{"id":"https://openalex.org/W4411224726","doi":"https://doi.org/10.1016/j.cmpb.2026.109435","title":"The challenge of data scarcity and imbalanced classes in radiomics performance","display_name":"The challenge of data scarcity and imbalanced classes in radiomics performance","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W4411224726","doi":"https://doi.org/10.1016/j.cmpb.2026.109435","pmid":"https://pubmed.ncbi.nlm.nih.gov/42143960"},"language":"en","primary_location":{"id":"doi:10.1016/j.cmpb.2026.109435","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.cmpb.2026.109435","pdf_url":null,"source":{"id":"https://openalex.org/S138959710","display_name":"Computer Methods and Programs in Biomedicine","issn_l":"0169-2607","issn":["0169-2607","1872-7565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer Methods and Programs in Biomedicine","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1016/j.cmpb.2026.109435","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5038909009","display_name":"Pablo Rodr\u00edguez-Belenguer","orcid":"https://orcid.org/0000-0001-5270-7452"},"institutions":[{"id":"https://openalex.org/I4210166036","display_name":"Instituto de Investigaci\u00f3n Sanitaria La Fe","ror":"https://ror.org/05n7v5997","country_code":"ES","type":"nonprofit","lineage":["https://openalex.org/I4210166036"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Pablo Rodriguez-Belenguer","raw_affiliation_strings":["Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain"],"raw_orcid":"https://orcid.org/0000-0001-5270-7452","affiliations":[{"raw_affiliation_string":"Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain","institution_ids":["https://openalex.org/I4210166036"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028700235","display_name":"Mar\u00eda Beser\u2010Robles","orcid":"https://orcid.org/0000-0002-0072-5525"},"institutions":[{"id":"https://openalex.org/I4210166036","display_name":"Instituto de Investigaci\u00f3n Sanitaria La Fe","ror":"https://ror.org/05n7v5997","country_code":"ES","type":"nonprofit","lineage":["https://openalex.org/I4210166036"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Mar\u00eda Beser-Robles","raw_affiliation_strings":["Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain"],"raw_orcid":"https://orcid.org/0000-0002-0072-5525","affiliations":[{"raw_affiliation_string":"Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain","institution_ids":["https://openalex.org/I4210166036"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5117399588","display_name":"Manuel Marfil-Trujillo","orcid":null},"institutions":[{"id":"https://openalex.org/I4210166036","display_name":"Instituto de Investigaci\u00f3n Sanitaria La Fe","ror":"https://ror.org/05n7v5997","country_code":"ES","type":"nonprofit","lineage":["https://openalex.org/I4210166036"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Manuel Marfil-Trujillo","raw_affiliation_strings":["Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain"],"raw_orcid":"https://orcid.org/0009-0001-8573-1827","affiliations":[{"raw_affiliation_string":"Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain","institution_ids":["https://openalex.org/I4210166036"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106672327","display_name":"L. Cerd\u00e1 Alberich","orcid":"https://orcid.org/0000-0002-5567-4278"},"institutions":[{"id":"https://openalex.org/I4210166036","display_name":"Instituto de Investigaci\u00f3n Sanitaria La Fe","ror":"https://ror.org/05n7v5997","country_code":"ES","type":"nonprofit","lineage":["https://openalex.org/I4210166036"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Leonor Cerd\u00e1-Alberich","raw_affiliation_strings":["Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain. Electronic address: leonor_cerda@iislafe.es"],"raw_orcid":"https://orcid.org/0000-0002-5567-4278","affiliations":[{"raw_affiliation_string":"Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain. Electronic address: leonor_cerda@iislafe.es","institution_ids":["https://openalex.org/I4210166036"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5072387429","display_name":"Luis Mart\u00ed\u2010Bonmat\u00ed","orcid":"https://orcid.org/0000-0002-8234-010X"},"institutions":[{"id":"https://openalex.org/I4210166036","display_name":"Instituto de Investigaci\u00f3n Sanitaria La Fe","ror":"https://ror.org/05n7v5997","country_code":"ES","type":"nonprofit","lineage":["https://openalex.org/I4210166036"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Luis Mart\u00ed-Bonmat\u00ed","raw_affiliation_strings":["Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain"],"raw_orcid":"https://orcid.org/0000-0002-8234-010X","affiliations":[{"raw_affiliation_string":"Biomedical Imaging Research Group (GIBI230), Instituto de Investigaci\u00f3n Sanitaria La Fe, Valencia, Spain","institution_ids":["https://openalex.org/I4210166036"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210166036"],"apc_list":{"value":2880,"currency":"USD","value_usd":2880},"apc_paid":{"value":2880,"currency":"USD","value_usd":2880},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00333542,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"283","issue":null,"first_page":"109435","last_page":"109435"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12422","display_name":"Radiomics