{"id":"https://openalex.org/W7204918210","doi":"https://doi.org/10.48550/arxiv.2608.30390","title":"Uncertainty of Vision Medical Foundation Models","display_name":"Uncertainty of Vision Medical Foundation Models","publication_year":2026,"publication_date":"2026-08-31","ids":{"openalex":"https://openalex.org/W7204918210","doi":"https://doi.org/10.48550/arxiv.2608.30390"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.30390","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.30390","pdf_url":"https://arxiv.org/pdf/2608.30390","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.30390","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022918672","display_name":"Haoxu Huang","orcid":"https://orcid.org/0009-0006-6962-2520","observed_orcids":["https://orcid.org/0009-0006-6962-2520"]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Huang, Haoxu","raw_affiliation_strings":["Center for Data Science , New York University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Data Science , New York University","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5079997593","display_name":"Narges Razavian","orcid":"https://orcid.org/0000-0002-9922-6370","observed_orcids":["https://orcid.org/0000-0002-9922-6370"]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Razavian, Narges","raw_affiliation_strings":["Department of Population Health , NYU Grossman School of Medicine","Department of Radiology , NYU Grossman School of Medicine"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Population Health , NYU Grossman School of Medicine","institution_ids":["https://openalex.org/I57206974"]},{"raw_affiliation_string":"Department of Radiology , NYU Grossman School of Medicine","institution_ids":["https://openalex.org/I57206974"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I57206974"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.4794999957084656,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.4794999957084656,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.09440000355243683,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.08990000188350677,"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/uncertainty-quantification","display_name":"uncertainty quantification","score":0.9620000123977661},{"id":"https://openalex.org/keywords/conformal-prediction","display_name":"conformal prediction","score":0.9300000071525574},{"id":"https://openalex.org/keywords/self-supervised-learning","display_name":"self-supervised learning","score":0.8730000257492065},{"id":"https://openalex.org/keywords/calibration","display_name":"calibration","score":0.6830000281333923},{"id":"https://openalex.org/keywords/fundus-images","display_name":"fundus images","score":0.781000018119812},{"id":"https://openalex.org/keywords/histopathology","display_name":"histopathology","score":0.8450000286102295},{"id":"https://openalex.org/keywords/chest-x-ray","display_name":"chest radiography","score":0.9670000076293945}],"concepts":[{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.6409000158309937},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6182000041007996},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6139000058174133},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6123999953269958},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5414000153541565},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.5185999870300293},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5166000127792358},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.49970000982284546},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.484499990940094},{"id":"https://openalex.org/C32230216","wikidata":"https://www.wikidata.org/wiki/Q7882499","display_name":"Uncertainty quantification","level":2,"score":0.4440999925136566},{"id":"https://openalex.org/C146849305","wikidata":"https://www.wikidata.org/wiki/Q370766","display_name":"Ground truth","level":2,"score":0.40209999680519104},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3953000009059906},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.350600004196167},{"id":"https://openalex.org/C72161134","wikidata":"https://www.wikidata.org/wiki/Q5421219","display_name":"Expert elicitation","level":2,"score":0.3472999930381775},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.33480000495910645},{"id":"https://openalex.org/C137209882","wikidata":"https://www.wikidata.org/wiki/Q1403517","display_name":"Measurement uncertainty","level":2,"score":0.334199994802475},{"id":"https://openalex.org/C41426520","wikidata":"https://www.wikidata.org/wiki/Q1192065","display_name":"Point estimation","level":2,"score":0.3327000141143799},{"id":"https://openalex.org/C177803969","wikidata":"https://www.wikidata.org/wiki/Q29205","display_name":"Uncertainty analysis","level":2,"score":0.2978000044822693},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.2955000102519989},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.27219998836517334},{"id":"https://openalex.org/C107908354","wikidata":"https://www.wikidata.org/wiki/Q4781456","display_name":"Applicability