{"id":"https://openalex.org/W7204076747","doi":"https://doi.org/10.48550/arxiv.2608.21140","title":"A Modular Agent for Reliable and Auditable Spatial Relation Verification in CT Scans","display_name":"A Modular Agent for Reliable and Auditable Spatial Relation Verification in CT Scans","publication_year":2026,"publication_date":"2026-08-21","ids":{"openalex":"https://openalex.org/W7204076747","doi":"https://doi.org/10.48550/arxiv.2608.21140"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.21140","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.21140","pdf_url":"https://arxiv.org/pdf/2608.21140","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":"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/2608.21140","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148033936","display_name":"Simon Vincent Abel","orcid":null,"observed_orcids":[]},"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":"Abel, Simon Vincent","raw_affiliation_strings":["Visual Computing Group , Institute of Media Informatics , Ulm University , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Visual Computing Group , Institute of Media Informatics , Ulm University , Germany","institution_ids":["https://openalex.org/I196349391"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119675354","display_name":"Heiko Hillenhagen","orcid":"https://orcid.org/0000-0002-2749-6726","observed_orcids":["https://orcid.org/0000-0002-2749-6726"]},"institutions":[{"id":"https://openalex.org/I4210155144","display_name":"University Hospital Ulm","ror":"https://ror.org/05emabm63","country_code":"DE","type":"healthcare","lineage":["https://openalex.org/I4210155144"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Hillenhagen, Heiko","raw_affiliation_strings":["Diagnostic and Interventional Radiology , Ulm University Hospital , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Diagnostic and Interventional Radiology , Ulm University Hospital , Germany","institution_ids":["https://openalex.org/I4210155144"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110148964","display_name":"MICHAEL A. GOETZ","orcid":"https://orcid.org/0000-0003-0984-224X","observed_orcids":["https://orcid.org/0000-0003-0984-224X"]},"institutions":[{"id":"https://openalex.org/I4210155144","display_name":"University Hospital Ulm","ror":"https://ror.org/05emabm63","country_code":"DE","type":"healthcare","lineage":["https://openalex.org/I4210155144"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"G\u00f6tz, Michael","raw_affiliation_strings":["Diagnostic and Interventional Radiology , Ulm University Hospital , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Diagnostic and Interventional Radiology , Ulm University Hospital , Germany","institution_ids":["https://openalex.org/I4210155144"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061059879","display_name":"Timo Ropinski","orcid":"https://orcid.org/0000-0002-7857-5512","observed_orcids":["https://orcid.org/0000-0002-7857-5512"]},"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":"Ropinski, Timo","raw_affiliation_strings":["Visual Computing Group , Institute of Media Informatics , Ulm University , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Visual Computing Group , Institute of Media Informatics , Ulm University , Germany","institution_ids":["https://openalex.org/I196349391"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041700318","display_name":"Ayhan Can Erdur","orcid":"https://orcid.org/0009-0006-0290-0434","observed_orcids":["https://orcid.org/0009-0006-0290-0434"]},"institutions":[{"id":"https://openalex.org/I2802619606","display_name":"TUM Klinikum","ror":"https://ror.org/04jc43x05","country_code":"DE","type":"healthcare","lineage":["https://openalex.org/I2802619606","https://openalex.org/I62916508"]},{"id":"https://openalex.org/I62916508","display_name":"Technical University of Munich","ror":"https://ror.org/02kkvpp62","country_code":"DE","type":"education","lineage":["https://openalex.org/I62916508"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Erdur, Ayhan Can","raw_affiliation_strings":["Chair for AI in Healthcare and Medicine , Technical University of Munich (TUM) and","Department of Radiation Oncology , TUM University Hospital , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chair for AI in Healthcare and Medicine , Technical University of Munich (TUM) and","institution_ids":["https://openalex.org/I62916508"]},{"raw_affiliation_string":"Department of Radiation Oncology , TUM University Hospital , Germany","institution_ids":["https://openalex.org/I2802619606","https://openalex.org/I62916508"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5147085619","display_name":"Daniel Santak Wolf","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wolf, Daniel Santak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9976000189781189,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9976000189781189,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.0019000000320374966,"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/T10052","display_name":"Medical Image Segmentation Techniques","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/computed-tomography","display_name":"computed tomography","score":0.6790000200271606},{"id":"https://openalex.org/keywords/vision-language-models","display_name":"vision-language models","score":0.7770000100135803},{"id":"https://openalex.org/keywords/anatomical-localization","display_name":"anatomical localization","score":0.6779999732971191},{"id":"https://openalex.org/keywords/yolo","display_name":"YOLO","score":0.8029999732971191},{"id":"https://openalex.org/keywords/geometric-verification","display_name":"geometric verification","score":0.5329999923706055},{"id":"https://openalex.org/keywords/natural-language-questions","display_name":"natural language queries","score":0.4410000145435333},{"id":"https://openalex.org/keywords/interpretability","display_name":"interpretability","score":0.6930000185966492}],"concepts":[{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.766700029373169},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6919999718666077},{"id":"https://openalex.org/C27511587","wikidata":"https://www.wikidata.org/wiki/Q2178623","display_name":"Spatial