{"id":"https://openalex.org/W3207138772","doi":null,"title":"Opening the Black-Box: Extracting Medical Reasoning from Machine Learning Predictions","display_name":"Opening the Black-Box: Extracting Medical Reasoning from Machine Learning Predictions","alternate_titles":[],"publication_year":2021,"publication_date":"2021-09-01","ids":{"openalex":"https://openalex.org/W3207138772","mag":"3207138772"},"language":"en","primary_location":{"id":"pmh:oai:doaj.org/article:2c74bcc05e1d4ad2b47376f75891dcf4","doi":null,"is_oa":true,"landing_page_url":"https://doaj.org/article/2c74bcc05e1d4ad2b47376f75891dcf4","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":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Applied Medical Informatics, Vol 43, Iss Suppl. S1, Pp 20-20 (2021)","raw_type":"article"},"parent_works":[],"lineage":["https://openalex.org/W3207138772"],"child_works_count":0,"type":"article","is_traditional":true,"indexed_in":["doaj"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doaj.org/article/2c74bcc05e1d4ad2b47376f75891dcf4","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5009711522","display_name":"Marius Fersigan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I208470662","display_name":"Iuliu Ha\u021bieganu University of Medicine and Pharmacy","ror":"https://ror.org/051h0cw83","country_code":"RO","type":"education","lineage":["https://openalex.org/I208470662"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Marius FERSIGAN","raw_affiliation_strings":["Iuliu Ha\u0163ieganu University of Medicine and Pharmacy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Iuliu Ha\u0163ieganu University of Medicine and Pharmacy","institution_ids":["https://openalex.org/I208470662"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051591093","display_name":"Marius M\u0103ru\u0219teri","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I3125032317","display_name":"Universitatea de Medicin\u0103, Farmacie, \u0218tiin\u021be \u0219i Tehnologie \u201eGeorge Emil Palade\u201d din T\u00e2rgu Mure\u0219","ror":"https://ror.org/03gwbzf29","country_code":"RO","type":"education","lineage":["https://openalex.org/I3125032317"]}],"countries":["RO"],"is_corresponding":false,"raw_author_name":"Marius M\u0102RU\u0218TERI","raw_affiliation_strings":["University of Medicine, Pharmacy, Science and Technology of T\u00e2rgu Mure\u015f"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Medicine, Pharmacy, Science and Technology of T\u00e2rgu Mure\u015f","institution_ids":["https://openalex.org/I3125032317"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_pulled":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":1.0,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":1.0,"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":0.0,"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"}},{"id":"https://openalex.org/T13702","display_name":"Machine Learning in Healthcare","score":0.0,"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/explainable-machine-learning","display_name":"explainable machine learning","score":0.843999981880188},{"id":"https://openalex.org/keywords/model-interpretability","display_name":"model interpretability","score":0.746999979019165},{"id":"https://openalex.org/keywords/black-box-modeling","display_name":"black-box models","score":0.7329999804496765},{"id":"https://openalex.org/keywords/clinical-decision-support","display_name":"clinical decision support","score":0.746999979019165},{"id":"https://openalex.org/keywords/automated-machine-learning-pipeline","display_name":"automated machine learning pipeline","score":0.8679999709129333},{"id":"https://openalex.org/keywords/julia-programming-language","display_name":"Julia","score":0.6669999957084656},{"id":"https://openalex.org/keywords/lime","display_name":"lime","score":0.8629999756813049},{"id":"https://openalex.org/keywords/shap","display_name":"SHAP","score":0.9990000128746033},{"id":"https://openalex.org/keywords/convolutional-neural-networks","display_name":"convolutional neural networks","score":0.7580000162124634},{"id":"https://openalex.org/keywords/random-forest","display_name":"random forest","score":0.8650000095367432},{"id":"https://openalex.org/keywords/health-data","display_name":"health data","score":0.48500001430511475}],"concepts":[{"id":"https://openalex.org/C94966114","wikidata":"https://www.wikidata.org/wiki/Q29256","display_name":"Black