{"id":"https://openalex.org/W7131621125","doi":"https://doi.org/10.48550/arxiv.2602.21374","title":"Small Language Models for Privacy-Preserving Clinical Information Extraction in Low-Resource Languages","display_name":"Small Language Models for Privacy-Preserving Clinical Information Extraction in Low-Resource Languages","publication_year":2026,"publication_date":"2026-02-24","ids":{"openalex":"https://openalex.org/W7131621125","doi":"https://doi.org/10.48550/arxiv.2602.21374"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.21374","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"preprint","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/A5107783916","display_name":"Mohammadreza Ghaffarzadeh-Esfahani","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ghaffarzadeh-Esfahani, Mohammadreza","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126286306","display_name":"Nahid Yousefian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yousefian, Nahid","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055550355","display_name":"Ebrahim HeidariFarsani","orcid":"https://orcid.org/0000-0001-5399-9266"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heidari-Farsani, Ebrahim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126883028","display_name":"Ali Akbar Omidvarian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Omidvarian, Ali Akbar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126908154","display_name":"Sepehr Ghahraei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ghahraei, Sepehr","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126864633","display_name":"Atena Farangi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Farangi, Atena","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5126877254","display_name":"AmirBahador Boroumand","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Boroumand, AmirBahador","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/T10028","display_name":"Topic Modeling","score":0.6039999723434448,"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/T10028","display_name":"Topic Modeling","score":0.6039999723434448,"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/T11710","display_name":"Biomedical Text Mining and Ontologies","score":0.12120000272989273,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"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/T13702","display_name":"Machine Learning in Healthcare","score":0.050999999046325684,"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/pipeline","display_name":"Pipeline (software)","score":0.616599977016449},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.48399999737739563},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.47870001196861267},{"id":"https://openalex.org/keywords/information-extraction","display_name":"Information extraction","score":0.4632999897003174},{"id":"https://openalex.org/keywords/annotation","display_name":"Annotation","score":0.4523000121116638},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.39809998869895935},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.39160001277923584},{"id":"https://openalex.org/keywords/blueprint","display_name":"Blueprint","score":0.3716999888420105}],"concepts":[{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.7121999859809875},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6766999959945679},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6468999981880188},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.616599977016449},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.48399999737739563},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.47870001196861267},{"id":"https://openalex.org/C195807954","wikidata":"https://www.wikidata.org/wiki/Q1662562","display_name":"Information extraction","level":2,"score":0.4632999897003174},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.4523000121116638},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.39809998869895935},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.39160001277923584},{"id":"https://openalex.org/C155911762","wikidata":"https://www.wikidata.org/wiki/Q422321","display_name":"Blueprint","level":2,"score":0.3716999888420105},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3626999855041504},{"id":"https://openalex.org/C153604712","wikidata":"https://www.wikidata.org/wiki/Q7310755","display_name":"Relationship extraction","level":3,"score":0.3352000117301941},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.31790000200271606},{"id":"https://openalex.org/C117220453","wikidata":"https://www.wikidata.org/wiki/Q5172842","display_name":"Correlation","level":2,"score":0.3138999938964844},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.31220000982284546},{"id":"https://openalex.org/C69505689","wikidata":"https://www.wikidata.org/wiki/Q455338","display_name":"Unified