2022
Improving Pharmacovigilance Signal Detection from Clinical Notes with Locality Sensitive Neural Concept Embeddings.
Mower J, Bernstam E, Xu H, Myneni S, Subramanian D, Cohen T. Improving Pharmacovigilance Signal Detection from Clinical Notes with Locality Sensitive Neural Concept Embeddings. AMIA Joint Summits On Translational Science Proceedings 2022, 2022: 349-358. PMID: 35854716, PMCID: PMC9285153.Peer-Reviewed Original ResearchNatural language processingClinical notesRetrieval tasksConcept embeddingsNeural embeddingsLeverage informationLanguage processingEmbedding methodPharmacovigilance signal detectionADR signalsInherent complexityEmbeddingSignal detectionSignal recoveryAdverse drug reactionsStatistical measuresInformationDetection
2017
A comparative study of different methods for automatic identification of clopidogrel-induced bleedings in electronic health records.
Lee H, Jiang M, Wu Y, Shaffer C, Cleator J, Friedman E, Lewis J, Roden D, Denny J, Xu H. A comparative study of different methods for automatic identification of clopidogrel-induced bleedings in electronic health records. AMIA Joint Summits On Translational Science Proceedings 2017, 2017: 185-192. PMID: 28815128, PMCID: PMC5543340.Peer-Reviewed Original ResearchElectronic health recordsAdverse drug reactionsMachine learning-based methodsLearning-based methodsHealth recordsRule-based methodReasonable recallAutomatic identificationValuable data sourceAutomatic methodTemporality informationCertain adverse drug reactionsData sourcesIdentification of patientsPharmacogenomic studiesManual chart reviewInformatics approachAdverse eventsChart reviewDrug reactionsHigh precisionFunction-based methodScoring methodDifferent typesBleedingIdentification of adverse drug-drug interactions through causal association rule discovery from spontaneous adverse event reports
Cai R, Liu M, Hu Y, Melton B, Matheny M, Xu H, Duan L, Waitman L. Identification of adverse drug-drug interactions through causal association rule discovery from spontaneous adverse event reports. Artificial Intelligence In Medicine 2017, 76: 7-15. PMID: 28363289, PMCID: PMC6438384, DOI: 10.1016/j.artmed.2017.01.004.Peer-Reviewed Original ResearchConceptsDrug-drug interactionsTraditional association rule mining methodsAssociation rule mining methodAssociation rule discoveryAssociation rule miningRule mining methodAdverse Event Reporting SystemAdverse drug-drug interactionsAdverse event reportsAdverse eventsData-driven discoveryHigher-order associationsRule miningRule discoveryDrug safety surveillanceMining methodsBayesian networkDrug combinationsChallenging taskCausal associationDrug Administration Adverse Event Reporting SystemDDI identificationAdverse drug reactionsCombination of drugsEvent Reporting System
2012
Data mining methodologies for pharmacovigilance
Liu M, Matheny M, Hu Y, Xu H. Data mining methodologies for pharmacovigilance. ACM SIGKDD Explorations Newsletter 2012, 14: 35-42. DOI: 10.1145/2408736.2408742.Peer-Reviewed Original ResearchAdverse drug reactionsElectronic medical recordsLong-term adverse drug reactionsTerm adverse drug reactionPrevention of ADRsAdverse drug eventsPatient-reported dataPotential adverse drug reactionsNational surveillance systemEmergency departmentDrug eventsDrug reactionsPreclinical dataMedical recordsADR monitoringClinical trialsMedication safetyPreclinical characteristicsSpontaneous reportsPostmarketing phaseOnline health forumsPostmarketing stageDrug developmentHealth forumsPre-marketing stagesComparative analysis of pharmacovigilance methods in the detection of adverse drug reactions using electronic medical records
Liu M, Hinz E, Matheny M, Denny J, Schildcrout J, Miller R, Xu H. Comparative analysis of pharmacovigilance methods in the detection of adverse drug reactions using electronic medical records. Journal Of The American Medical Informatics Association 2012, 20: 420-426. PMID: 23161894, PMCID: PMC3628053, DOI: 10.1136/amiajnl-2012-001119.Peer-Reviewed Original ResearchConceptsAdverse drug reactionsElectronic medical recordsProportional reporting ratioVanderbilt University Medical CenterSpontaneous reporting systemDrug-event pairsDrug reactionsMedical recordsMedication ordersAbnormal laboratory resultsDrug-exposed groupNew adverse drug reactionsUniversity Medical CenterSpecific drug administrationReference standardLaboratory resultsUnexposed groupGamma Poisson ShrinkerMedical CenterPatient harmDrug AdministrationPharmacovigilance measuresBayesian confidence propagation neural networkEarly detectionReporting ratioDetecting Adverse Drug Reactions using Inpatient Medication Orders and Laboratory Tests Data
Liu M, Matheny M, Wu Y, Hinz E, Denny J, Schildcrout J, Miller R, Xu H. Detecting Adverse Drug Reactions using Inpatient Medication Orders and Laboratory Tests Data. 2012, 1: 131-131. DOI: 10.1109/hisb.2012.56.Peer-Reviewed Original ResearchAdverse drug reactionsElectronic medical recordsSpontaneous reporting systemProportional reporting ratioDrug reactionsMedication ordersDrug ordersInpatient medication ordersAbnormal laboratory resultsTime of admissionDrug-exposed groupChi-square testLaboratory resultsGamma Poisson ShrinkerUnexposed groupMedical recordsOdds ratioMedication safetyPatient harmBayesian confidence propagation neural networkEarly detectionReporting ratioEMR dataDay zeroReporting systemLarge-scale prediction of adverse drug reactions using chemical, biological, and phenotypic properties of drugs
Liu M, Wu Y, Chen Y, Sun J, Zhao Z, Chen X, Matheny M, Xu H. Large-scale prediction of adverse drug reactions using chemical, biological, and phenotypic properties of drugs. Journal Of The American Medical Informatics Association 2012, 19: e28-e35. PMID: 22718037, PMCID: PMC3392844, DOI: 10.1136/amiajnl-2011-000699.Peer-Reviewed Original ResearchConceptsAdverse drug reactionsPost-marketing phaseDrug reactionsSevere adverse drug reactionsImportant adverse drug reactionsWithdrawal of rofecoxibPotential adverse drug reactionsPost-marketing surveillanceADR predictionPatient morbidityClinical trialsMajor causeLarge-scale studiesDrugsMolecular pathwaysDrug developmentPhenotypic featuresSignificant improvementPhenotypic characteristicsEarly stages