{"id":"https://openalex.org/W7161756922","doi":"https://doi.org/10.1109/isbi61048.2026.11516032","title":"When Vision-Language Models Look but Don't See: Anatomical Bias in Endoscopic Spatial Reasoning","display_name":"When Vision-Language Models Look but Don't See: Anatomical Bias in Endoscopic Spatial Reasoning","publication_year":2026,"publication_date":"2026-04-08","ids":{"openalex":"https://openalex.org/W7161756922","doi":"https://doi.org/10.1109/isbi61048.2026.11516032"},"language":null,"primary_location":{"id":"doi:10.1109/isbi61048.2026.11516032","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isbi61048.2026.11516032","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE 23rd International Symposium on Biomedical Imaging (ISBI)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5048727603","display_name":"Diego Bravo","orcid":"https://orcid.org/0000-0003-1957-1615"},"institutions":[{"id":"https://openalex.org/I36243813","display_name":"Universidad Nacional de Colombia","ror":"https://ror.org/059yx9a68","country_code":"CO","type":"education","lineage":["https://openalex.org/I36243813"]}],"countries":["CO"],"is_corresponding":false,"raw_author_name":"Diego Bravo","raw_affiliation_strings":["Universidad Nacional de Colombia,Computer Imaging and Medical Applications Laboratory (CIM@LAB)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Universidad Nacional de Colombia,Computer Imaging and Medical Applications Laboratory (CIM@LAB)","institution_ids":["https://openalex.org/I36243813"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064977863","display_name":"Dani\u00ebl De Wolf","orcid":"https://orcid.org/0000-0002-0846-8914"},"institutions":[{"id":"https://openalex.org/I196349391","display_name":"Universit\u00e4t Ulm","ror":"https://ror.org/032000t02","country_code":"DE","type":"education","lineage":["https://openalex.org/I196349391"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Daniel Wolf","raw_affiliation_strings":["Institute of Media Informatics, Ulm University,Visual Computing Group,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Media Informatics, Ulm University,Visual Computing Group,Germany","institution_ids":["https://openalex.org/I196349391"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120020437","display_name":"J C Hurtado-Tobar","orcid":null},"institutions":[{"id":"https://openalex.org/I36243813","display_name":"Universidad Nacional de Colombia","ror":"https://ror.org/059yx9a68","country_code":"CO","type":"education","lineage":["https://openalex.org/I36243813"]}],"countries":["CO"],"is_corresponding":false,"raw_author_name":"Juan Hurtado-Tobar","raw_affiliation_strings":["Universidad Nacional de Colombia,Computer Imaging and Medical Applications Laboratory (CIM@LAB)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Universidad Nacional de Colombia,Computer Imaging and Medical Applications Laboratory (CIM@LAB)","institution_ids":["https://openalex.org/I36243813"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022107537","display_name":"M B Juan G\u00f3mez","orcid":null},"institutions":[{"id":"https://openalex.org/I36243813","display_name":"Universidad Nacional de Colombia","ror":"https://ror.org/059yx9a68","country_code":"CO","type":"education","lineage":["https://openalex.org/I36243813"]}],"countries":["CO"],"is_corresponding":false,"raw_author_name":"Mart\u00edn G\u00f3mez","raw_affiliation_strings":["Universidad Nacional de Colombia,Computer Imaging and Medical Applications Laboratory (CIM@LAB)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Universidad Nacional de Colombia,Computer Imaging and Medical Applications Laboratory (CIM@LAB)","institution_ids":["https://openalex.org/I36243813"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5136509183","display_name":"Eduardo Romero","orcid":null},"institutions":[{"id":"https://openalex.org/I36243813","display_name":"Universidad Nacional de Colombia","ror":"https://ror.org/059yx9a68","country_code":"CO","type":"education","lineage":["https://openalex.org/I36243813"]}],"countries":["CO"],"is_corresponding":false,"raw_author_name":"Eduardo Romero","raw_affiliation_strings":["Universidad Nacional de Colombia,Computer Imaging and Medical Applications Laboratory (CIM@LAB)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Universidad Nacional de Colombia,Computer Imaging and Medical Applications Laboratory (CIM@LAB)","institution_ids":["https://openalex.org/I36243813"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"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":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11904","display_name":"Spatial Cognition and Navigation","score":0.5685999989509583,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11904","display_name":"Spatial Cognition and