Deep-Transfer-Learning–Based Natural Language Processing of Serial Free-Text Computed Tomography Reports for Predicting Survival of Patients With Pancreatic Cancer
Kim S, Kim S, Kim E, Cecchini M, Park M, Choi J, Kim S, Hwang H, Kang C, Choi H, Shin S, Kang J, Lee C. Deep-Transfer-Learning–Based Natural Language Processing of Serial Free-Text Computed Tomography Reports for Predicting Survival of Patients With Pancreatic Cancer. JCO Clinical Cancer Informatics 2024, 8: e2400021. PMID: 39151114, DOI: 10.1200/cci.24.00021.Peer-Reviewed Original ResearchMeSH KeywordsAgedDeep LearningFemaleHumansMaleMiddle AgedNatural Language ProcessingPancreatic NeoplasmsPrognosisRetrospective StudiesROC CurveTomography, X-Ray ComputedConceptsArea under the receiver operating characteristic curveSurvival of patientsCT reportsPancreatic cancerNatural language processingC-indexPredicting survivalOverall survival of patientsTertiary hospitalPredicting 1-year survivalPredicting survival of patientsImproved C-indexSurvival informationPancreatic cancer survivalReceiver operating characteristic curveInternal test data setNLP modelsComputed tomography reportsLanguage processingKorean tertiary hospitalOverall survivalConsecutive patientsActual survivalConcordance indexPatients