Abstract: Some embodiments are directed to a categorization system for categorizing a sensitive data field in a dataset, e.g., a disease classification according to the ICD classification. A client device is to obtain categories for one or more records of the dataset. The client device determines categorization data for the categorization. The categorization data comprises homomorphic encryptions of possible values of the sensitive data field and encodings of the categories associated to the respective possible values, thus keeping the categorization secret. A data provider device stores the dataset and determines homomorphic encryption indicating differences between the value of the sensitive data field for a record and respective possible values. A categorization device determines which of those encryptions indicates a match and provides a category encoding associated with a matching possible value to the client device. The client device associates the encoded category to the record.

Abstract: Some embodiments are directed to a method for anonymizing a genomic data set. The method comprises receiving (410) the genomic data set and obtaining (420) a phenotypic probability for at least one phenotype informative single nucleotide polymorphism (SNP) of the genomic data set and a proportion of a population which exhibits a corresponding phenotypic trait. A re-identification risk score is computed (430) based on the genomic data set from the obtained phenotypic probability and the obtained proportion of the population which exhibits the phenotypic trait. If the re-identification risk score does not meet a threshold risk criterion, the genomic data set is anonymized by selecting (450) a phenotype informative SNP and masking (460) the selected phenotype informative SNP, and the re-identification risk score is re-computed. If the re-identification risk score meets the threshold risk criterion, the anonymized genomic data set is output (470).

Abstract: A mechanism for determining the trustworthiness of training a first neural network, and thereby of the trained first neural network. Values of a set of weights of the first neural network are monitored during the training process. The monitored values are used to determine the trustworthiness of the training of the first neural network.

Abstract: A method performed by a device for identifying a network node within a network to which data will be replicated is disclosed. The method comprises encrypting a session key according to an attribute-based encryption scheme; broadcasting the encrypted session key within the network; receiving at least one message encrypted using the session key from at least one network node within the network; and selecting a network node from the at least one network node to which data will be replicated. A further method, a device and a non-transitory machine-readable medium are also disclosed.

Abstract: A system or method generates de-identified output from a data set of patient data comprising unstructured text (100) in natural language phrases. A blacklist (105) has word items that are not allowed. The unstructured text is processed to determine a word count (110) comprising a list of low-rate word items that have a number of occurrences (k) in the unstructured text below a threshold (120). Subsequently, the low-rate word items and the blacklist word items are masked (130) in the unstructured text to generate the de-identified output (140).

Abstract: Some embodiments are directed to a categorization system for 100 categorizing a sensitive data field in a dataset, e.g., a disease classification according to the ICD classification. A client device is to obtain categories for one or more records of the dataset. The client device determines categorization data for the categorization. The categorization data comprises homomorphic encryptions of possible values of the sensitive data field and encodings of the categories associated to the respective possible values, thus keeping the categorization secret. A data provider device stores the dataset and determines homomorphic encryption indicating differences between the value of the sensitive data field for a record and respective possible values. A categorization device determines which of those encryptions indicates a match and provides a category encoding associated with a matching possible value to the client device. The client device associates the encoded category to the record.

Abstract: A system and method for anonymization of a data set of patient data from multiple patients provides k-anonymity, a concatenation of indirect identifiers of a patient enabling identifying an outlying patient in the data set if there are less than k patients having a same concatenation of indirect identifiers. The patient data as provided (302) is longitudinal and has events related to a disease or a treatment of a disease, and time stamps related to the events. At least one first indirect identifier representing a property of the data distribution of the time stamps, and at least one second indirect identifier representing a number of events regarding a respective patient, are determined (303). For all patients in the data set, the respective concatenations comprising the first indirect identifier and the second indirect identifier, are determined (305). Then, the patient data of each outlying patient is removed from the data set (306).

Abstract: A method, a system and a computer program product for anonymization of genetic data from at least one individual wherein the genetic data are grouped into a subset of genetic data being directly related to a disease and one or more subsets of genetic data being distantly related to the disease based upon the genome pathways network, and wherein the subsets of genetic data being distantly related to the disease are anonymized.

Abstract: A method performed by a device for identifying a network node within a network to which data will be replicated is disclosed. The method comprises encrypting a session key according to an attribute-based encryption scheme; broadcasting the encrypted session key within the network; receiving at least one message encrypted using the session key from at least one network node within the network; and selecting a network node from the at least one network node to which data will be replicated. A further method, a device and a non-transitory machine-readable medium are also disclosed.