{"id":"https://openalex.org/W7224091719","doi":"https://doi.org/10.3389/frai.2026.1983619","title":"From task-specific networks to foundation models in diabetes care: a critical review of opportunities, evaluation gaps, and safety frontiers","display_name":"From task-specific networks to foundation models in diabetes care: a critical review of opportunities, evaluation gaps, and safety frontiers","publication_year":2026,"publication_date":"2026-10-28","ids":{"openalex":"https://openalex.org/W7224091719","doi":"https://doi.org/10.3389/frai.2026.1983619"},"language":"en","primary_location":{"id":"doi:10.3389/frai.2026.1983619","doi":"https://doi.org/10.3389/frai.2026.1983619","is_oa":true,"landing_page_url":"https://doi.org/10.3389/frai.2026.1983619","pdf_url":"https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2026.1983619/pdf","source":{"id":"https://openalex.org/S4210197006","display_name":"Frontiers in Artificial Intelligence","issn_l":"2624-8212","issn":["2624-8212"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["anvur-class-a","anvur-scientific","cwts-core","doaj","jufo-1","ki-jl-1","norway-1","poland-20","russia-white-list-1","vabb-shw"],"host_organization":"https://openalex.org/P4310320527","host_organization_name":"Frontiers Media","host_organization_lineage":["https://openalex.org/P4310320527"],"host_organization_lineage_names":["Frontiers Media"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence","raw_type":"journal-article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2026.1983619/pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5144482560","display_name":"M. Nanda Kishore","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I876193797","display_name":"Vellore Institute of Technology University","ror":"https://ror.org/00qzypv28","country_code":"IN","type":"education","lineage":["https://openalex.org/I876193797"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"M. Nanda Kishore","raw_affiliation_strings":["School of Computer Science Engineering and Information Systems, Vellore Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","institution_ids":["https://openalex.org/I876193797"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5034971373","display_name":"C. Navaneethan","orcid":"https://orcid.org/0000-0001-5079-058X","observed_orcids":["https://orcid.org/0000-0001-5079-058X"]},"institutions":[{"id":"https://openalex.org/I876193797","display_name":"Vellore Institute of Technology University","ror":"https://ror.org/00qzypv28","country_code":"IN","type":"education","lineage":["https://openalex.org/I876193797"]}],"countries":["IN"],"is_corresponding":true,"raw_author_name":"C. Navaneethan","raw_affiliation_strings":["School of Computer Science Engineering and Information Systems, Vellore Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","institution_ids":["https://openalex.org/I876193797"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5034971373"],"corresponding_institution_ids":["https://openalex.org/I876193797"],"apc_list":{"value":1950,"currency":"CHF","value_usd":2353},"apc_paid":{"value":1950,"currency":"CHF","value_usd":2353},"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.74320114,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"9","issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.6942999958992004,"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"}},"topics":[{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.6942999958992004,"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"}},{"id":"https://openalex.org/T13702","display_name":"Machine Learning in Healthcare","score":0.260699987411499,"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/T11396","display_name":"Artificial Intelligence in Healthcare","score":0.019500000402331352,"subfield":{"id":"https://openalex.org/subfields/3605","display_name":"Health Information Management"},"field":{"id":"https://openalex.org/fields/36","display_name":"Health Professions"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/foundation-models","display_name":"foundation models","score":0.843999981880188},{"id":"https://openalex.org/keywords/diabetes-management","display_name":"diabetes management","score":0.9409999847412109},{"id":"https://openalex.org/keywords/diabetic-retinopathy-grading","display_name":"diabetic