{"id":"https://openalex.org/W7134072215","doi":"https://doi.org/10.48550/arxiv.2603.04878","title":"Structure Observation Driven Image-Text Contrastive Learning for Computed Tomography Report Generation","display_name":"Structure Observation Driven Image-Text Contrastive Learning for Computed Tomography Report Generation","publication_year":2026,"publication_date":"2026-03-05","ids":{"openalex":"https://openalex.org/W7134072215","doi":"https://doi.org/10.48550/arxiv.2603.04878"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.04878","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.04878","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.04878","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128258006","display_name":"Hong Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Hong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128232202","display_name":"Dong Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Dong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128220946","display_name":"Qiong Peng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng, Qiong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Huang, Yawen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Yawen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128224897","display_name":"Xian Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128246364","display_name":"Yefeng Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Yefeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100613491","display_name":"Liansheng Wang","orcid":"https://orcid.org/0000-0002-2096-454X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Liansheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":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":0.7900999784469604,"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":0.7900999784469604,"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/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.027799999341368675,"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/T10028","display_name":"Topic Modeling","score":0.026499999687075615,"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/set","display_name":"Set (abstract data type)","score":0.5350000262260437},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4690999984741211},{"id":"https://openalex.org/keywords/structured-prediction","display_name":"Structured prediction","score":0.45260000228881836},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.447299987077713},{"id":"https://openalex.org/keywords/data-set","display_name":"Data set","score":0.42590001225471497},{"id":"https://openalex.org/keywords/computed-tomography","display_name":"Computed tomography","score":0.4187000095844269}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7767999768257141},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6772000193595886},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5350000262260437},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4690999984741211},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.45260000228881836},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.447299987077713},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.42590001225471497},{"id":"https://openalex.org/C544519230","wikidata":"https://www.wikidata.org/wiki/Q32566","display_name":"Computed tomography","level":2,"score":0.4187000095844269},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.39430001378059387},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.387800008058548},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.37209999561309814},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.36079999804496765},{"id":"https://openalex.org/C31601959","wikidata":"https://www.wikidata.org/wiki/Q931309","display_name":"Medical imaging","level":2,"score":0.3402000069618225},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.3319000005722046},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.28200000524520874},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.25769999623298645},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.2522999942302704}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.04878","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.04878","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.04878","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.04878","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":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.5938040018081665,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Computed":[0],"Tomography":[1],"Report":[2],"Generation":[3],"(CTRG)":[4],"aims":[5],"to":[6,44,56,126,165,171,187],"automate":[7],"the":[8,15,78,107,128,146,167,176,179,189],"clinical":[9,235],"radiology":[10,109],"reporting":[11],"process,":[12],"thereby":[13],"reducing":[14,204],"workload":[16],"of":[17,48,83,130],"report":[18,34,110,216],"writing":[19],"and":[20,51,66,138,143,155,185,213,237],"facilitating":[21],"patient":[22],"care.":[23],"While":[24],"deep":[25],"learning":[26,170],"approaches":[27],"have":[28],"achieved":[29],"remarkable":[30],"advances":[31],"in":[32,41,91,169,234],"X-ray":[33],"generation,":[35],"their":[36],"effectiveness":[37],"may":[38],"be":[39],"limited":[40],"CTRG":[42,71,233],"due":[43],"larger":[45],"data":[46],"volumes":[47],"CT":[49,93,153],"images":[50,142,154],"more":[52],"intricate":[53],"details":[54],"required":[55],"describe":[57],"them.":[58],"This":[59],"work":[60],"introduces":[61],"a":[62,81,92,112,158,208],"novel":[63],"two-stage":[64],"(structure-":[65],"report-learning)":[67],"framework":[68,227],"tailored":[69],"for":[70,215,232],"featuring":[72],"effective":[73],"structure-wise":[74,113],"image-text":[75,114],"contrasting.":[76],"In":[77,117,175],"first":[79],"stage,":[80,178],"set":[82],"learnable":[84],"structure-specific":[85,102],"visual":[86,180],"queries":[87,182],"observe":[88],"corresponding":[89],"structures":[90,137],"image.":[94],"The":[95],"resulting":[96],"observation":[97],"tokens":[98],"are":[99,124,183,240],"contrasted":[100],"with":[101,111],"textual":[103],"features":[104],"extracted":[105],"from":[106,140,200],"accompanying":[108],"contrastive":[115],"loss.":[116],"addition,":[118],"text-text":[119],"similarity-based":[120],"soft":[121],"pseudo":[122],"targets":[123],"proposed":[125,164],"mitigate":[127],"impact":[129],"false":[131],"negatives,":[132],"i.e.,":[133],"semantically":[134],"identical":[135],"image":[136,191],"texts":[139],"non-paired":[141],"reports.":[144,156],"Thus,":[145],"model":[147],"learns":[148],"structure-level":[149],"semantic":[150],"correspondences":[151],"between":[152],"Further,":[157],"dynamic,":[159],"diversity-enhanced":[160],"negative":[161],"queue":[162],"is":[163,211],"guide":[166],"network":[168],"discriminate":[172],"various":[173],"abnormalities.":[174],"second":[177],"structure":[181],"frozen":[184],"used":[186],"select":[188],"critical":[190],"patch":[192],"embeddings":[193],"depicting":[194],"each":[195],"anatomical":[196],"structure,":[197],"minimizing":[198],"distractions":[199],"irrelevant":[201],"areas":[202],"while":[203],"memory":[205],"consumption.":[206],"Also,":[207],"text":[209],"decoder":[210],"added":[212],"trained":[214],"generation.Our":[217],"extensive":[218],"experiments":[219],"on":[220],"two":[221],"public":[222],"datasets":[223],"demonstrate":[224],"that":[225],"our":[226],"establishes":[228],"new":[229],"state-of-the-art":[230],"performance":[231],"efficiency,":[236],"its":[238],"components":[239],"effective.":[241]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-07T00:00:00"}