{"id":"https://openalex.org/W7208838808","doi":null,"title":"Sharpening the Ensemble: An SSIM-Aligned Residual Refiner for Brain-MRI Inpainting Post-Processing","display_name":"Sharpening the Ensemble: An SSIM-Aligned Residual Refiner for Brain-MRI Inpainting Post-Processing","publication_year":2026,"publication_date":"2026-09-03","ids":{"openalex":"https://openalex.org/W7208838808"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.03981","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.03981","pdf_url":"https://arxiv.org/pdf/2609.03981","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","indexed_in":["arxiv"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.03981","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5012673487","display_name":"Kubilay K. K\u00f6m\u00fcrc\u00fc","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I48912391","display_name":"Istanbul Technical University","ror":"https://ror.org/059636586","country_code":"TR","type":"education","lineage":["https://openalex.org/I48912391"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"K\u00f6m\u00fcrc\u00fc, Kubilay Ka\u011fan","raw_affiliation_strings":["Department of Computer Engineering , Istanbul Technical University , Istanbul , T\u00fcrkiye"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Engineering , Istanbul Technical University , Istanbul , T\u00fcrkiye","institution_ids":["https://openalex.org/I48912391"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5080738214","display_name":"\u0130lkay \u00d6ks\u00fcz","orcid":"https://orcid.org/0000-0001-6478-0534","observed_orcids":["https://orcid.org/0000-0001-6478-0534"]},"institutions":[{"id":"https://openalex.org/I48912391","display_name":"Istanbul Technical University","ror":"https://ror.org/059636586","country_code":"TR","type":"education","lineage":["https://openalex.org/I48912391"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"\u00d6ks\u00fcz, \u0130lkay","raw_affiliation_strings":["Department of Computer Engineering , Istanbul Technical University , Istanbul , T\u00fcrkiye"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Engineering , Istanbul Technical University , Istanbul , T\u00fcrkiye","institution_ids":["https://openalex.org/I48912391"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I48912391"],"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.5630000233650208,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.5630000233650208,"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/T10378","display_name":"Advanced MRI Techniques and Applications","score":0.23190000653266907,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"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":0.07190000265836716,"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/structural-similarity-index","display_name":"structural similarity index","score":0.75},{"id":"https://openalex.org/keywords/ensemble-deep-learning","display_name":"ensemble deep learning","score":0.6399999856948853},{"id":"https://openalex.org/keywords/post-processing","display_name":"post-processing","score":0.7590000033378601},{"id":"https://openalex.org/keywords/brats","display_name":"BraTS","score":0.4569999873638153},{"id":"https://openalex.org/keywords/structural-similarity-loss","display_name":"structural similarity loss","score":0.4690000116825104},{"id":"https://openalex.org/keywords/unsharp-masking","display_name":"unsharp masking","score":0.7760000228881836},{"id":"https://openalex.org/keywords/image-sharpening","display_name":"image sharpening","score":0.625}],"concepts":[{"id":"https://openalex.org/C2781137444","wikidata":"https://www.wikidata.org/wiki/Q237105","display_name":"Sharpening","level":2,"score":0.8953999876976013},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.7839999794960022},{"id":"https://openalex.org/C11727466","wikidata":"https://www.wikidata.org/wiki/Q1628157","display_name":"Inpainting","level":3,"score":0.7134000062942505},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6599000096321106},{"id":"https://openalex.org/C79052406","wikidata":"https://www.wikidata.org/wiki/Q365715","display_name":"Unsharp masking","level":4,"score":0.5824000239372253},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5486999750137329},{"id":"https://openalex.org/C139945424","wikidata":"https://www.wikidata.org/wiki/Q1940696","display_name":"Mean squared error","level":2,"score":0.49639999866485596},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.48089998960494995},{"id":"https://openalex.org/C2777402240","wikidata":"https://www.wikidata.org/wiki/Q6783436","display_name":"Masking (illustration)","level":2,"score":0.45840001106262207},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4505999982357025},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.42910000681877136},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.38609999418258667},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3815999925136566},{"id":"https://openalex.org/C167085575","wikidata":"https://www.wikidata.org/wiki/Q6803654","display_name":"Mean