{"id":"https://openalex.org/W7151885866","doi":"https://doi.org/10.48550/arxiv.2604.06028","title":"A Multi-Stage Validation Framework for Trustworthy Large-scale Clinical Information Extraction using Large Language Models","display_name":"A Multi-Stage Validation Framework for Trustworthy Large-scale Clinical Information Extraction using Large Language Models","publication_year":2026,"publication_date":"2026-04-07","ids":{"openalex":"https://openalex.org/W7151885866","doi":"https://doi.org/10.48550/arxiv.2604.06028"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.06028","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.06028","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.06028","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5063640628","display_name":"Maria Mahbub","orcid":"https://orcid.org/0000-0002-3422-9650"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Mahbub, Maria","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088832526","display_name":"Gregory M. Dams","orcid":"https://orcid.org/0000-0002-8546-014X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dams, Gregory M.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041832528","display_name":"Josh A. Arnold","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Arnold, Josh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5094208274","display_name":"Caitlin Rizy","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rizy, Caitlin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065143008","display_name":"Sudarshan Srinivasan","orcid":"https://orcid.org/0009-0002-8662-5820"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Srinivasan, Sudarshan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103908022","display_name":"Elliot M. Fielstein","orcid":"https://orcid.org/0000-0002-0244-3281"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fielstein, Elliot M.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133228666","display_name":"Minu A. Aghevli","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aghevli, Minu A.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074566036","display_name":"Kamonica Craig","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Craig, Kamonica L.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042596176","display_name":"Elizabeth M. Oliva","orcid":"https://orcid.org/0000-0001-8318-3049"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Oliva, Elizabeth M.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051255361","display_name":"Joseph Erdos","orcid":"https://orcid.org/0000-0003-4442-0683"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Erdos, Joseph","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058909594","display_name":"Jodie Trafton","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Trafton, Jodie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5021608374","display_name":"Ioana Danciu","orcid":"https://orcid.org/0000-0002-0164-1403"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Danciu, Ioana","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":12,"corresponding_author_ids":["https://openalex.org/A5063640628"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.2800999879837036,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.2800999879837036,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11710","display_name":"Biomedical Text Mining and Ontologies","score":0.25870001316070557,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T13702","display_name":"Machine Learning in Healthcare","score":0.2021999955177307,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/medical-diagnosis","display_name":"Medical diagnosis","score":0.5144000053405762},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.5080999732017517},{"id":"https://openalex.org/keywords/subject-matter-expert","display_name":"Subject-matter expert","score":0.44780001044273376},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.43320000171661377},{"id":"https://openalex.org/keywords/information-extraction","display_name":"Information extraction","score":0.42410001158714294},{"id":"https://openalex.org/keywords/comparability","display_name":"Comparability","score":0.41819998621940613},{"id":"https://openalex.org/keywords/population","display_name":"Population","score":0.3889000117778778},{"id":"https://openalex.org/keywords/trustworthiness","display_name":"Trustworthiness","score":0.3402999937534332}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7267000079154968},{"id":"https://openalex.org/C534262118","wikidata":"https://www.wikidata.org/wiki/Q177719","display_name":"Medical diagnosis","level":2,"score":0.5144000053405762},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.5080999732017517},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5048999786376953},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4860000014305115},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47290000319480896},{"id":"https://openalex.org/C105002631","wikidata":"https://www.wikidata.org/wiki/Q4833645","display_name":"Subject-matter expert","level":3,"score":0.44780001044273376},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.43320000171661377},{"id":"https://openalex.org/C195807954","wikidata":"https://www.wikidata.org/wiki/Q1662562","display_name":"Information