{"id":"https://openalex.org/W7160428058","doi":"https://doi.org/10.1016/j.ijmedinf.2026.106464","title":"How improper dataset split hinders model generalizability: a systematic comparison in Human activity recognition and exercise evaluation tasks","display_name":"How improper dataset split hinders model generalizability: a systematic comparison in Human activity recognition and exercise evaluation tasks","publication_year":2026,"publication_date":"2026-05-03","ids":{"openalex":"https://openalex.org/W7160428058","doi":"https://doi.org/10.1016/j.ijmedinf.2026.106464","pmid":"https://pubmed.ncbi.nlm.nih.gov/42097112"},"language":"en","primary_location":{"id":"doi:10.1016/j.ijmedinf.2026.106464","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.ijmedinf.2026.106464","pdf_url":null,"source":{"id":"https://openalex.org/S53142634","display_name":"International Journal of Medical Informatics","issn_l":"1386-5056","issn":["1386-5056","1872-8243"],"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":"International Journal of Medical Informatics","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.ijmedinf.2026.106464","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5106957744","display_name":"Samuele Pe","orcid":null},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":true,"raw_author_name":"Samuele Pe","raw_affiliation_strings":["Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy. Electronic address: samuele.pe01@universitadipavia.it"],"raw_orcid":"https://orcid.org/0009-0002-7877-0294","affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy. Electronic address: samuele.pe01@universitadipavia.it","institution_ids":["https://openalex.org/I25217355"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068449833","display_name":"Arianna Dagliati","orcid":"https://orcid.org/0000-0002-5041-0409"},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Arianna Dagliati","raw_affiliation_strings":["Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy","institution_ids":["https://openalex.org/I25217355"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090028219","display_name":"Enea Parimbelli","orcid":"https://orcid.org/0000-0003-0679-828X"},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Enea Parimbelli","raw_affiliation_strings":["Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy; Telfer School of Management, University of Ottawa, Ottawa, Ontario K1N 6N5, Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy; Telfer School of Management, University of Ottawa, Ottawa, Ontario K1N 6N5, Canada","institution_ids":["https://openalex.org/I25217355"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017956054","display_name":"Bruna Maria Vittoria Guerra","orcid":"https://orcid.org/0000-0002-5169-4431"},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Bruna Maria Vittoria Guerra","raw_affiliation_strings":["Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy","institution_ids":["https://openalex.org/I25217355"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080420778","display_name":"Stefania Sozzi","orcid":"https://orcid.org/0000-0002-6837-651X"},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Stefania Sozzi","raw_affiliation_strings":["Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy","institution_ids":["https://openalex.org/I25217355"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135440882","display_name":"Riccardo Bellazzi","orcid":null},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Riccardo Bellazzi","raw_affiliation_strings":["Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy","institution_ids":["https://openalex.org/I25217355"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5044714123","display_name":"Giovanna Nicora","orcid":"https://orcid.org/0000-0001-7007-0862"},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Giovanna Nicora","raw_affiliation_strings":["Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer, and Biomedical Engineering, University of Pavia 27100 Pavia, Italy","institution_ids":["https://openalex.org/I25217355"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5106957744"],"corresponding_institution_ids":["https://openalex.org/I25217355"],"apc_list":{"value":3210,"currency":"USD","value_usd":3210},"apc_paid":{"value":3210,"currency":"USD","value_usd":3210},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.55489606,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"216","issue":null,"first_page":"106464","last_page":"106464"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.4921000003814697,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.4921000003814697,"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/T10444","display_name":"Context-Aware Activity Recognition Systems","score":0.3646000027656555,"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/T10667","display_name":"Emotion and Mood