domain","level":3,"score":0.26179999113082886},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.2549000084400177}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.30390","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.30390","pdf_url":"https://arxiv.org/pdf/2608.30390","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.30390","doi":"https://doi.org/10.48550/arxiv.2608.30390","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.30390","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.30390","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.30390","pdf_url":"https://arxiv.org/pdf/2608.30390","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G5693414920","display_name":"NRT-HDR: FUTURE Foundations, Translation, and Responsibility for Data Science Impact","funder_award_id":"1922658","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5900765980","display_name":null,"funder_award_id":"P30AG066512","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320332161","display_name":"National Institutes of Health","ror":"https://ror.org/01cwqze88"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204918210.pdf","grobid_xml":"https://content.openalex.org/works/W7204918210.grobid-xml"},"referenced_works_count":17,"referenced_works":["https://openalex.org/W1553101044","https://openalex.org/W2108598243","https://openalex.org/W2957498340","https://openalex.org/W2977650145","https://openalex.org/W3013277995","https://openalex.org/W3135057764","https://openalex.org/W3135367836","https://openalex.org/W4288809219","https://openalex.org/W4295916804","https://openalex.org/W4361012985","https://openalex.org/W4385948838","https://openalex.org/W4392947521","https://openalex.org/W4396758996","https://openalex.org/W4400889241","https://openalex.org/W4402220065","https://openalex.org/W4405669265","https://openalex.org/W4406353025"],"related_works":["https://openalex.org/W4405253844","https://openalex.org/W7163073477","https://openalex.org/W4387399257","https://openalex.org/W7154099048","https://openalex.org/W4403069550","https://openalex.org/W2991523350","https://openalex.org/W6979223302","https://openalex.org/W7214811496","https://openalex.org/W4399040553","https://openalex.org/W4403184078"],"abstract_inverted_index":{"Accurate":[0],"uncertainty":[1,100,167,227],"estimation":[2],"is":[3,71],"essential":[4],"for":[5,222],"machine":[6],"learning":[7,148],"systems":[8],"de-":[9],"ployed":[10],"in":[11,229],"high-stakes":[12],"domains":[13],"such":[14],"as":[15],"medicine.":[16],"Traditional":[17],"approaches":[18],"primarily":[19],"rely":[20],"on":[21,34,95,124,141,172],"probability":[22],"outputs":[23],"from":[24],"trained":[25,123,171],"models":[26,108,122,170],"(point":[27],"predictions),":[28],"which":[29],"provide":[30],"no":[31],"formal":[32],"guarantees":[33],"prediction":[35,50,59,205],"coverage":[36],"and":[37,93,98,127,196,203,210,239],"often":[38],"require":[39],"additional":[40],"calibra-":[41],"tion":[42],"techniques":[43],"to":[44,150,182,206,226],"improve":[45],"reliability.":[46],"In":[47,81],"contrast,":[48],"conformal":[49,185],"(region":[51],"prediction)":[52],"offers":[53],"a":[54,77,117,223],"principled":[55],"alternative":[56],"by":[57,102],"generating":[58],"sets":[60],"with":[61,146],"finite-":[62],"sample":[63],"validity":[64],"guarantees,":[65],"ensuring":[66,237],"that":[67,138],"the":[68,74,86,190,197,208,220],"ground":[69],"truth":[70],"contained":[72],"within":[73],"set":[75],"at":[76],"specified":[78],"confidence":[79],"level.":[80],"this":[82],"study,":[83],"we":[84],"explore":[85],"impact":[87],"of":[88,192,200,213,232],"pre-training":[89,140],"approach,":[90],"dataset":[91],"scale":[92],"domain":[94,111,156],"both":[96,201],"point":[97,152,202],"region-level":[99],"quantification,":[101],"studying":[103],"domain-specific":[104,143,177],"vision":[105,112,234],"medical":[106,214,233],"foundation":[107,113,121,178,235],"vs.":[109],"general":[110,155],"models.":[114],"We":[115],"conduct":[116],"comprehensive":[118],"evaluation":[119],"across":[120,169],"retinal,":[125],"histopathological,":[126],"Chest":[128],"X-Rays":[129],"data,":[130],"applying":[131],"various":[132],"calibration":[133],"techniques.":[134],"Our":[135,217],"results":[136],"demonstrate":[137],"(1)":[139],"higher-quality":[142],"datasets":[144],"along":[145],"self-supervised":[147],"leads":[149],"better-calibrated":[151],"predictions":[153],"than":[154],"pre-training,":[157],"(2)":[158],"stan-":[159],"dard":[160],"re-calibration":[161],"methods":[162],"alone":[163],"cannot":[164],"fully":[165],"mitigate":[166],"discrepancies":[168],"different":[173],"data":[174],"sources,":[175],"(3)":[176],"model":[179,194],"can":[180],"lead":[181],"more":[183],"efficient":[184],"prediction.":[186],"These":[187],"findings":[188],"highlight":[189],"importance":[191],"careful":[193],"selection":[195],"inte-":[198],"gration":[199],"region":[204],"enhance":[207],"reliability":[209],"trust-":[211],"worthiness":[212],"AI":[215],"systems.":[216],"work":[218],"underscores":[219],"need":[221],"holistic":[224],"approach":[225],"quantification":[228],"recent":[230],"development":[231],"model,":[236],"robust":[238],"interpretable":[240],"AI-driven":[241],"decision-making.":[242]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-02T00:00:00"}