and Machine Learning in Medical Imaging","score":0.9991000294685364,"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"}},"topics":[{"id":"https://openalex.org/T12422","display_name":"Radiomics and Machine Learning in Medical Imaging","score":0.9991000294685364,"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/T10862","display_name":"AI in cancer detection","score":0.9688000082969666,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.9097999930381775,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/radiomics","display_name":"Radiomics","score":0.7934933304786682},{"id":"https://openalex.org/keywords/scarcity","display_name":"Scarcity","score":0.7316553592681885},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.4178132712841034},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38897496461868286},{"id":"https://openalex.org/keywords/data-science","display_name":"Data science","score":0.3713109493255615},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.32190221548080444},{"id":"https://openalex.org/keywords/economics","display_name":"Economics","score":0.22776994109153748},{"id":"https://openalex.org/keywords/microeconomics","display_name":"Microeconomics","score":0.06447136402130127}],"concepts":[{"id":"https://openalex.org/C2778559731","wikidata":"https://www.wikidata.org/wiki/Q23808793","display_name":"Radiomics","level":2,"score":0.7934933304786682},{"id":"https://openalex.org/C109747225","wikidata":"https://www.wikidata.org/wiki/Q815758","display_name":"Scarcity","level":2,"score":0.7316553592681885},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4178132712841034},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38897496461868286},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3713109493255615},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32190221548080444},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.22776994109153748},{"id":"https://openalex.org/C175444787","wikidata":"https://www.wikidata.org/wiki/Q39072","display_name":"Microeconomics","level":1,"score":0.06447136402130127}],"mesh":[{"descriptor_ui":"D000069550","descriptor_name":"Machine Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D000097188","descriptor_name":"Radiomics","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000097188","descriptor_name":"Radiomics","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000098446","descriptor_name":"Autoencoder","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D000098446","descriptor_name":"Autoencoder","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D005909","descriptor_name":"Glioblastoma","qualifier_ui":"Q000000981","qualifier_name":"diagnostic imaging","is_major_topic":false},{"descriptor_ui":"D005909","descriptor_name":"Glioblastoma","qualifier_ui":"Q000000981","qualifier_name":"diagnostic imaging","is_major_topic":false},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D008175","descriptor_name":"Lung Neoplasms","qualifier_ui":"Q000000981","qualifier_name":"diagnostic imaging","is_major_topic":false},{"descriptor_ui":"D008175","descriptor_name":"Lung Neoplasms","qualifier_ui":"Q000000981","qualifier_name":"diagnostic imaging","is_major_topic":false},{"descriptor_ui":"D008297","descriptor_name":"Male","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D008297","descriptor_name":"Male","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D011471","descriptor_name":"Prostatic Neoplasms","qualifier_ui":"Q000000981","qualifier_name":"diagnostic imaging","is_major_topic":false},{"descriptor_ui":"D011471","descriptor_name":"Prostatic Neoplasms","qualifier_ui":"Q000000981","qualifier_name":"diagnostic imaging","is_major_topic":false},{"descriptor_ui":"D015203","descriptor_name":"Reproducibility of Results","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D015203","descriptor_name":"Reproducibility of Results","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false}],"locations_count":3,"locations":[{"id":"doi:10.1016/j.cmpb.2026.109435","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.cmpb.2026.109435","pdf_url":null,"source":{"id":"https://openalex.org/S138959710","display_name":"Computer Methods and Programs in Biomedicine","issn_l":"0169-2607","issn":["0169-2607","1872-7565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer Methods and Programs in Biomedicine","raw_type":"journal-article"},{"id":"doi:10.2139/ssrn.5281759","is_oa":true,"landing_page_url":"https://doi.org/10.2139/ssrn.5281759","pdf_url":null,"source":{"id":"https://openalex.org/S4210172589","display_name":"SSRN Electronic Journal","issn_l":"1556-5068","issn":["1556-5068"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1318003438","host_organization_name":"RELX Group (Netherlands)","host_organization_lineage":["https://openalex.org/I1318003438"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"posted-content"},{"id":"pmid:42143960","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/42143960","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer methods and programs in