Recognition","score":0.011599999852478504,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/activity-recognition","display_name":"Activity recognition","score":0.7085000276565552},{"id":"https://openalex.org/keywords/physical-activity","display_name":"Physical activity","score":0.34540000557899475},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.34040001034736633},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.2978000044822693},{"id":"https://openalex.org/keywords/task-analysis","display_name":"Task analysis","score":0.2718000113964081}],"concepts":[{"id":"https://openalex.org/C121687571","wikidata":"https://www.wikidata.org/wiki/Q4677630","display_name":"Activity recognition","level":2,"score":0.7085000276565552},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6407999992370605},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4535999894142151},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3774000108242035},{"id":"https://openalex.org/C3020255362","wikidata":"https://www.wikidata.org/wiki/Q747883","display_name":"Physical activity","level":2,"score":0.34540000557899475},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.34040001034736633},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.30660000443458557},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2978000044822693},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2718000113964081},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2694999873638153},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.25369998812675476}],"mesh":[{"descriptor_ui":"D000069550","descriptor_name":"Machine Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D000069550","descriptor_name":"Machine Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D000069550","descriptor_name":"Machine Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000069550","descriptor_name":"Machine Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000077321","descriptor_name":"Deep Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D000077321","descriptor_name":"Deep Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D000077321","descriptor_name":"Deep Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000077321","descriptor_name":"Deep Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000098412","descriptor_name":"Predictive Learning Models","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D000098412","descriptor_name":"Predictive Learning Models","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D003198","descriptor_name":"Computer Simulation","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":"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":"D006801","descriptor_name":"Humans","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D006802","descriptor_name":"Human Activities","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D006802","descriptor_name":"Human Activities","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D006802","descriptor_name":"Human Activities","qualifier_ui":"Q000145","qualifier_name":"classification","is_major_topic":true},{"descriptor_ui":"D006802","descriptor_name":"Human Activities","qualifier_ui":"Q000145","qualifier_name":"classification","is_major_topic":true},{"descriptor_ui":"D012046","descriptor_name":"Rehabilitation","qualifier_ui":"Q000379","qualifier_name":"methods","is_major_topic":false},{"descriptor_ui":"D015444","descriptor_name":"Exercise","qualifier_ui":"Q000502","qualifier_name":"physiology","is_major_topic":true},{"descriptor_ui":"D015444","descriptor_name":"Exercise","qualifier_ui":"Q000502","qualifier_name":"physiology","is_major_topic":true},{"descriptor_ui":"D015444","descriptor_name":"Exercise","qualifier_ui":"Q000502","qualifier_name":"physiology","is_major_topic":false},{"descriptor_ui":"D015444","descriptor_name":"Exercise","qualifier_ui":"Q000502","qualifier_name":"physiology","is_major_topic":true},{"descriptor_ui":"D016208","descriptor_name":"Databases, Factual","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D066264","descriptor_name":"Datasets as Topic","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D066264","descriptor_name":"Datasets as Topic","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true}],"locations_count":2,"locations":[{"id":"doi:10.1016/j.ijmedinf.2026.106464","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.ijmedinf.2026.106464","pdf_url":null,"source":{"id":"https://openalex.org/S53142634","display_name":"International Journal of Medical Informatics","issn_l":"1386-5056","issn":["1386-5056","1872-8243"],"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":"International Journal of Medical Informatics","raw_type":"journal-article"},{"id":"pmid:42097112","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/42097112","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":"International journal of medical