extraction","level":2,"score":0.42410001158714294},{"id":"https://openalex.org/C197947376","wikidata":"https://www.wikidata.org/wiki/Q5155608","display_name":"Comparability","level":2,"score":0.41819998621940613},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4163999855518341},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4074000120162964},{"id":"https://openalex.org/C2908647359","wikidata":"https://www.wikidata.org/wiki/Q2625603","display_name":"Population","level":2,"score":0.3889000117778778},{"id":"https://openalex.org/C153701036","wikidata":"https://www.wikidata.org/wiki/Q659974","display_name":"Trustworthiness","level":2,"score":0.3402999937534332},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.33660000562667847},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.3325999975204468},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.30649998784065247},{"id":"https://openalex.org/C69505689","wikidata":"https://www.wikidata.org/wiki/Q455338","display_name":"Unified Medical Language System","level":2,"score":0.3057999908924103},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.29660001397132874},{"id":"https://openalex.org/C2778493491","wikidata":"https://www.wikidata.org/wiki/Q7449072","display_name":"Semantic matching","level":3,"score":0.29030001163482666},{"id":"https://openalex.org/C2777466982","wikidata":"https://www.wikidata.org/wiki/Q5227287","display_name":"Data extraction","level":3,"score":0.2847000062465668},{"id":"https://openalex.org/C2986862884","wikidata":"https://www.wikidata.org/wiki/Q7553","display_name":"Language translation","level":3,"score":0.27160000801086426},{"id":"https://openalex.org/C63527458","wikidata":"https://www.wikidata.org/wiki/Q5133829","display_name":"Clinical decision support system","level":3,"score":0.2540000081062317},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.25040000677108765},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.06028","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.06028","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.06028","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.06028","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7102295756340027,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"show":[4],"promise":[5],"for":[6,56],"extracting":[7],"clinically":[8],"meaningful":[9],"information":[10,59,206],"from":[11,123],"unstructured":[12],"health":[13],"records,":[14],"yet":[15],"their":[16],"translation":[17],"into":[18],"real-world":[19],"settings":[20],"is":[21,208],"constrained":[22],"by":[23],"the":[24,147,166],"lack":[25],"of":[26,113,134,173,203],"scalable":[27],"and":[28,91,99,129],"trustworthy":[29,201],"validation":[30,54],"approaches.":[31],"Conventional":[32],"evaluation":[33,81],"methods":[34],"rely":[35],"heavily":[36],"on":[37],"annotation-intensive":[38,211],"reference":[39],"standards":[40],"or":[41,141],"incomplete":[42],"structured":[43],"data,":[44],"limiting":[45],"feasibility":[46],"at":[47],"population":[48],"scale.":[49],"We":[50,107],"propose":[51],"a":[52],"multi-stage":[53],"framework":[55,69,110],"LLM-based":[57,204],"clinical":[58,125,205],"extraction":[60,112,207],"that":[61,137,199],"enables":[62],"rigorous":[63],"assessment":[64],"under":[65,175],"weak":[66],"supervision.":[67],"The":[68],"integrates":[70],"prompt":[71],"calibration,":[72],"rule-based":[73],"plausibility":[74],"filtering,":[75],"semantic":[76,130],"grounding":[77,131],"assessment,":[78],"targeted":[79],"confirmatory":[80],"using":[82],"an":[83,170],"independent":[84],"higher-capacity":[85],"judge":[86,148],"LLM,":[87],"selective":[88],"expert":[89,157],"review,":[90],"external":[92],"predictive":[93],"validity":[94],"analysis":[95],"to":[96,111],"quantify":[97],"uncertainty":[98],"characterize":[100],"error":[101],"modes":[102],"without":[103,210],"exhaustive":[104],"manual":[105],"annotation.":[106],"applied":[108],"this":[109],"substance":[114,121],"use":[115],"disorder":[116],"(SUD)":[117],"diagnoses":[118,181],"across":[119],"11":[120],"categories":[122],"919,783":[124],"notes.":[126],"Rule-based":[127],"filtering":[128],"removed":[132],"14.59%":[133],"LLM-positive":[135],"extractions":[136],"were":[138],"unsupported,":[139],"irrelevant,":[140],"structurally":[142],"implausible.":[143],"For":[144],"high-uncertainty":[145],"cases,":[146],"LLM's":[149],"assessments":[150],"showed":[151],"substantial":[152],"agreement":[153],"with":[154],"subject":[155],"matter":[156],"review":[158],"(Gwet's":[159],"AC1=0.80).":[160],"Using":[161],"judge-evaluated":[162],"outputs":[163],"as":[164],"references,":[165],"primary":[167],"LLM":[168],"achieved":[169],"F1":[171],"score":[172],"0.80":[174],"relaxed":[176],"matching":[177],"criteria.":[178],"LLM-extracted":[179],"SUD":[180,187],"also":[182],"predicted":[183],"subsequent":[184],"engagement":[185],"in":[186],"specialty":[188],"care":[189],"more":[190],"accurately":[191],"than":[192],"structured-data":[193],"baselines":[194],"(AUC=0.80).":[195],"These":[196],"findings":[197],"demonstrate":[198],"scalable,":[200],"deployment":[202],"feasible":[209],"evaluation.":[212]},"counts_by_year":[],"updated_date":"2026-05-05T08:41:31.759640","created_date":"2026-04-09T00:00:00"}