2023
WICOX: Weight-Based Integrated Cox Model for Time-to-Event Data in Distributed Databases Without Data-Sharing
Park J, Kim T, Kim J, Park Y. WICOX: Weight-Based Integrated Cox Model for Time-to-Event Data in Distributed Databases Without Data-Sharing. IEEE Journal Of Biomedical And Health Informatics 2023, 27: 526-537. PMID: 36318551, DOI: 10.1109/jbhi.2022.3218585.Peer-Reviewed Original ResearchConceptsLarge-scale biomedical dataPrivacy-preserving mannerPrivacy-protecting methodsCommon data modelPrivacy issuesDistributed DatabaseBiomedical dataData modelData networksIterative communicationCentralized modelPatient-level informationEvent dataSimulation resultsNon-iterative methodPredictive performanceRobust characteristicsRobust resultsMultiple institutionsDatabaseIntegrated modelAlgorithmNetworkCommunicationModel
2022
The evolving privacy and security concerns for genomic data analysis and sharing as observed from the iDASH competition
Kuo T, Jiang X, Tang H, Wang X, Harmanci A, Kim M, Post K, Bu D, Bath T, Kim J, Liu W, Chen H, Ohno-Machado L. The evolving privacy and security concerns for genomic data analysis and sharing as observed from the iDASH competition. Journal Of The American Medical Informatics Association 2022, 29: 2182-2190. PMID: 36164820, PMCID: PMC9667175, DOI: 10.1093/jamia/ocac165.Peer-Reviewed Original ResearchConceptsSensitive personal informationGenomic data analysisPotential future research directionsPersonal informationSecurity concernsGenomics data repositoryData repositoryReport lessonsProtection techniquesFuture research directionsPrivacyResearch directionsData usePractical challengesGenomic dataData analysisAnonymizationCommunity effortsRepositorySecurityBiomedical researchInformationDataChallenges
2021
VERTIcal Grid lOgistic regression with Confidence Intervals (VERTIGO-CI).
Kim J, Li W, Bath T, Jiang X, Ohno-Machado L. VERTIcal Grid lOgistic regression with Confidence Intervals (VERTIGO-CI). AMIA Joint Summits On Translational Science Proceedings 2021, 2021: 355-364. PMID: 34457150, PMCID: PMC8378611.Peer-Reviewed Original ResearchConceptsDual spaceCovariance matrixVariance estimationSpace modelTest statisticKernel matrixLinear modelReal dataTolerable performanceNovel extensionDual-space modelPoint estimatesEquivalent accuracyCentralized versionStatisticsRegression modelsModelMatrixExtensionDual objectivesCentralized settingFederated LearningEstimationPrivacy-preserving mannerSpace
2020
Privacy-preserving model learning on a blockchain network-of-networks
Kuo T, Kim J, Gabriel R. Privacy-preserving model learning on a blockchain network-of-networks. Journal Of The American Medical Informatics Association 2020, 27: 343-354. PMID: 31943009, PMCID: PMC7025358, DOI: 10.1093/jamia/ocz214.Peer-Reviewed Original ResearchConceptsNetwork topologyExecution timeArt methodsPredictive correctnessPrivacy-preserving learningPrivacy-preserving methodsPrivacy-preserving modelSmall training datasetBlockchain networkBlockchain platformBlockchain technologyPrivacy concernsModel learningComplex dataLearning iterationsLearning methodsTraining datasetConsensus algorithmGeneralizable predictive modelsCorrectness resultsModel disseminationHierarchical networkSmall dataHierarchical approachRecord model
2016
PRINCESS: Privacy-protecting Rare disease International Network Collaboration via Encryption through Software guard extensionS
Chen F, Wang S, Jiang X, Ding S, Lu Y, Kim J, Sahinalp S, Shimizu C, Burns J, Wright V, Png E, Hibberd M, Lloyd D, Yang H, Telenti A, Bloss C, Fox D, Lauter K, Ohno-Machado L. PRINCESS: Privacy-protecting Rare disease International Network Collaboration via Encryption through Software guard extensionS. Bioinformatics 2016, 33: 871-878. PMID: 28065902, PMCID: PMC5860394, DOI: 10.1093/bioinformatics/btw758.Peer-Reviewed Original ResearchConceptsSoftware Guard ExtensionsHomomorphic encryptionDistributed ComputationCollaboration frameworkTrustworthy computationCollaboration modelSupplementary dataNetwork collaborationEncryptionExperimental resultsHealth informationAlternative solutionComputationInternational collaboration frameworkHardwareAccurate analysisPerformanceBioinformaticsImplementationFrameworkDataExtensionInformationCollaboration
2011
Anomaly and signature filtering improve classifier performance for detection of suspicious access to EHRs.
Kim J, Grillo J, Boxwala A, Jiang X, Mandelbaum R, Patel B, Mikels D, Vinterbo S, Ohno-Machado L. Anomaly and signature filtering improve classifier performance for detection of suspicious access to EHRs. AMIA Annual Symposium Proceedings 2011, 2011: 723-31. PMID: 22195129, PMCID: PMC3243249.Peer-Reviewed Original ResearchConceptsSuspicious accessAccess recordsRule-based techniquesMachine learning methodsConstruction of classifiersAnomaly detectionInformative instancesLearning methodsSymbolic clusteringClassifier performanceSignature detectionIndependent test setInappropriate accessTest setEHRFiltering methodIntegrated filtering strategyFiltering strategyClassifierFilteringNegative rateFalse negative rateAccessDetectionClustering