biomedicine","raw_type":null}],"best_oa_location":{"id":"doi:10.1016/j.cmpb.2026.109435","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.cmpb.2026.109435","pdf_url":null,"source":{"id":"https://openalex.org/S138959710","display_name":"Computer Methods and Programs in Biomedicine","issn_l":"0169-2607","issn":["0169-2607","1872-7565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer Methods and Programs in Biomedicine","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320334923","display_name":"Instituto de Salud Carlos III","ror":"https://ror.org/00ca2c886"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W1556779599","https://openalex.org/W2011301426","https://openalex.org/W2137070227","https://openalex.org/W2936503027","https://openalex.org/W2972550981","https://openalex.org/W3014904955","https://openalex.org/W3035965352","https://openalex.org/W3120137913","https://openalex.org/W4386001451","https://openalex.org/W4388408234","https://openalex.org/W4393236711","https://openalex.org/W4400508677","https://openalex.org/W4402482840","https://openalex.org/W4406199868"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W4387369504","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"BACKGROUND":[0],"AND":[1],"OBJECTIVES:":[2],"Radiomics-based":[3],"machine":[4,108],"learning":[5,109],"models":[6,110,197],"are":[7],"increasingly":[8],"used":[9],"for":[10,116,122],"clinical":[11],"decision-making,":[12],"yet":[13],"their":[14,36],"reliability":[15],"is":[16,341],"often":[17],"constrained":[18],"by":[19],"limited":[20],"sample":[21,100,240],"sizes":[22],"and":[23,47,54,82,103,118,120,136,168,177,226,265,309,327],"class":[24,55,97,194],"imbalance":[25,56,322],"in":[26,80,165,184,190,200,211,250,254,267,288],"imaging":[27],"datasets.":[28],"While":[29],"prior":[30],"studies":[31],"addressed":[32,131],"these":[33],"challenges":[34],"independently,":[35],"combined":[37,105,280,344],"impact":[38,335],"remains":[39],"insufficiently":[40],"explored.":[41],"This":[42],"study":[43],"quantifies":[44],"the":[45,204,279],"individual":[46],"joint":[48],"effects":[49],"of":[50,336],"progressive":[51,96,99,193,239],"data":[52,137,339],"scarcity":[53,138,340],"on":[57,125],"radiomics":[58,68,325],"model":[59,166],"performance,":[60],"aiming":[61],"to":[62],"identify":[63],"mitigation":[64],"strategies.":[65,333],"METHODS:":[66],"Three":[67],"datasets-PI-CAI":[69],"(clinically":[70],"significant":[71,248],"prostate":[72],"cancer":[73,186],"detection),":[74],"BraTS2021":[75,123,219,255,299],"(MGMT":[76],"promoter":[77],"methylation":[78],"prediction":[79],"glioblastoma),":[81],"Hunter2023":[83,119,227,268,310],"(lung":[84],"nodule":[85],"malignancy":[86],"classification)-were":[87],"analyzed":[88],"under":[89,338,349],"four":[90],"conditions:":[91],"baseline":[92],"(balanced,":[93],"fixed-size":[94],"dataset),":[95],"imbalance,":[98,195],"size":[101,241],"reduction,":[102,242],"a":[104,162],"scenario.":[106],"Five":[107],"were":[111,174],"evaluated,":[112],"with":[113],"XGB":[114],"selected":[115],"PI-CAI":[117,212,289],"MLP":[121],"based":[124],"highest":[126],"balanced":[127,201,209,286],"accuracy.":[128,202],"Imbalance":[129],"was":[130,139,148],"using":[132,141,150],"state-of-the-art":[133],"sampling":[134,206,284],"techniques,":[135],"mitigated":[140,330],"Tabular":[142],"Variational":[143],"Autoencoders":[144],"(TVAE).":[145],"Statistical":[146],"significance":[147],"evaluated":[149],"paired":[151],"permutation":[152],"t-tests":[153],"across":[154],"repeated":[155],"cross-validation":[156],"splits.":[157],"RESULTS:":[158],"Feature":[159],"selection":[160],"played":[161],"key":[163],"role":[164],"performance":[167,326],"interpretability.":[169],"The":[170,334],"most":[171],"predictive":[172],"features":[173],"biologically":[175],"plausible":[176],"dataset-specific,":[178],"such":[179],"as":[180],"perinodular":[181],"texture":[182],"heterogeneity":[183],"lung":[185],"or":[187],"gray-level":[188],"non-uniformity":[189],"glioblastoma.":[191],"Under":[192,238],"unsampled":[196],"showed":[198,244],"degradation":[199,253],"Applying":[203],"best-performing":[205],"strategy":[207],"improved":[208],"accuracy":[210,287],"(\u0394":[213,220,228,256,269,290,300,311],"=":[214,217,221,224,229,232,257,260,270,273,291,294,297,301,304,312,315],"0.065;":[215],"dz":[216,223,231,259,272,293,303,314],"1.54),":[218],"0.057;":[222],"1.25),":[225],"0.087;":[230],"1.35),":[233],"all":[234],"p":[235,262,275,296,306,317],"<":[236,263,276,307,318],"0.001.":[237],"TVAE":[243,282],"dataset-dependent":[245],"effects:":[246],"no":[247],"change":[249],"PI-CAI,":[251],"moderate":[252],"-0.042;":[258],"-0.60;":[261],"0.001),":[264,308],"improvement":[266],"0.029;":[271],"0.46;":[274],"0.001).":[277],"In":[278],"scenario,":[281],"plus":[283],"increased":[285],"0.023;":[292],"0.42;":[295],"0.03),":[298],"0.070;":[302],"0.95;":[305],"0.041;":[313],"0.61;":[316],"0.01).":[319],"CONCLUSION:":[320],"Class":[321],"consistently":[323],"impairs":[324],"may":[328],"be":[329],"through":[331],"undersampling-based":[332],"augmentation":[337],"dataset-dependent,":[342],"but":[343],"correction":[345],"strategies":[346],"yield":[347],"gains":[348],"compounded":[350],"constraints.":[351]},"counts_by_year":[],"updated_date":"2026-08-11T07:18:39.950985","created_date":"2025-10-10T00:00:00"}