relation","level":2,"score":0.656499981880188},{"id":"https://openalex.org/C155911833","wikidata":"https://www.wikidata.org/wiki/Q3817354","display_name":"Spatial intelligence","level":2,"score":0.6281999945640564},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5839999914169312},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.5329999923706055},{"id":"https://openalex.org/C31601959","wikidata":"https://www.wikidata.org/wiki/Q931309","display_name":"Medical imaging","level":2,"score":0.51419997215271},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.47119998931884766},{"id":"https://openalex.org/C159620131","wikidata":"https://www.wikidata.org/wiki/Q1938983","display_name":"Spatial analysis","level":2,"score":0.45399999618530273},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4528000056743622},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.4072999954223633},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4007999897003174},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.37549999356269836},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3605000078678131},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.3560999929904938},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3458999991416931},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3260999917984009},{"id":"https://openalex.org/C203689450","wikidata":"https://www.wikidata.org/wiki/Q2302053","display_name":"Spatial database","level":3,"score":0.3046000003814697},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.2881999909877777},{"id":"https://openalex.org/C111498074","wikidata":"https://www.wikidata.org/wiki/Q173326","display_name":"Formal verification","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2702000141143799},{"id":"https://openalex.org/C65180967","wikidata":"https://www.wikidata.org/wiki/Q130901","display_name":"Binary relation","level":2,"score":0.2621999979019165},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.2567000091075897}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.21140","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.21140","pdf_url":"https://arxiv.org/pdf/2608.21140","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.21140","doi":"https://doi.org/10.48550/arxiv.2608.21140","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.21140","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:2608.21140","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.21140","pdf_url":"https://arxiv.org/pdf/2608.21140","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G7743612666","display_name":null,"funder_award_id":"520750254","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"}],"funders":[{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204076747.pdf","grobid_xml":"https://content.openalex.org/works/W7204076747.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W4403345778","https://openalex.org/W4403444126","https://openalex.org/W4403853618","https://openalex.org/W4404330874","https://openalex.org/W4414687708","https://openalex.org/W4415198530","https://openalex.org/W6947991961","https://openalex.org/W7154719127","https://openalex.org/W7155053861"],"related_works":["https://openalex.org/W7135060579","https://openalex.org/W7204820958","https://openalex.org/W7172427996","https://openalex.org/W7160158375","https://openalex.org/W7160911644","https://openalex.org/W7154719127","https://openalex.org/W7160090851","https://openalex.org/W7138125993","https://openalex.org/W7124159817","https://openalex.org/W7165391909"],"abstract_inverted_index":{"Reliable":[0],"spatial":[1,44,52,73,89,100,138,157,203],"understanding":[2,63],"is":[3,140],"an":[4],"important":[5],"prerequisite":[6],"for":[7,87,212],"future":[8,213],"medical":[9,33,84,215],"vision-language":[10,25],"systems":[11],"that":[12,58,200],"aim":[13],"to":[14,49,77],"support":[15],"radiological":[16,59],"report":[17],"generation":[18],"and":[19,46,70,115,135,160,175,193],"structured":[20,124],"image":[21,55],"understanding.":[22],"While":[23],"modern":[24],"models":[26],"(VLMs)":[27],"show":[28],"promising":[29,209],"performance":[30],"on":[31,62,153],"many":[32],"imaging":[34,85,216],"tasks,":[35],"recent":[36],"evidence":[37],"suggests":[38],"they":[39],"remain":[40],"weak":[41],"in":[42,54,92,186],"controlled":[43],"reasoning":[45,60,195],"often":[47],"fail":[48],"reliably":[50],"ground":[51],"relations":[53],"evidence.":[56],"Given":[57],"hinges":[61],"the":[64,103,106,136,151,154],"relative":[65],"positions":[66],"of":[67,97],"anatomical":[68,113],"structures":[69],"findings,":[71],"this":[72],"weakness":[74],"poses":[75],"risks":[76],"diagnostic":[78],"accuracy.":[79],"We":[80,149],"present":[81],"a":[82,132,208],"modular":[83,202],"agent":[86],"binary":[88],"relation":[90,125],"verification":[91,204],"axial":[93],"CT":[94],"slices.":[95],"Instead":[96],"directly":[98],"predicting":[99],"answers":[101],"end-to-end,":[102],"system":[104],"decomposes":[105],"task":[107],"into":[108,123],"explicit":[109,201],"stages:":[110],"language":[111],"parsing,":[112],"localization,":[114],"deterministic":[116,146],"geometric":[117,147],"verification.":[118],"Natural-language":[119],"queries":[120],"are":[121,129],"converted":[122],"tuples,":[126],"queried":[127],"organs":[128],"localized":[130],"with":[131],"YOLO-based":[133],"detector,":[134],"final":[137],"decision":[139],"computed":[141],"from":[142],"object":[143],"centers":[144],"using":[145],"rules.":[148],"evaluate":[150],"approach":[152],"held-out":[155],"MIRP":[156],"QA":[158],"benchmark":[159],"compare":[161],"it":[162],"against":[163],"representative":[164],"end-to-end":[165],"VLM":[166],"baselines.":[167],"The":[168,197],"best-performing":[169],"hybrid":[170],"configuration":[171],"reaches":[172],"94.1%":[173],"accuracy":[174],"94.2%":[176],"F1,":[177],"outperforming":[178],"direct":[179],"Qwen2-VL":[180],"prompting":[181],"by":[182],"42.5":[183],"percentage":[184],"points":[185],"accuracy,":[187],"while":[188],"preserving":[189],"interpretable":[190],"intermediate":[191],"representations":[192],"auditable":[194],"stages.":[196],"results":[198],"suggest":[199],"can":[205],"serve":[206],"as":[207],"building":[210],"block":[211],"report-oriented":[214],"agents.":[217]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-08-25T00:00:00"}