Medical Language System","level":2,"score":0.3109000027179718},{"id":"https://openalex.org/C148524875","wikidata":"https://www.wikidata.org/wiki/Q6975395","display_name":"F1 score","level":2,"score":0.28220000863075256},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.28110000491142273},{"id":"https://openalex.org/C2777466982","wikidata":"https://www.wikidata.org/wiki/Q5227287","display_name":"Data extraction","level":3,"score":0.2770000100135803},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.27410000562667847},{"id":"https://openalex.org/C111696304","wikidata":"https://www.wikidata.org/wiki/Q2303697","display_name":"Sorting","level":2,"score":0.26460000872612},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2556999921798706},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2554999887943268},{"id":"https://openalex.org/C160735492","wikidata":"https://www.wikidata.org/wiki/Q31207","display_name":"Health care","level":2,"score":0.2538999915122986},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.21374","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.21374","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.21374","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":"pmh:doi:10.48550/arxiv.2602.21374","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[{"score":0.48844242095947266,"display_name":"Industry, innovation and infrastructure","id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Extracting":[0],"clinical":[1,52,190],"information":[2],"from":[3,54],"medical":[4],"transcripts":[5,58,130],"in":[6,13,117,188],"low-resource":[7],"languages":[8],"remains":[9],"a":[10,22,28,61,68,180],"significant":[11],"challenge":[12],"healthcare":[14,214],"natural":[15],"language":[16,36,210],"processing":[17],"(NLP).":[18],"This":[19],"study":[20],"evaluates":[21],"two-step":[23],"pipeline":[24],"combining":[25],"Aya-expanse-8B":[26,125],"as":[27],"Persian-to-English":[29],"translation":[30],"model":[31,206],"with":[32,193],"five":[33],"open-source":[34,186],"small":[35],"models":[37,74,110],"(SLMs)":[38],"--":[39,46],"Qwen2.5-7B-Instruct,":[40],"Llama-3.1-8B-Instruct,":[41],"Llama-3.2-3B-Instruct,":[42],"Qwen2.5-1.5B-Instruct,":[43],"and":[44,85,119,138,153,171,196,199,208],"Gemma-3-1B-it":[45,104],"for":[47,89,184,212],"binary":[48],"extraction":[49,159],"of":[50,124,149,160,203],"13":[51],"features":[53,174],"1,221":[55],"anonymized":[56],"Persian":[57,129],"collected":[59],"at":[60,146],"cancer":[62],"palliative":[63],"care":[64],"call":[65],"center.":[66],"Using":[67],"few-shot":[69],"prompting":[70],"strategy":[71,211],"without":[72],"fine-tuning,":[73],"were":[75],"assessed":[76],"on":[77],"macro-averaged":[78],"F1-score,":[79],"Matthews":[80],"Correlation":[81],"Coefficient":[82],"(MCC),":[83],"sensitivity,":[84,134],"specificity":[86,152],"to":[87,131,142],"account":[88],"class":[90,143],"imbalance.":[91],"Qwen2.5-7B-Instruct":[92],"achieved":[93],"the":[94,106,147,201],"highest":[95],"overall":[96],"performance":[97],"(median":[98],"macro-F1:":[99],"0.899;":[100],"MCC:":[101],"0.797),":[102],"while":[103],"showed":[105,157],"weakest":[107],"results.":[108],"Larger":[109],"(7B--8B":[111],"parameters)":[112],"consistently":[113],"outperformed":[114],"smaller":[115],"counterparts":[116],"sensitivity":[118],"MCC.":[120],"A":[121],"bilingual":[122],"analysis":[123],"revealed":[126],"that":[127],"translating":[128],"English":[132],"improved":[133],"reduced":[135],"missing":[136],"outputs,":[137],"boosted":[139],"metrics":[140],"robust":[141],"imbalance,":[144],"though":[145],"cost":[148],"slightly":[150],"lower":[151],"precision.":[154],"Feature-level":[155],"results":[156],"reliable":[158],"physiological":[161],"symptoms":[162],"across":[163],"most":[164],"models,":[165],"whereas":[166],"psychological":[167],"complaints,":[168],"administrative":[169],"requests,":[170],"complex":[172],"somatic":[173],"remained":[175],"challenging.":[176],"These":[177],"findings":[178],"establish":[179],"practical,":[181],"privacy-preserving":[182],"blueprint":[183],"deploying":[185],"SLMs":[187],"multilingual":[189],"NLP":[191],"settings":[192],"limited":[194],"infrastructure":[195],"annotation":[197],"resources,":[198],"highlight":[200],"importance":[202],"jointly":[204],"optimizing":[205],"scale":[207],"input":[209],"sensitive":[213],"applications.":[215]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-02-27T00:00:00"}