retinopathy grading","score":0.8529999852180481},{"id":"https://openalex.org/keywords/clinical-large-language-models","display_name":"clinical large language models","score":0.9639999866485596},{"id":"https://openalex.org/keywords/generative-models","display_name":"generative models","score":0.4230000078678131},{"id":"https://openalex.org/keywords/electronic-health-records","display_name":"electronic health records","score":0.6110000014305115},{"id":"https://openalex.org/keywords/model-evaluation","display_name":"model evaluation","score":0.4519999921321869},{"id":"https://openalex.org/keywords/benchmark-contamination","display_name":"benchmark contamination","score":0.8939999938011169},{"id":"https://openalex.org/keywords/distribution-shift","display_name":"distribution shift","score":0.734000027179718},{"id":"https://openalex.org/keywords/calibration-error","display_name":"calibration error","score":0.7720000147819519},{"id":"https://openalex.org/keywords/hallucinations","display_name":"hallucinations","score":0.597000002861023},{"id":"https://openalex.org/keywords/automation-bias","display_name":"automation bias","score":0.7419999837875366},{"id":"https://openalex.org/keywords/privacy-risks","display_name":"privacy risks","score":0.609000027179718},{"id":"https://openalex.org/keywords/external-validation","display_name":"external validation","score":0.45500001311302185},{"id":"https://openalex.org/keywords/treatment-effect-estimation","display_name":"treatment effect estimation","score":0.4490000009536743}],"concepts":[{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.5270000100135803},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5110999941825867},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.46560001373291016},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.4440000057220459},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.42410001158714294},{"id":"https://openalex.org/C2779134260","wikidata":"https://www.wikidata.org/wiki/Q12136","display_name":"Disease","level":2,"score":0.3977000117301941},{"id":"https://openalex.org/C2778313320","wikidata":"https://www.wikidata.org/wiki/Q550455","display_name":"Retinopathy","level":3,"score":0.37400001287460327},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3716999888420105},{"id":"https://openalex.org/C555293320","wikidata":"https://www.wikidata.org/wiki/Q12206","display_name":"Diabetes mellitus","level":2,"score":0.37119999527931213},{"id":"https://openalex.org/C177713679","wikidata":"https://www.wikidata.org/wiki/Q679690","display_name":"Intensive care medicine","level":1,"score":0.36719998717308044},{"id":"https://openalex.org/C2986379492","wikidata":"https://www.wikidata.org/wiki/Q1638492","display_name":"Continuous glucose monitoring","level":4,"score":0.3562000095844269},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3368000090122223},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.3125},{"id":"https://openalex.org/C2779829184","wikidata":"https://www.wikidata.org/wiki/Q631361","display_name":"Diabetic retinopathy","level":3,"score":0.30660000443458557},{"id":"https://openalex.org/C2779473830","wikidata":"https://www.wikidata.org/wiki/Q1540899","display_name":"MEDLINE","level":2,"score":0.304500013589859},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.2937000095844269},{"id":"https://openalex.org/C189708586","wikidata":"https://www.wikidata.org/wiki/Q1504425","display_name":"Systematic review","level":3,"score":0.2863999903202057},{"id":"https://openalex.org/C12174686","wikidata":"https://www.wikidata.org/wiki/Q1058438","display_name":"Risk assessment","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C44210515","wikidata":"https://www.wikidata.org/wiki/Q16968978","display_name":"Bespoke","level":2,"score":0.2757999897003174},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.26919999718666077},{"id":"https://openalex.org/C160735492","wikidata":"https://www.wikidata.org/wiki/Q31207","display_name":"Health care","level":2,"score":0.25369998812675476},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.25279998779296875},{"id":"https://openalex.org/C2779100554","wikidata":"https://www.wikidata.org/wiki/Q25303769","display_name":"Microperimetry","level":3,"score":0.25}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.3389/frai.2026.1983619","doi":"https://doi.org/10.3389/frai.2026.1983619","is_oa":true,"landing_page_url":"https://doi.org/10.3389/frai.2026.1983619","pdf_url":"https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2026.1983619/pdf","source":{"id":"https://openalex.org/S4210197006","display_name":"Frontiers