Chia-Hsuan Chang
Postdoctoral Associate in Biomedical Informatics and Data ScienceAbout
Research
Publications
2025
LITA: An Efficient LLM-Assisted Iterative Topic Augmentation Framework
Chang C, Tsai J, Tsai Y, Hwang S. LITA: An Efficient LLM-Assisted Iterative Topic Augmentation Framework. Lecture Notes In Computer Science 2025, 15870: 449-460. DOI: 10.1007/978-981-96-8170-9_35.Peer-Reviewed Original ResearchAugmentation frameworkClustering performance metricsText clusteringTopic qualityLanguage modelSeed wordsTopic modelsText corpusPerformance metricsBaseline modelIterative refinementAPIGuided approachLabor-intensiveFrameworkTraditional modelsThematic structureTextDatasetMetricsLDAIterationClustersCostModelEnhancing Patient-Physician Communication: Simulating African American Vernacular English in Medical Diagnostics with Large Language Models
Lee Y, Chang C, Yang C. Enhancing Patient-Physician Communication: Simulating African American Vernacular English in Medical Diagnostics with Large Language Models. Journal Of Healthcare Informatics Research 2025, 9: 119-153. PMID: 40309129, PMCID: PMC12037967, DOI: 10.1007/s41666-025-00194-9.Peer-Reviewed Original ResearchAfrican American Vernacular EnglishLanguage modelLinguistic differencesLinguistic styleLinguistic cuesPrompt typeImprove cultural sensitivityEnglishCommunication gapCultural sensitivityLanguageAmerican demographicsHealthcare communicationStereotypesEffective communicationCommunicationClinical dialogueCuesUnited States Medical Licensing ExaminationDialogueStyleCultureReduce health disparitiesInformation formDemographic information
2024
Beyond Self-consistency: Ensemble Reasoning Boosts Consistency and Accuracy of LLMs in Cancer Staging
Chang C, Lucas M, Lee Y, Yang C, Lu-Yao G. Beyond Self-consistency: Ensemble Reasoning Boosts Consistency and Accuracy of LLMs in Cancer Staging. Lecture Notes In Computer Science 2024, 14844: 224-228. DOI: 10.1007/978-3-031-66538-7_23.Peer-Reviewed Original ResearchExplainable AI for Fair Sepsis Mortality Predictive Model
Chang C, Wang X, Yang C. Explainable AI for Fair Sepsis Mortality Predictive Model. Lecture Notes In Computer Science 2024, 14845: 267-276. DOI: 10.1007/978-3-031-66535-6_29.Peer-Reviewed Original ResearchTransfer learning processFairness of model predictionsExplainable AIExplainability methodsAI applicationsPrediction modelArtificial intelligenceImportance algorithmTransform clinical decision-makingExplainabilityDiverse patient demographicsFairnessHealthcare stakeholdersPredictive performanceLearning processMitigate biasAlgorithmHealthcare deliveryDecision-makingIntelligenceClinical decision-makingHealthcare professionalsMortality prediction modelMethodHealthcareAn ExplainableFair Framework for Prediction of Substance Use Disorder Treatment Completion
Lucas M, Wang X, Chang C, Yang C, Braughton J, Ngo Q. An ExplainableFair Framework for Prediction of Substance Use Disorder Treatment Completion. 2024, 00: 157-166. DOI: 10.1109/ichi61247.2024.00028.Peer-Reviewed Original ResearchSensitive attributesMachine learning modelsIn-process approachLearning modelsResource allocationFairness frameworkFairnessEnhancement processSubstance use disorder treatment completionLikelihood of adoptionClinical decision-makingFrameworkHealthcare providersLevels of governmentModel changesExplainabilityDecision-makingTreatment completionTrustworthinessHealthcareConstructing Cross-Lingual Consumer Health Vocabulary with Word-Embedding from Comparable User Generated Content
Chang C, Wang L, Yang C. Constructing Cross-Lingual Consumer Health Vocabulary with Word-Embedding from Comparable User Generated Content. 2024, 00: 275-284. DOI: 10.1109/ichi61247.2024.00043.Peer-Reviewed Original ResearchWord vector spaceOnline health communitiesConsumer health vocabularyOAC CHVHealth vocabularySkip-gram algorithmSemantically similar wordsConsumer-generated contentVector spaceUser Generated ContentMonolingual spacesIsometry assumptionSkip-gramRecognition frameworkWord embeddingsLanguage modelCosine similarityLanguageSimilar wordsEnglishNon-EnglishHuman effortVocabularyMedical translationMedical expressionsAchieving Equity via Transfer Learning With Fairness Optimization
Wang X, Chang C, Yang C. Achieving Equity via Transfer Learning With Fairness Optimization. IEEE Access 2024, 12: 195229-195241. DOI: 10.1109/access.2024.3519465.Peer-Reviewed Original ResearchBias mitigation approachesFairness optimizationTransfer learningReal-world datasetsMachine learning algorithmsMachine learning modelsDecision-making systemMinimal performance degradationFairness enhancementFairness constraintsAccurate classifierLearning algorithmsAI systemsTraining processMitigation approachesLearning modelsTrade-OffsPerformance degradationPerformance impactSuperior fairnessPerformance optimizationFairnessPredictive performanceLearningMachine
2023
Examining Public Awareness of Ageist Terms on Twitter: Content Analysis
Schramm E, Yang C, Chang C, Mulhorn K, Yoshinaga S, Huh-Yoo J. Examining Public Awareness of Ageist Terms on Twitter: Content Analysis. JMIR Aging 2023, 6: e41448. PMID: 37698119, PMCID: PMC10507520, DOI: 10.2196/41448.Peer-Reviewed Original ResearchAgeist languageAppropriate language useContent analysisLanguage usePublic discourseDerogatory termSocial mediaTwitter usersAgeist discoursesTwitterTweetsLanguageDiscourseOlder adultsOlder peopleMessagesSociety of AmericaCenters for Disease Control and PreventionDisease Control and PreventionOlder adult populationGerontological Society of AmericaControl and PreventionPeopleOlder adults' vulnerabilityTermPrompting for Few-shot Adverse Drug Reaction Recognition from Online Reviews
Chang C, Chang F, Hwang S, Yang C. Prompting for Few-shot Adverse Drug Reaction Recognition from Online Reviews. 2023, 00: 168-175. DOI: 10.1109/ichi57859.2023.00032.Peer-Reviewed Original ResearchNamed Entity RecognitionUser-generated contentSocial networking sitesRecognition performanceSize of labeled dataFew-shot learningGeneral-purpose datasetsFew-shotEntity recognitionReview datasetLabeled dataLanguage modelDetection modelLearned knowledgeTransfer knowledgeADR detectionDatasetOnline reviewsNetworking sitesExperimental resultsRecognitionClinical notesLabor-intensivePerformanceLearningResampling for Mitigating Bias in Predictive Model for Substance Use Disorder Treatment Completion
Lucas M, Chang C, Yang C. Resampling for Mitigating Bias in Predictive Model for Substance Use Disorder Treatment Completion. 2023, 00: 709-711. DOI: 10.1109/ichi57859.2023.00128.Peer-Reviewed Original Research