receiving, by a first encryption entity, first plaintext data including a matrix of numbers; determining, by the first encryption entity, an encryption key including an integer matrix; homomorphically encrypting, by the first encryption entity, the first plaintext data based on a matrix multiplication of the first plaintext data and the encryption key, to generate first encrypted data; sending, by the first encryption entity, the first encrypted data to a processing entity; receiving, by a decryption entity, encrypted processed data from the processing entity, the encrypted processed data being based on the first encrypted data; decrypting, by the decryption entity, the encrypted processed data based on a matrix multiplication of the processed data and an inverse of the encryption key, to generate processed data; and providing, by the decryption entity, the processed data.

Abstract: A computer-implemented method is for providing processed data. In an embodiment, the method includes receiving, by a first encryption entity, first plaintext data including a matrix of numbers; determining, by the first encryption entity, an encryption key including an integer matrix; homomorphically encrypting, by the first encryption entity, the first plaintext data based on a matrix multiplication of the first plaintext data and the encryption key, to generate first encrypted data; sending, by the first encryption entity, the first encrypted data to a processing entity; receiving, by a decryption entity, encrypted processed data from the processing entity, the encrypted processed data being based on the first encrypted data; decrypting, by the decryption entity, the encrypted processed data based on a matrix multiplication of the processed data and an inverse of the encryption key, to generate processed data; and providing, by the decryption entity, the processed data.

Abstract: Various examples of embodiments of the invention generally relate to automating a clinical workflow, the clinical workflow including an analysis of one or more medical datasets and generation of a medical report based on the analysis. For example, machine-learning algorithms may be used for the analysis. The medical report may be generated based on one or more report templates.