informatics","raw_type":null}],"best_oa_location":{"id":"doi:10.1016/j.ijmedinf.2026.106464","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.ijmedinf.2026.106464","pdf_url":null,"source":{"id":"https://openalex.org/S53142634","display_name":"International Journal of Medical Informatics","issn_l":"1386-5056","issn":["1386-5056","1872-8243"],"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":"International Journal of Medical Informatics","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1983705368","https://openalex.org/W2076118331","https://openalex.org/W2302501749","https://openalex.org/W2595557058","https://openalex.org/W2802211033","https://openalex.org/W2944006115","https://openalex.org/W2980175514","https://openalex.org/W2989619171","https://openalex.org/W3007506370","https://openalex.org/W3009800047","https://openalex.org/W3017181332","https://openalex.org/W3037356471","https://openalex.org/W3044326989","https://openalex.org/W3109410822","https://openalex.org/W3210072454","https://openalex.org/W4292869395","https://openalex.org/W4293093464","https://openalex.org/W4294811436","https://openalex.org/W4306893738","https://openalex.org/W4387267005","https://openalex.org/W4404741787","https://openalex.org/W4409287446"],"related_works":[],"abstract_inverted_index":{"BACKGROUND:":[0],"Human":[1],"Activity":[2],"Recognition":[3],"(HAR)":[4],"and":[5,19,66,74,85,95,108,112,119,134,145,154,159,180,184,203,232,260,274,284,290],"exercise":[6],"assessment":[7],"models":[8,98,148,273],"are":[9],"increasingly":[10],"used":[11],"in":[12,63,182,208,288],"healthcare":[13,292],"to":[14,31,43,71,116,175,187,216],"support":[15],"clinical":[16,262],"evaluation,":[17,256],"rehabilitation,":[18],"remote":[20],"monitoring.":[21],"However,":[22],"their":[23,155],"real-world":[24],"applicability":[25],"critically":[26],"depends":[27],"on":[28],"the":[29,59,177],"ability":[30],"generalize":[32],"across":[33,92,151],"unseen":[34],"subjects,":[35],"whose":[36],"movement":[37],"patterns":[38],"may":[39],"differ":[40],"substantially":[41],"due":[42],"inter-individual":[44],"variability.":[45],"Despite":[46],"this,":[47],"many":[48],"studies":[49],"adopt":[50],"random":[51],"noncross-subject":[52],"(NCS)":[53],"data":[54,106],"splits,":[55,218],"where":[56],"samples":[57],"from":[58],"same":[60],"individual":[61],"appear":[62],"both":[64,152],"training":[65,111,183,231],"test":[67,113,185,233],"sets,":[68],"potentially":[69],"leading":[70],"overly":[72],"optimistic":[73],"clinically":[75],"misleading":[76],"performance":[77,90,156,214],"estimates.":[78],"OBJECTIVE:":[79],"We":[80],"investigate":[81],"(i)":[82],"how":[83,105],"NCS":[84,192,213],"cross-subject":[86],"(CS)":[87],"splits":[88,193,252],"affect":[89],"estimation":[91],"machine":[93,143],"learning":[94,97,144,147],"deep":[96,146],"under":[99],"tasks":[100],"of":[101,141],"increasing":[102,200],"complexity,":[103],"(ii)":[104],"splitting":[107,241],"differences":[109,181],"between":[110,230],"sets":[114,234],"contribute":[115],"predictive":[117,237],"variance":[118,166,243],"stability.":[120,190],"METHODS:":[121],"Experiments":[122],"were":[123,149],"performed":[124],"using":[125,161],"a":[126,135,162],"large-scale":[127],"HAR":[128],"benchmark":[129],"dataset":[130,137,251],"(NTU":[131],"RGB+D":[132],"120)":[133],"rehabilitation-specific":[136],"(IntelliRehabDS).":[138],"A":[139],"total":[140],"12":[142],"trained":[150],"tasks,":[153],"was":[157,173],"estimated":[158],"compared":[160,215],"simulation-based":[163],"approach.":[164],"Predictive":[165],"decomposition,":[167],"via":[168],"Generalized":[169],"Linear":[170],"Mixed-Effects":[171],"models,":[172],"applied":[174],"link":[176],"split":[178],"strategy":[179],"instances":[186],"model":[188,196,204,255],"output":[189],"RESULTS:":[191],"consistently":[194],"overestimated":[195],"performances,":[197],"with":[198,220],"discrepancies":[199],"alongside":[201],"task":[202],"complexity.":[205],"DL":[206],"architectures,":[207],"particular,":[209],"showed":[210],"markedly":[211],"higher":[212],"CS":[217,240],"generally":[219],"statistical":[221],"significance.":[222],"Variance":[223],"decomposition":[224],"revealed":[225],"that":[226],"greater":[227],"subject":[228],"difference":[229],"often":[235],"enhances":[236],"instability,":[238],"while":[239],"reduces":[242],"by":[244],"promoting":[245],"more":[246],"generalizable":[247],"representations.":[248],"CONCLUSIONS:":[249],"Improper":[250],"can":[253],"mislead":[254],"exaggerate":[257],"generalization":[258],"capabilities,":[259],"undermine":[261],"trust.":[263],"Our":[264],"study":[265],"provides":[266],"empirical":[267],"evidence":[268],"for":[269,278],"computer":[270],"vision-based":[271],"rehabilitation":[272,289],"offers":[275],"methodological":[276],"guidance":[277],"robust":[279],"evaluation":[280],"practices,":[281],"supporting":[282],"reproducible":[283],"trustworthy":[285],"AI":[286],"deployment":[287],"broader":[291],"applications.":[293]},"counts_by_year":[],"updated_date":"2026-07-27T08:26:11.824852","created_date":"2026-05-07T00:00:00"}