{"id":"https://openalex.org/W7133890756","doi":"https://doi.org/10.48550/arxiv.2603.04081","title":"Revisiting the Role of Foundation Models in Cell-Level Histopathological Image Analysis under Small-Patch Constraints -- Effects of Training Data Scale and Blur Perturbations on CNNs and Vision Transformers","display_name":"Revisiting the Role of Foundation Models in Cell-Level Histopathological Image Analysis under Small-Patch Constraints -- Effects of Training Data Scale and Blur Perturbations on CNNs and Vision Transformers","publication_year":2026,"publication_date":"2026-03-04","ids":{"openalex":"https://openalex.org/W7133890756","doi":"https://doi.org/10.48550/arxiv.2603.04081"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2603.04081","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},"type":"article","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5057943172","display_name":"Hiroki Kagiyama","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kagiyama, Hiroki","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128215345","display_name":"Toru Nagasaka","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nagasaka, Toru","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104250379","display_name":"Yukari Adachi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Adachi, Yukari","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121852267","display_name":"Takaaki Tachibana","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tachibana, Takaaki","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065670775","display_name":"Ryota Ito","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ito, Ryota","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041704850","display_name":"Mitsugu Fujita","orcid":"https://orcid.org/0000-0002-0000-9879"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fujita, Mitsugu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019586725","display_name":"Kimihiro Yamashita","orcid":"https://orcid.org/0000-0002-4208-7138"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yamashita, Kimihiro","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5024742184","display_name":"Yoshihiro Kakeji","orcid":"https://orcid.org/0000-0002-2727-0241"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kakeji, Yoshihiro","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":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.24862018,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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/T10862","display_name":"AI in cancer detection","score":0.7342000007629395,"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/T10862","display_name":"AI in cancer detection","score":0.7342000007629395,"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/T12859","display_name":"Cell Image Analysis Techniques","score":0.2029999941587448,"subfield":{"id":"https://openalex.org/subfields/1304","display_name":"Biophysics"},"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/T12874","display_name":"Digital Imaging for Blood Diseases","score":0.019099999219179153,"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/robustness","display_name":"Robustness (evolution)","score":0.6887000203132629},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5444999933242798},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.46709999442100525},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.46230000257492065},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.4618000090122223},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.43149998784065247},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.42320001125335693},{"id":"https://openalex.org/keywords/gaussian-blur","display_name":"Gaussian blur","score":0.41839998960494995}],"concepts":[{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7016000151634216},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6887000203132629},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6215999722480774},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5444999933242798},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.46709999442100525},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.46230000257492065},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.4618000090122223},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4390000104904175},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.43149998784065247},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.42320001125335693},{"id":"https://openalex.org/C104317376","wikidata":"https://www.wikidata.org/wiki/Q1894545","display_name":"Gaussian blur","level":5,"score":0.41839998960494995},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3725000023841858},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.3619999885559082},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.3377000093460083},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.32120001316070557},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.3118000030517578},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.2946000099182129},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.29339998960494995},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.2831999957561493},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.26829999685287476},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2606000006198883},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.259799987077713},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.2540999948978424}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2603.04081","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},{"id":"doi:10.48550/arxiv.2603.04081","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.04081","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2603.04081","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure","score":0.49613189697265625}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Background":[0],"and":[1,29,35,46,84,93,108,177,196],"objective:":[2],"Cell-level":[3],"pathological":[4],"image":[5,12],"analysis":[6],"requires":[7],"working":[8],"with":[9,60,117,145,156],"extremely":[10],"small":[11,134],"patches":[13,135],"(40x40":[14],"pixels),":[15],"far":[16],"below":[17],"standard":[18],"ImageNet":[19],"resolutions.":[20],"It":[21],"remains":[22],"unclear":[23],"whether":[24],"modern":[25],"deep":[26],"learning":[27],"architectures":[28,70,173],"foundation":[30,86,122,143,162,180],"models":[31,87,114,123,144,181,199],"can":[32],"learn":[33],"robust":[34],"scalable":[36],"representations":[37],"under":[38,168],"this":[39],"constraint.":[40],"We":[41,54],"systematically":[42],"evaluated":[43,89],"architectural":[44],"suitability":[45],"data-scale":[47],"effects":[48],"for":[49,133,161],"small-patch":[50,205],"cell":[51,66],"classification.":[52],"Methods:":[53],"analyzed":[55],"303":[56],"colorectal":[57],"cancer":[58],"specimens":[59],"CD103/CD8":[61],"immunostaining,":[62],"generating":[63],"185,432":[64],"annotated":[65],"images.":[67],"Eight":[68],"task-specific":[69,172],"were":[71,88],"trained":[72],"from":[73],"scratch":[74],"at":[75,125],"multiple":[76],"data":[77,119,185],"scales":[78],"(FlagLimit:":[79],"256--16,384":[80],"samples":[81],"per":[82],"class),":[83],"three":[85],"via":[90],"linear":[91],"probing":[92],"fine-tuning":[94],"after":[95],"resizing":[96],"inputs":[97],"to":[98,102],"224x224":[99],"pixels.":[100],"Robustness":[101],"blur":[103],"was":[104,152],"assessed":[105],"using":[106],"pre-":[107],"post-resize":[109],"Gaussian":[110],"perturbations.":[111],"Results:":[112],"Task-specific":[113],"improved":[115],"consistently":[116],"increasing":[118],"scale,":[120],"whereas":[121],"saturated":[124],"moderate":[126],"sample":[127],"sizes.":[128],"A":[129],"Vision":[130],"Transformer":[131],"optimized":[132],"(CustomViT)":[136],"achieved":[137],"the":[138,204],"highest":[139],"accuracy,":[140],"outperforming":[141],"all":[142],"substantially":[146],"lower":[147],"inference":[148],"cost.":[149],"Blur":[150],"robustness":[151],"comparable":[153],"across":[154],"architectures,":[155],"no":[157],"qualitative":[158],"advantage":[159],"observed":[160],"models.":[163],"Conclusion:":[164],"For":[165],"cell-level":[166],"classification":[167],"extreme":[169],"spatial":[170],"constraints,":[171],"are":[174,186],"more":[175],"effective":[176],"efficient":[178],"than":[179],"once":[182],"sufficient":[183],"training":[184],"available.":[187],"Higher":[188],"clean":[189],"accuracy":[190],"does":[191],"not":[192],"imply":[193],"superior":[194],"robustness,":[195],"large":[197],"pre-trained":[198],"offer":[200],"limited":[201],"benefit":[202],"in":[203],"regime.":[206]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-03-06T00:00:00"}