Navigation","score":0.5685999989509583,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.11500000208616257,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10656","display_name":"Child and Animal Learning Development","score":0.04659999907016754,"subfield":{"id":"https://openalex.org/subfields/3204","display_name":"Developmental and Educational Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.29840001463890076},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.2881999909877777},{"id":"https://openalex.org/keywords/spatial-intelligence","display_name":"Spatial intelligence","score":0.2628999948501587},{"id":"https://openalex.org/keywords/spatial-relation","display_name":"Spatial relation","score":0.262800008058548},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.2599000036716461}],"concepts":[{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6312999725341797},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5612999796867371},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4528000056743622},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.29840001463890076},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.2881999909877777},{"id":"https://openalex.org/C155911833","wikidata":"https://www.wikidata.org/wiki/Q3817354","display_name":"Spatial intelligence","level":2,"score":0.2628999948501587},{"id":"https://openalex.org/C27511587","wikidata":"https://www.wikidata.org/wiki/Q2178623","display_name":"Spatial relation","level":2,"score":0.262800008058548},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.25189998745918274},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.2459000051021576}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/isbi61048.2026.11516032","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isbi61048.2026.11516032","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE 23rd International Symposium on Biomedical Imaging (ISBI)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":7,"referenced_works":["https://openalex.org/W50237773","https://openalex.org/W1976684898","https://openalex.org/W2110288232","https://openalex.org/W2326962940","https://openalex.org/W3102722996","https://openalex.org/W4406495075","https://openalex.org/W4414348963"],"related_works":[],"abstract_inverted_index":{"Reliable":[0],"spatial":[1,49],"understanding":[2],"is":[3,83],"essential":[4],"for":[5,19,28,170],"complete":[6],"mucosal":[7],"inspection":[8],"during":[9],"upper":[10],"gastrointestinal":[11],"endoscopy,":[12],"particularly":[13],"under":[14,137],"the":[15,20,58,168],"Systematic":[16],"Screening":[17],"Protocol":[18],"Stomach":[21],"(SSS).":[22],"Vision-Language":[23],"Models":[24],"(VLMs)":[25],"show":[26],"promise":[27],"assisting":[29],"navigation":[30],"but":[31,151],"often":[32],"rely":[33,173],"on":[34,57,85,176],"textual":[35],"priors":[36],"rather":[37],"than":[38],"visual":[39,116,177],"information,":[40],"a":[41,54],"limitation":[42],"known":[43],"as":[44],"anatomical":[45,141,164],"bias.":[46],"To":[47],"evaluate":[48,124],"reasoning,":[50],"we":[51],"introduce":[52],"EndoSSS-RP,":[53],"benchmark":[55],"based":[56],"GastroHUN":[59],"dataset":[60],"(380":[61],"patients,":[62],"2,796":[63],"images,":[64],"3,678":[65],"binary":[66],"questions).":[67],"Each":[68],"question,":[69],"targeting":[70],"left/right":[71],"or":[72],"above/below":[73],"relationships":[74],"between":[75],"gastric":[76],"surfaces":[77],"(e.g.,":[78],"anterior":[79],"vs.":[80],"posterior":[81],"wall),":[82],"tested":[84],"original,":[86,146],"flipped,":[87,148],"and":[88,118,130,132,166,186],"rotated":[89],"images":[90],"to":[91,172],"assess":[92],"model":[93],"robustness":[94],"in":[95,179],"determining":[96],"relative":[97],"position":[98],"across":[99,105],"different":[100],"orientations.":[101],"Prompts":[102],"are":[103,189],"structured":[104],"three":[106],"levels:":[107],"L1":[108],"(anatomical":[109,113],"terms":[110,114],"only),":[111],"L2":[112],"with":[115,155],"markers),":[117],"L3":[119],"(visual":[120],"markers":[121],"only).":[122],"We":[123],"four":[125],"VLMs":[126],"(GPT-4o,":[127],"Gemini-2.5-Flash,":[128],"JanusPro-7B,":[129],"LLaMA-3.2)":[131],"find":[133],"that":[134],"accuracy":[135],"declines":[136],"transformations":[138],"when":[139],"using":[140],"prompts":[142,158],"(L1,":[143],"GPT-4o:":[144],"68%":[145],"61%":[147],"46%":[149],"rotated)":[150],"perform":[152],"consistently":[153],"better":[154],"visual\u2013marker":[156],"only":[157],"(L3:":[159],"87%).":[160],"These":[161],"results":[162],"reveal":[163],"bias":[165],"underscore":[167],"need":[169],"models":[171],"more":[174],"heavily":[175],"evidence":[178],"endoscopic":[180],"AI":[181],"systems.":[182],"Code,":[183],"data,":[184],"prompts,":[185],"evaluation":[187],"scripts":[188],"available":[190],"at:":[191],"https://github.com/DiegoBravoH/EndoNavQA.":[192]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-05-21T00:00:00"}
