Supervised machine learning-based bias risk of prognostic models for total knee or hip arthroplasty patients: A systematic review

As various machine learning (ML) algorithms have become more popular in orthopedic surgery, the research quality of these models requires further evaluation, and the methodological quality of the models still needs to be clarified. This study aimed to comprehensively analyze and evaluate the potenti...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Medicine (Baltimore) Ročník 104; číslo 42; s. e45230
Hlavní autori:	Zhang, Hongxia, Jiang, Lina, Zheng, Jiang, Li, Chuanbo
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	United States 17.10.2025
Predmet:	Arthroplasty, Replacement, Hip - statistics & numerical data Arthroplasty, Replacement, Knee - statistics & numerical data Bias Humans Prediction Algorithms Predictive Learning Models - standards Prognosis Risk Assessment - methods Risk Assessment - standards Software Validation Supervised Machine Learning - standards supervised machine learning risk bias systematic review prognostic model arthroplasty
ISSN:	1536-5964, 1536-5964
On-line prístup:	Zistit podrobnosti o prístupe
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	As various machine learning (ML) algorithms have become more popular in orthopedic surgery, the research quality of these models requires further evaluation, and the methodological quality of the models still needs to be clarified. This study aimed to comprehensively analyze and evaluate the potential bias and applicability of existing studies of supervised ML-driven prognostic risk prediction models focusing on total knee arthroplasty/total hip arthroplasty individuals. The China National Knowledge Infrastructure, Wanfang, PubMed, Web of Science, Cochrane Library, and Embase databases were searched from inception to January 20, 2024. The following data were extracted from the selected studies: author, year, joints, sample size, data source, study design, ML methods, ML performance, and primary outcome. The PROBAST checklist was applied to evaluate the risk of bias and applicability in prediction model studies. The protocol is registered in the PROSPERO database (CRD42024501747). A total of 2909 indexed records were obtained and 32 studies were included, 30 of which had models for internal validation only, 1 study for development and validation, and 1 study for external validation only. The PROBAST evaluation results showed that 1 externally validated model and 29/31 (93%) development models were rated as having a high risk of bias. There was a high risk of bias in the participant and analysis domains. Almost all supervised ML models have the potential for a high bias risk. Factors contributing to a high bias risk include inadequate sample size, missing data during recruitment, model overfitting, and limited external validation. Adhering to strict standards and implementing comprehensive improvements when constructing prognosis models using supervised ML is crucial.
AbstractList	As various machine learning (ML) algorithms have become more popular in orthopedic surgery, the research quality of these models requires further evaluation, and the methodological quality of the models still needs to be clarified. This study aimed to comprehensively analyze and evaluate the potential bias and applicability of existing studies of supervised ML-driven prognostic risk prediction models focusing on total knee arthroplasty/total hip arthroplasty individuals.BACKGROUNDAs various machine learning (ML) algorithms have become more popular in orthopedic surgery, the research quality of these models requires further evaluation, and the methodological quality of the models still needs to be clarified. This study aimed to comprehensively analyze and evaluate the potential bias and applicability of existing studies of supervised ML-driven prognostic risk prediction models focusing on total knee arthroplasty/total hip arthroplasty individuals.The China National Knowledge Infrastructure, Wanfang, PubMed, Web of Science, Cochrane Library, and Embase databases were searched from inception to January 20, 2024. The following data were extracted from the selected studies: author, year, joints, sample size, data source, study design, ML methods, ML performance, and primary outcome. The PROBAST checklist was applied to evaluate the risk of bias and applicability in prediction model studies. The protocol is registered in the PROSPERO database (CRD42024501747).METHODSThe China National Knowledge Infrastructure, Wanfang, PubMed, Web of Science, Cochrane Library, and Embase databases were searched from inception to January 20, 2024. The following data were extracted from the selected studies: author, year, joints, sample size, data source, study design, ML methods, ML performance, and primary outcome. The PROBAST checklist was applied to evaluate the risk of bias and applicability in prediction model studies. The protocol is registered in the PROSPERO database (CRD42024501747).A total of 2909 indexed records were obtained and 32 studies were included, 30 of which had models for internal validation only, 1 study for development and validation, and 1 study for external validation