2023
Suicide Tendency Prediction from Psychiatric Notes Using Transformer Models
Li Z, Ameer I, Hu Y, Abdelhameed A, Tao C, Selek S, Xu H. Suicide Tendency Prediction from Psychiatric Notes Using Transformer Models. 2023, 00: 481-483. DOI: 10.1109/ichi57859.2023.00074.Peer-Reviewed Original ResearchWeighted F1 scoreF1 scoreMachine learning modelsElectronic health recordsLearning modelsState-of-the-art modelsState-of-the-artBinary classification taskHealth recordsBinary classification modelStandard diagnosis codesClassification taskMulticlass classificationHealth informaticsClassification modelMental health informaticsTransformation modelPrediction algorithmPsychiatric notesInitial psychiatric evaluationSuicidal tendenciesMachineRandom forest modelSuicidal ideationPerformance
2018
Psychiatric stressor recognition from clinical notes to reveal association with suicide
Zhang Y, Zhang O, Li R, Flores A, Selek S, Zhang X, Xu H. Psychiatric stressor recognition from clinical notes to reveal association with suicide. Health Informatics Journal 2018, 25: 1846-1862. PMID: 30328378, DOI: 10.1177/1460458218796598.Peer-Reviewed Original ResearchConceptsElectronic health recordsSuicidal behaviorHealth recordsSuicide ideation/attemptsTremendous economic burdenPsychiatric stressorsSuicide risk factorsRisk factorsEconomic burdenPsychiatric stressClinical notesLarge-scale studiesPsychiatric notesSuicideAssociationSignificant stressorsStressorsPrior studiesPercentPrevious studiesStudyAdapting Word Embeddings from Multiple Domains to Symptom Recognition from Psychiatric Notes.
Zhang Y, Li H, Wang J, Cohen T, Roberts K, Xu H. Adapting Word Embeddings from Multiple Domains to Symptom Recognition from Psychiatric Notes. AMIA Joint Summits On Translational Science Proceedings 2018, 2017: 281-289. PMID: 29888086, PMCID: PMC5961810.Peer-Reviewed Original ResearchWord embeddingsClinical textTarget domainSource domainNatural language processing techniquesLanguage processing techniquesMultiple word embeddingsBaseline methodsBiomedical literatureFirst workProcessing techniquesEmbeddingPsychiatric notesMultiple domainsExperimental resultsDifferent weightsSuch informationImportant topicRecognitionDifferent approachesWikipediaInformationPersonalizationDomainTextLeveraging existing corpora for de-identification of psychiatric notes using domain adaptation.
Lee H, Zhang Y, Roberts K, Xu H. Leveraging existing corpora for de-identification of psychiatric notes using domain adaptation. AMIA Annual Symposium Proceedings 2018, 2017: 1070-1079. PMID: 29854175, PMCID: PMC5977650.Peer-Reviewed Original Research
2017
Psychiatric symptom recognition without labeled data using distributional representations of phrases and on-line knowledge
Zhang Y, Zhang O, Wu Y, Lee H, Xu J, Xu H, Roberts K. Psychiatric symptom recognition without labeled data using distributional representations of phrases and on-line knowledge. Journal Of Biomedical Informatics 2017, 75: s129-s137. PMID: 28624644, PMCID: PMC5705397, DOI: 10.1016/j.jbi.2017.06.014.Peer-Reviewed Original ResearchConceptsPsychiatric symptomsCandidate symptomMental disordersPsychiatric notesList of symptomsMayo ClinicClinical dataSymptom recognitionPatient experiencePersonalized preventionSymptomsAmerican Psychiatric AssociationAbstractTextClinical conceptsPsychiatric AssociationPhenotypic classificationDisordersClinical textMIMIC-IIHealthcare knowledgeConclusionSubjective descriptionsClinicDiseaseDiagnosisA hybrid approach to automatic de-identification of psychiatric notes
Lee H, Wu Y, Zhang Y, Xu J, Xu H, Roberts K. A hybrid approach to automatic de-identification of psychiatric notes. Journal Of Biomedical Informatics 2017, 75: s19-s27. PMID: 28602904, PMCID: PMC5705430, DOI: 10.1016/j.jbi.2017.06.006.Peer-Reviewed Original ResearchConceptsPsychiatric notesCEGS N-GRIDNatural language processing systemsRule-based componentTask Track 1Language processing systemRule-based approachDe-identificationDomain adaptationRich featuresProcessing systemHybrid approachN gridTrack 1Clinical dataTest setSystem performanceMachineHealth informationHybrid systemSystemClinical applicationTaskInformationDataInterweaving Domain Knowledge and Unsupervised Learning for Psychiatric Stressor Extraction from Clinical Notes
Zhang O, Zhang Y, Xu J, Roberts K, Zhang X, Xu H. Interweaving Domain Knowledge and Unsupervised Learning for Psychiatric Stressor Extraction from Clinical Notes. Lecture Notes In Computer Science 2017, 10351: 396-406. DOI: 10.1007/978-3-319-60045-1_41.Peer-Reviewed Original ResearchNatural language processing systemsWord representation featuresPsychiatric stressorsLanguage processing systemDeep learningDomain knowledgeElectronic health recordsUnsupervised learningInexact matchingClinical notesF-measureRepresentation featuresProcessing systemHealth recordsPsychiatric notesImportant problemMultiple sourcesExperimental resultsLearningAlgorithmChallengesMatchingNarrative textStressor dataRecall