in Artificial Intelligence","issn_l":"2624-8212","issn":["2624-8212"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["anvur-class-a","anvur-scientific","cwts-core","doaj","jufo-1","ki-jl-1","norway-1","poland-20","russia-white-list-1","vabb-shw"],"host_organization":"https://openalex.org/P4310320527","host_organization_name":"Frontiers Media","host_organization_lineage":["https://openalex.org/P4310320527"],"host_organization_lineage_names":["Frontiers Media"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:6a2278041607495a8e26a5bb9a46bed4","doi":null,"is_oa":false,"landing_page_url":"https://doaj.org/article/6a2278041607495a8e26a5bb9a46bed4","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"Frontiers in Artificial Intelligence, Vol 9 (2026)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.3389/frai.2026.1983619","doi":"https://doi.org/10.3389/frai.2026.1983619","is_oa":true,"landing_page_url":"https://doi.org/10.3389/frai.2026.1983619","pdf_url":"https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2026.1983619/pdf","source":{"id":"https://openalex.org/S4210197006","display_name":"Frontiers in Artificial Intelligence","issn_l":"2624-8212","issn":["2624-8212"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["anvur-class-a","anvur-scientific","cwts-core","doaj","jufo-1","ki-jl-1","norway-1","poland-20","russia-white-list-1","vabb-shw"],"host_organization":"https://openalex.org/P4310320527","host_organization_name":"Frontiers Media","host_organization_lineage":["https://openalex.org/P4310320527"],"host_organization_lineage_names":["Frontiers Media"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/3","display_name":"Good health and well-being","score":0.98580002784729}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224091719.pdf","grobid_xml":"https://content.openalex.org/works/W7224091719.grobid-xml"},"referenced_works_count":60,"referenced_works":["https://openalex.org/W1969496006","https://openalex.org/W1986495510","https://openalex.org/W2025923126","https://openalex.org/W2079283960","https://openalex.org/W2106311978","https://openalex.org/W2107638293","https://openalex.org/W2111352982","https://openalex.org/W2115531174","https://openalex.org/W2117191092","https://openalex.org/W2129995532","https://openalex.org/W2396881363","https://openalex.org/W2466367758","https://openalex.org/W2557738935","https://openalex.org/W2598442119","https://openalex.org/W2625625371","https://openalex.org/W2626354203","https://openalex.org/W2727650337","https://openalex.org/W2751687090","https://openalex.org/W2765304416","https://openalex.org/W2769131622","https://openalex.org/W2769264260","https://openalex.org/W2772246530","https://openalex.org/W2790501791","https://openalex.org/W2888171911","https://openalex.org/W2888424632","https://openalex.org/W2902802452","https://openalex.org/W2907554860","https://openalex.org/W2908201961","https://openalex.org/W2922577842","https://openalex.org/W2942586440","https://openalex.org/W2945976633","https://openalex.org/W2948261890","https://openalex.org/W2981869278","https://openalex.org/W3004612364","https://openalex.org/W3012501605","https://openalex.org/W3017637887","https://openalex.org/W3020619771","https://openalex.org/W3106920072","https://openalex.org/W3158522809","https://openalex.org/W3160137267","https://openalex.org/W3177617320","https://openalex.org/W3200840849","https://openalex.org/W3209901185","https://openalex.org/W4233026002","https://openalex.org/W4280572991","https://openalex.org/W4306783667","https://openalex.org/W4309674289","https://openalex.org/W4365143687","https://openalex.org/W4367310920","https://openalex.org/W4384071683","https://openalex.org/W4384561707","https://openalex.org/W4386697749","https://openalex.org/W4387001850","https://openalex.org/W4394857110","https://openalex.org/W4396553357","https://openalex.org/W4406152279","https://openalex.org/W4416404524","https://openalex.org/W4417121383","https://openalex.org/W7133211372","https://openalex.org/W7133228402"],"related_works":["