squared prediction error","level":2,"score":0.3163999915122986},{"id":"https://openalex.org/C119898033","wikidata":"https://www.wikidata.org/wiki/Q3433888","display_name":"Ensemble forecasting","level":2,"score":0.30640000104904175},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3061000108718872},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.28769999742507935},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.275299996137619},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.271699994802475},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.258899986743927},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.25}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:arXiv.org:2609.03981","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.03981","pdf_url":"https://arxiv.org/pdf/2609.03981","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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.03981","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.03981","pdf_url":"https://arxiv.org/pdf/2609.03981","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":[],"funders":[{"id":"https://openalex.org/F4320309480","display_name":"Nvidia","ror":"https://ror.org/03jdj4y14"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7208838808.pdf","grobid_xml":"https://content.openalex.org/works/W7208838808.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4415878258","https://openalex.org/W7171959641","https://openalex.org/W4405852678","https://openalex.org/W7203440928","https://openalex.org/W7106805842","https://openalex.org/W7220984894","https://openalex.org/W3189088695","https://openalex.org/W7208742756","https://openalex.org/W7152001073","https://openalex.org/W4417129430"],"abstract_inverted_index":{"Brain-MRI":[0],"inpainting":[1],"replaces":[2],"a":[3,7,92,103,118,127,142,178,230],"masked":[4],"region":[5],"of":[6,76,95,145,173],"scan":[8],"with":[9,117,159,177],"synthesized,":[10],"anatomically":[11],"plausible":[12],"healthy":[13,21],"tissue,":[14],"so":[15,218],"that":[16,235],"analysis":[17],"tools":[18],"built":[19],"for":[20],"brains":[22],"can":[23],"be":[24],"applied":[25],"to":[26,72,139,152,197,208],"images":[27],"they":[28],"would":[29],"otherwise":[30],"reject.":[31],"On":[32],"the":[33,41,46,51,57,73,77,96,108,130,135,146,155,174,183,191,198,219],"BraTS":[34],"local-synthesis":[35],"benchmark,":[36],"which":[37],"ranks":[38],"submissions":[39],"on":[40,107,141,154],"structural":[42,184],"similarity":[43],"index":[44],"(SSIM),":[45],"peak":[47],"signal-to-noise":[48],"ratio,":[49],"and":[50,69,79,101,149,203],"mean":[52],"squared":[53],"error":[54],"(MSE)":[55],"jointly,":[56],"strongest":[58],"recent":[59],"models":[60,100],"are":[61],"accurate,":[62],"but":[63,169],"several":[64],"report":[65],"blurry":[66],"synthesized":[67],"regions":[68],"attribute":[70],"this":[71,88],"mean-seeking":[74],"behavior":[75],"$\\ell_1$":[78,114],"MSE":[80],"terms":[81],"in":[82,89,163],"their":[83],"training":[84,102],"losses.":[85],"We":[86],"address":[87],"post-processing,":[90],"forming":[91],"deep":[93],"ensemble":[94,200,240],"two":[97],"co-first-place":[98],"2025":[99],"lightweight":[104],"residual":[105],"refiner":[106,131],"ensemble's":[109],"own":[110],"outputs":[111],"under":[112],"an":[113,237],"loss":[115],"augmented":[116],"structural-similarity":[119],"term":[120,185],"whose":[121],"weight":[122],"$\u03bb$":[123,129],"we":[124],"vary.":[125],"At":[126],"moderate":[128],"improves":[132,236],"SSIM":[133,210],"over":[134],"ensemble,":[136],"from":[137,150],"$0.8767$":[138],"$0.8780$":[140],"held-out":[143,175],"reproduction":[144],"official":[147,156],"scorer":[148],"$0.8555$":[151],"$0.8572$":[153],"validation":[157],"leaderboard,":[158],"essentially":[160],"no":[161],"change":[162],"MSE.":[164],"The":[165,227],"gain":[166,220],"is":[167,229],"small":[168],"consistent,":[170],"improving":[171],"$62.6\\%$":[172],"cases":[176],"signed-rank":[179],"$p=2.2\\times10^{-7}$,":[180],"whereas":[181],"over-weighting":[182],"reverses":[186],"it.":[187],"Two":[188],"ablations":[189],"bound":[190],"effect.":[192],"Adding":[193],"any":[194,212,242],"third":[195],"model":[196],"two-model":[199],"degrades":[201],"it,":[202],"classical":[204],"unsharp":[205],"masking":[206],"fails":[207],"improve":[209],"at":[211],"strength":[213],"(best":[214],"$0.8765$":[215],"against":[216],"$0.8767$),":[217],"reflects":[221],"learned":[222],"rather":[223],"than":[224],"indiscriminate":[225],"sharpening.":[226],"result":[228],"cheap,":[231],"reproducible":[232],"post-processing":[233],"stage":[234],"already":[238],"strong":[239],"without":[241],"large-scale":[243],"retraining.":[244]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-09-06T00:00:00"}