box","level":2,"score":0.8446769714355469},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.517087459564209},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5010507106781006},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.45156314969062805}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:doaj.org/article:2c74bcc05e1d4ad2b47376f75891dcf4","doi":null,"is_oa":true,"landing_page_url":"https://doaj.org/article/2c74bcc05e1d4ad2b47376f75891dcf4","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":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Applied Medical Informatics, Vol 43, Iss Suppl. S1, Pp 20-20 (2021)","raw_type":"article"},{"id":"pmh:oai:doaj.org/article:ce82e8e2a5254275a2dba7ad6d0509fb","doi":null,"is_oa":false,"landing_page_url":"https://doaj.org/article/ce82e8e2a5254275a2dba7ad6d0509fb","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":"Applied Medical Informatics, Vol 43, Iss Suppl. S1 (2021)","raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:doaj.org/article:2c74bcc05e1d4ad2b47376f75891dcf4","doi":null,"is_oa":true,"landing_page_url":"https://doaj.org/article/2c74bcc05e1d4ad2b47376f75891dcf4","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":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Applied Medical Informatics, Vol 43, Iss Suppl. S1, Pp 20-20 (2021)","raw_type":"article"},"sustainable_development_goals":[{"display_name":"Good health and well-being","score":0.504800021648407,"id":"https://metadata.un.org/sdg/3"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391149387","https://openalex.org/W4285105932","https://openalex.org/W7214637217","https://openalex.org/W4381929541","https://openalex.org/W3165167124","https://openalex.org/W2295559204","https://openalex.org/W4285108450","https://openalex.org/W3182295700","https://openalex.org/W7154460534","https://openalex.org/W6960278486"],"abstract_inverted_index":{"Background:":[0],"A":[1],"transparent":[2,241],"machine":[3,28,116,142,210],"learning":[4,29,117,143,211],"model":[5,52,69,81],"enables":[6,164],"the":[7,11,19,23,38,43,46,64,68,73,80,91,95,106,120,126,130,136,148,165,172,177,181,187,222,236,247,253,257],"end-user":[8],"to":[9,18,202],"investigate":[10],"decision":[12,150,242],"process":[13,151,245],"from":[14,119,129],"(clinical)":[15,149],"input":[16],"data":[17,216],"final":[20],"prediction":[21,92],"of":[22,115,167,209,224,256],"model.":[24,258],"Modern":[25],"(Deep":[26],"Learning)":[27],"models":[30,155,169,212],"are":[31,217],"highly":[32],"complex":[33],"and":[34,45,70,99,110,195,226,230,243,246],"opaque":[35],"-":[36,191],"lacking":[37],"transparency":[39,225],"required":[40],"by":[41],"both":[42],"clinician":[44],"patient.":[47],"Not":[48],"knowing":[49],"why":[50],"a":[51,54,58,84,112,198,240],"predicted":[53],"particular":[55],"outcome":[56],"in":[57,67],"clinical":[59],"context":[60],"would":[61],"likely":[62],"hinder":[63],"clinician's":[65,237],"trust":[66],"ultimately":[71],"impact":[72],"patient's":[74],"health":[75],"outcome.":[76],"Aim:":[77],"To":[78],"integrate":[79],"training":[82,166],"into":[83,197],"unique":[85],"automated":[86,141],"pipeline":[87,144,185],"that":[88,145,251],"will":[89],"output":[90],"along":[93],"with":[94,125],"relevant":[96],"explanation.":[97],"Materials":[98],"Methods:":[100],"The":[101,123,206],"workflow":[102],"was":[103,133],"build":[104],"using":[105,135],"Julia":[107],"programming":[108],"language":[109],"employing":[111],"wide":[113],"range":[114],"packages":[118,128],"MLJ":[121],"universe.":[122],"interoperability":[124],"explainer":[127,200],"Python":[131],"ecosystem":[132],"enabled":[134],"PyCall.jl":[137],"package.":[138],"Results:":[139],"An":[140],"a)":[146],"exposes":[147],"inside":[152],"various":[153],"black-box":[154],"(convolutional":[156],"neural":[157],"networks,":[158],"random":[159],"forest,":[160],"boosted":[161],"models),":[162],"b)":[163],"explainable":[168],"without":[170],"compromising":[171],"performance":[173,207,249],"metrics,":[174],"c)":[175],"validates":[176],"resulted":[178],"method":[179],"against":[180],"medical":[182],"domain-knowledge.":[183],"Our":[184],"combines":[186],"known":[188],"non-healthcare-specific":[189],"explainers":[190],"LIME,":[192],"SHAP,":[193],"ShapML,":[194],"InterpretML":[196],"global":[199],"tailored":[201],"healthcare-specific":[203],"data.":[204],"Conclusions:":[205],"metrics":[208,250],"trained":[213],"on":[214,221],"healthcare":[215],"not":[218],"necessarily":[219],"sacrificed":[220],"altar":[223],"interpretability.":[227],"Using":[228],"model-agnostic":[229],"model-specific":[231],"explainers,":[232],"we":[233],"can":[234],"satisfy":[235],"need":[238],"for":[239],"diagnosis":[244],"superior":[248],"ensure":[252],"predictive":[254],"power":[255]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W3207138772","counts_by_year":[],"updated_date":"2026-11-11T12:16:26.678958","created_date":"2021-10-25T00:00:00"}