https://openalex.org/W7154992295","https://openalex.org/W7154614828","https://openalex.org/W7223096838","https://openalex.org/W7162028048","https://openalex.org/W7135406865","https://openalex.org/W4408315845","https://openalex.org/W4413949768","https://openalex.org/W4405621434","https://openalex.org/W7214791563","https://openalex.org/W4417315754"],"abstract_inverted_index":{"Artificial":[0],"intelligence":[1],"in":[2,202,248],"diabetes":[3,166,250],"care":[4],"has":[5],"historically":[6],"relied":[7],"on":[8,34,59,71,113],"task-specific":[9,69,150],"supervised":[10],"learning,":[11],"such":[12,55,105],"as":[13,56,106],"bespoke":[14],"networks":[15,70],"for":[16,94,212],"retinopathy":[17,72],"grading,":[18],"kidney":[19],"disease":[20,76],"prediction,":[21,77],"and":[22,40,74,108,117,123,156,163,176,185,209,214,226,252,265,293],"glucose":[23,96],"forecasting.":[24],"The":[25],"emergence":[26],"of":[27,136,195,207,240,244],"foundation":[28,157,216,246],"models":[29,32,54,86,103,119,158],"(FMs:":[30],"large":[31,101],"pretrained":[33,58],"broad":[35],"data,":[36],"typically":[37],"with":[38,299],"self-supervision,":[39],"adapted":[41],"to":[42,67,223],"many":[43],"downstream":[44],"tasks)":[45],"is":[46,277],"displacing":[47],"this":[48,137],"paradigm":[49],"across":[50,179],"diabetes-relevant":[51],"modalities:":[52],"ophthalmic":[53],"RETFound,":[57],"1.6":[60],"million":[61],"retinal":[62,180,213],"photographs,":[63],"have":[64],"been":[65],"reported":[66],"outperform":[68],"grading":[73],"systemic":[75],"especially":[78],"when":[79],"labeled":[80],"data":[81],"are":[82,87,236],"scarce;":[83],"transformer-based":[84],"time-series":[85],"replacing":[88],"long":[89],"short-term":[90],"memory":[91],"(LSTM)":[92],"pipelines":[93],"continuous":[95],"monitoring":[97],"(CGM)":[98],"forecasting;":[99],"clinical":[100],"language":[102],"(LLMs)":[104],"Med-PaLM":[107],"GPT-4":[109],"attain":[110],"expert-level":[111],"scores":[112,147],"medical":[114,197],"question-answering":[115],"benchmarks;":[116],"generative":[118],"supply":[120],"synthetic":[121],"CGM":[122,215],"electronic":[124],"health":[125],"record":[126],"(EHR)":[127],"data.":[128],"This":[129],"review":[130],"provides":[131],"a":[132,142,199,245,249],"critical,":[133],"modality-organized":[134],"synthesis":[135],"transition.":[138],"We":[139,270],"(i)":[140],"formalize":[141],"three-era":[143],"taxonomy:":[144],"rule-based":[145],"risk":[146],"(era":[148,153,159],"I),":[149],"deep":[151],"learning":[152,281],"II,":[154],"2016\u20132020),":[155],"III,":[160],"from":[161],"2021),":[162],"map":[164],"the":[165,203,228,238,273],"literature":[167,204,230],"onto":[168],"it;":[169],"(ii)":[170],"survey":[171],"foundation-model":[172,297],"architectures,":[173],"pretraining":[174],"objectives,":[175],"adaptation":[177],"strategies":[178],"imaging,":[181],"CGM,":[182],"longitudinal":[183],"EHR,":[184],"multimodal":[186],"data;":[187],"(iii)":[188],"expose":[189],"an":[190],"evaluation":[191,243],"crisis:":[192],"benchmark":[193],"contamination":[194],"LLM":[196],"evaluations,":[198],"near-universal":[200],"absence":[201,239],"we":[205],"reviewed":[206,229],"external":[208],"prospective":[210,291],"validation":[211],"models,":[217],"miscalibration":[218],"under":[219],"dataset":[220],"shift,":[221],"and,":[222],"our":[224],"knowledge":[225],"within":[227],"(January":[231],"2016\u2013December":[232],"2025;":[233],"negative":[234],"findings":[235],"time-sensitive),":[237],"any":[241],"randomized":[242],"model":[247],"workflow;":[251],"(iv)":[253],"analyze":[254],"safety":[255],"risks:":[256],"hallucinated":[257],"insulin-dosing":[258],"advice,":[259],"automation":[260],"bias,":[261],"inherited":[262],"algorithmic":[263],"inequity,":[264],"privacy":[266],"exposure":[267],"during":[268],"adaptation.":[269],"conclude":[271],"that":[272],"decisive":[274],"open":[275],"problem":[276],"no":[278],"longer":[279],"representation":[280],"but":[282],"grounded,":[283],"calibrated,":[284],"causally":[285],"informed":[286],"deployment,":[287],"requiring":[288],"contamination-resistant":[289],"evaluation,":[290],"trials,":[292],"hybrid":[294],"architectures":[295],"coupling":[296],"representations":[298],"treatment-effect":[300],"estimation.":[301]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2026-10-29T00:00:00"}