Optimizing large language models in digestive disease: strategies and challenges to improve clinical outcomes
Giuffrè M, Kresevic S, Pugliese N, You K, Shung D. Optimizing large language models in digestive disease: strategies and challenges to improve clinical outcomes. Liver International 2024, 44: 2114-2124. PMID: 38819632, DOI: 10.1111/liv.15974.Peer-Reviewed Original ResearchSupervised fine-tuningHuman feedbackLanguage modelAccurate information retrievalInformation retrievalReinforcement learningDomain knowledgeText corpusNeural networkModel trainingAdaptive methodField of digestive diseasesFine-tuningWindow limitRetrievalIncreased accuracyAccuracyDigestive diseasesImprove healthcareDefinition of accuracyLearningIncrease patients' qualityClinical decision-makingSpecialized knowledgeDiverse sourcesOptimization of hepatological clinical guidelines interpretation by large language models: a retrieval augmented generation-based framework
Kresevic S, Giuffrè M, Ajcevic M, Accardo A, Crocè L, Shung D. Optimization of hepatological clinical guidelines interpretation by large language models: a retrieval augmented generation-based framework. Npj Digital Medicine 2024, 7: 102. PMID: 38654102, PMCID: PMC11039454, DOI: 10.1038/s41746-024-01091-y.Peer-Reviewed Original ResearchClinical decision support systemsFew-shot learningLanguage modelImprove clinical decision support systemsDecision support systemText similarity scoresText similarityAblation studiesImprove overall accuracyGeneration accuracyRight informationAccurate outputSupport systemTransform healthcareIncreasing levels of complexityHospital workflowOverall accuracyRight providerTurbo modelLearning strategiesLevel of complexityLearningRetrievalCorpus of textsAccuracy