Abstract: A computer-implemented method and a distributed computer system (100) for privacy-preserving distributed training of a global neural network model on distributed datasets (DS1 to DSn). The system has a plurality of data providers (DP1 to DPn) being communicatively coupled. Each data provider has a respective local training dataset (DS1 to DSn) and a vector of output labels (OL1 to OLn) for training the global model. Further, it has a portion of a cryptographic distributed secret key (SK1 to SKn) and a corresponding collective cryptographic public key (CPK) of a multiparty fully homomorphic encryption scheme, with the weights of the global model being encrypted with the collective public key. Each data provider (DP1) computes and aggregates, for each layer of the global model, encrypted local gradients (LG1) using the respective local training dataset (DS1) and output labels (OL1), with forward pass and backpropagation using stochastic gradient descent.

Abstract: A computer-implemented method and a distributed computer system (100) for privacy-preserving distributed training of a global model on distributed datasets (DS1 to DSn). The system has a plurality of data providers (DP1 to DPn) being communicatively coupled. Each data provider has a respective local model (LM1 to LMn) and a respective local training dataset (DS1 to DSn) for training the local model using an iterative training algorithm (IA). Further it has a portion of a cryptographic distributed secret key (SK1 to SKn) and a corresponding collective cryptographic public key (CPK) of a multiparty fully homomorphic encryption scheme, with the local and global model being encrypted with the collective public key. Each data provider (DP1) trains its local model (LM1) using the respective local training dataset (DS1) by executing gradient descent updates of its local model (LM1), and combining (1340) the updated local model (LM1?) with the current global model (GM) into a current local model (LM1c).

Abstract: A genomic data decoder may jointly compress and encrypt genomic data alignment information while preserving the privacy of sensitive genomic data elements at retrieval stage. Genomic data alignment information organized as a read-based alignment data stream may be transposed into a position-based alignment data stream. The position-based alignment information may be encoded into a reference-based alignment data stream. The reference-based alignment data stream may be encrypted with a combination of order-preserving encryption of the genomic position information and symmetric encryption of the reference-based alignment differential data. Differential encoding and entropy coding schemes may further compress the reference-based alignment data stream. The resulting compressed and encrypted stream may be indexed and stored in a biobank storage unit.

Abstract: A genomic data decoder may jointly compress and encrypt genomic data alignment information while preserving the privacy of sensitive genomic data elements at retrieval stage. Genomic data alignment information organized as a read-based alignment data stream may be transposed into a position-based alignment data stream. The position-based alignment information may been coded into a reference-based alignment data stream. The reference-based alignment data stream may be encrypted with a combination of order-preserving encryption of the genomic position information and symmetric encryption of the reference-based alignment differential data. Differential encoding and entropy coding schemes may further compress the reference-based alignment data stream. The resulting compressed and encrypted stream may be indexed and stored in a biobank storage unit.

Abstract: A method to manage raw genomic data (SAM/BAM files) in a privacy preserving manner in a biobank. By using order preserving encryption of the reads' positions, the method provides a requested range of nucleotides to a medical unit, without revealing the locations of the short reads (which include the requested nucleotides) to the biobank. The method prevents the leakage of extra information in the short reads to the medical unit by masking the encrypted short reads at the biobank. That is, specific parts of the genomic data for which the medical unit is not authorized or the patient prefers to keep secret are masked at the biobank, without revealing any information to the biobank.

Abstract: A genomic data decoder may jointly compress and encrypt genomic data alignment information while preserving the privacy of sensitive genomic data elements at retrieval stage. Genomic data alignment information organized as a read-based alignment data stream may be transposed into a position-based alignment data stream. The position-based alignment information may been coded into a reference-based alignment data stream. The reference-based alignment data stream may be encrypted with a combination of order-preserving encryption of the genomic position information and symmetric encryption of the reference-based alignment differential data. Differential encoding and entropy coding schemes may further compress the reference-based alignment data stream. The resulting compressed and encrypted stream may be indexed and stored in a biobank storage unit.

Abstract: In a communication system having a number of base stations and user devices, a sending user device comprising a processor and a memory is configured to generate a packet or other communication for forwarding to a receiving user device via one or more intermediary user devices of the system. The forwarding path of the communication may involve one or more of the base stations. A payment token is associated with the communication, such that at least one of the intermediary user devices can generate a payment claim based on the payment token. The payment token is independent of the particular identities of the one or more intermediary user devices of the system.

Abstract: A method to manage raw genomic data (SAM/BAM files) in a privacy preserving manner in a biobank. By using order preserving encryption of the reads' positions, the method provides a requested range of nucleotides to a medical unit, without revealing the locations of the short reads (which include the requested nucleotides) to the biobank. The method prevents the leakage of extra information in the short reads to the medical unit by masking the encrypted short reads at the biobank. That is, specific parts of the genomic data for which the medical unit is not authorized or the patient prefers to keep secret are masked at the biobank, without revealing any information to the biobank.

Abstract: A method to manage raw genomic data (SAM/BAM files) in a privacy preserving manner in a biobank. By using order preserving encryption of the reads' positions, the method provides a requested range of nucleotides to a medical unit, without revealing the locations of the short reads (which include the requested nucleotides) to the biobank. The method prevents the leakage of extra information in the short reads to the medical unit by masking the encrypted short reads at the biobank. That is, specific parts of the genomic data for which the medical unit is not authorized or the patient prefers to keep secret are masked at the biobank, without revealing any information to the biobank.

Abstract: In this invention, we propose privacy-enhancing technologies for medical tests and personalized medicine methods, which utilize patients' genomic data. Assuming the whole genome sequencing is done by a certified institution, we propose to store patients' genomic data encrypted by a patient's public keys at a Storage and Processing Unit (SPU). A part of the corresponding private key is also stored on the SPU. At the time of the test by a Medical Unit (MU), the patient provides the second part of the private key to the MU. A test with its associated markers is determined by the MU and sent to the SPU. The test is carried out on the encrypted values thanks to homomorphic operation and returned back to the MU. The latter uses the second part of the private key to access the result.

Abstract: A privacy-preserving method for performing a disease susceptibility test on a patient, said method comprising: (I) performing homomorphic computations, (J) obtaining a test result which is partly decrypted with a first part (prk1 resp. prk2) of a private key, (L) decrypting said partly decrypted result with a second part (prk2 resp. prk1) of said private key, wherein said homomorphic computations are based on encrypted genomic markers of the patient, on encrypted clinical and/or environmental markers, and on encrypted ancestry markers of the patient. The invention is also related to a method for inferring ancestry in the encrypted domain.

Abstract: A method to manage raw genomic data (SAM/BAM files) in a privacy preserving manner in a biobank. By using order preserving encryption of the reads' positions, the method provides a requested range of nucleotides to a medical unit, without revealing the locations of the short reads (which include the requested nucleotides) to the biobank. The method prevents the leakage of extra information in the short reads to the medical unit by masking the encrypted short reads at the biobank. That is, specific parts of the genomic data for which the medical unit is not authorized or the patient prefers to keep secret are masked at the biobank, without revealing any information to the biobank.

Abstract: In this invention, we propose privacy-enhancing technologies for medical tests and personalized medicine methods, which utilize patients' genomic data. Assuming the whole genome sequencing is done by a certified institution, we propose to store patients' genomic data encrypted by a patient's public keys at a Storage and Processing Unit (SPU). A part of the corresponding private key is also stored on the SPU. At the time of the test by a Medical Unit (MU), the patient provides the second part of the private key to the MU. A test with its associated markers is determined by the MU and sent to the SPU. The test is carried out on the encrypted values thanks to homomorphic operation and returned back to the MU. The latter uses the second part of the private key to access the result.

Abstract: A privacy-preserving method for performing a disease susceptibility test on a patient, said method comprising: (I) performing homomorphic computations, (J) obtaining a test result which is partly decrypted with a first part (prk1 resp. prk2) of a private key, (L) decrypting said partly decrypted result with a second part (prk2 resp. prk1) of said private key, wherein said homomorphic computations are based on encrypted genomic markers of the patient, on encrypted clinical and/or environmental markers, and on encrypted ancestry markers of the patient. The invention is also related to a method for inferring ancestry in the encrypted domain.

Abstract: In this invention, we propose privacy-enhancing technologies for medical tests and personalized medicine methods, which utilize patients' genomic data. Assuming the whole genome sequencing is done by a certified institution, we propose to store patients' genomic data encrypted by a patient's public keys at a Storage and Processing Unit (SPU). A part of the corresponding private key is also stored on the SPU. At the time of the test by a Medical Unit (MU), the patient provides the second part of the private key to the MU. A test with its associated markers is determined by the MU and sent to the SPU. The test is carried out on the encrypted values thanks to homomorphic operation and returned back to the MU. The latter uses the second part of the private key to access the result.

Abstract: In this invention, we propose privacy-enhancing technologies for medical tests and personalized medicine methods, which utilize patients' genomic data. Assuming the whole genome sequencing is done by a certified institution, we propose to store patients' genomic data encrypted by a patient's public keys at a Storage and Processing Unit (SPU). A part of the corresponding private key is also stored on the SPU. At the time of the test by a Medical Unit (MU), the patient provides the second part of the private key to the MU. A test with its associated markers is determined by the MU and sent to the SPU. The test is carried out on the encrypted values thanks to homomorphic operation and returned back to the MU. The latter uses the second part of the private key to access the result.

Abstract: An approach is provided for preserving privacy for appointment scheduling. A scheduling platform receives a request to schedule an appointment among one or more users. The scheduling platform determines availability information for the one or more users from one or more respective devices, wherein the availability information is encrypted using homomorphic encryption. The scheduling platform then processes and/or facilitates a processing of the availability information using, at least in part, one or more homomorphic functions to determine one or more recommended time slots for the appointment.

Abstract: An approach is provided for protecting a user identity in communication based on privacy information. The privacy engine selects one or more parameters associated with a privacy metric. Next, the privacy engine determines the parameters in a communication environment, the communication environment including a user device and a plurality of other devices. Next, the privacy engine computes a privacy level based, at least in part, on the parameters and the privacy metric. Next, the privacy engine compares the computed privacy level against a predetermined privacy level. Then, the privacy engine triggers a communication to one or more of the other devices in the communication environment based, at least in part, on the comparison.

Abstract: The aim of the present invention is to provide a method to detect misbehavior use of the IEEE 802.11 standard without modifying the standard itself. This aim is reached according to a method for detecting misbehavior in a contention based communication network, this method comprising the steps of: recording at least some of invalid frames with their respective station identification issued by the stations accessing an Access Point, recording at least some of valid frames with their respective station identification issued by the stations accessing the Access Point, determining, for each station, a scrambled ratio based on the number of invalid frames and the number of valid frames, detecting a misbehavior station based on a station which has a substantially lower ratio than the other stations.

Abstract: In a communication system having a number of base stations and user devices, a sending user device comprising a processor and a memory is configured to generate a packet or other communication for forwarding to a receiving user device via one or more intermediary user devices of the system. The forwarding path of the communication may involve one or more of the base stations. A payment token is associated with the communication, such that at least one of the intermediary user devices can generate a payment claim based on the payment token. The payment token is independent of the particular identities of the one or more intermediary user devices of the system.