only. The PROBAST evaluation results showed that 1 externally validated model and 29/31 (93%) development models were rated as having a high risk of bias. There was a high risk of bias in the participant and analysis domains.RESULTSA total of 2909 indexed records were obtained and 32 studies were included, 30 of which had models for internal validation only, 1 study for development and validation, and 1 study for external validation only. The PROBAST evaluation results showed that 1 externally validated model and 29/31 (93%) development models were rated as having a high risk of bias. There was a high risk of bias in the participant and analysis domains.Almost all supervised ML models have the potential for a high bias risk. Factors contributing to a high bias risk include inadequate sample size, missing data during recruitment, model overfitting, and limited external validation. Adhering to strict standards and implementing comprehensive improvements when constructing prognosis models using supervised ML is crucial.CONCLUSIONAlmost all supervised ML models have the potential for a high bias risk. Factors contributing to a high bias risk include inadequate sample size, missing data during recruitment, model overfitting, and limited external validation. Adhering to strict standards and implementing comprehensive improvements when constructing prognosis models using supervised ML is crucial. As various machine learning (ML) algorithms have become more popular in orthopedic surgery, the research quality of these models requires further evaluation, and the methodological quality of the models still needs to be clarified. This study aimed to comprehensively analyze and evaluate the potential bias and applicability of existing studies of supervised ML-driven prognostic risk prediction models focusing on total knee arthroplasty/total hip arthroplasty individuals. The China National Knowledge Infrastructure, Wanfang, PubMed, Web of Science, Cochrane Library, and Embase databases were searched from inception to January 20, 2024. The following data were extracted from the selected studies: author, year, joints, sample size, data source, study design, ML methods, ML performance, and primary outcome. The PROBAST checklist was applied to evaluate the risk of bias and applicability in prediction model studies. The protocol is registered in the PROSPERO database (CRD42024501747). A total of 2909 indexed records were obtained and 32 studies were included, 30 of which had models for internal validation only, 1 study for development and validation, and 1 study for external validation only. The PROBAST evaluation results showed that 1 externally validated model and 29/31 (93%) development models were rated as having a high risk of bias. There was a high risk of bias in the participant and analysis domains. Almost all supervised ML models have the potential for a high bias risk. Factors contributing to a high bias risk include inadequate sample size, missing data during recruitment, model overfitting, and limited external validation. Adhering to strict standards and implementing comprehensive improvements when constructing prognosis models using supervised ML is crucial.
Author	Li, Chuanbo Zhang, Hongxia Jiang, Lina Zheng, Jiang
Author_xml	– sequence: 1 givenname: Hongxia surname: Zhang fullname: Zhang, Hongxia organization: Department of Orthopedics, Banan Hospital Affiliated to Chongqing Medical University, Chongqing, China – sequence: 2 givenname: Lina surname: Jiang fullname: Jiang, Lina organization: Department of Orthopedics, Banan Hospital Affiliated to Chongqing Medical University, Chongqing, China – sequence: 3 givenname: Jiang surname: Zheng fullname: Zheng, Jiang organization: Leshan City Vocational and Technical College Nursing Department, Leshan, China – sequence: 4 givenname: Chuanbo surname: Li fullname: Li, Chuanbo organization: Department of Orthopedics, Banan Hospital Affiliated to Chongqing Medical University, Chongqing, China
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/41189192$$D View this record in MEDLINE/PubMed
BookMark	eNpNkE1LAzEQhoNU7If-AkFy9LI1yX57K_UTKh7Uc0k2kzZ2N1mTbKVH_7lbrOhcZt7h4YGZMRoYawChc0qmlJT51dPNlPxVkrKYHKERTeMsSsssGfybh2js_TshNM5ZcoKGCaVFSUs2Ql8vXQtuqz1I3PBqrQ3gGrgz2qwiwfdrobnHTvsNtgq3zq6M9UFXuLESao-VdTjYwGu8MQC4T2vdYu7C2tm25j7scMuDBhP8NZ5hv_MBGr4XONhq-DxFx4rXHs4OfYLe7m5f5w_R4vn-cT5bRFWcEhIJRXkqilgyUiU8LSnNVJnELCsYE7JgcS4VBSEFlyors5QoruJC5SCBKtYDE3T54-1P-OjAh2WjfQV1zQ3Yzi_3qiKhLM979OKAdqIBuWydbrjbLX_fxr4BJyN0YQ
ContentType	Journal Article
Copyright	Copyright © 2025 the Author(s). Published by Wolters Kluwer Health, Inc.
Copyright_xml	– notice: Copyright © 2025 the Author(s). Published by Wolters Kluwer Health, Inc.
DBID	CGR CUY CVF ECM EIF NPM 7X8
DOI	10.1097/MD.0000000000045230
DatabaseName	Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic
DatabaseTitle	MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic
DatabaseTitleList	MEDLINE - Academic MEDLINE
Database_xml	– sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	no_fulltext_linktorsrc
Discipline	Medicine
EISSN	1536-5964
ExternalDocumentID	41189192
Genre	Systematic Review Journal Article
GroupedDBID	--- .-D .XZ .Z2 01R 0R~ 354 40H 4Q1 4Q2 4Q3 5GY 5RE 5VS 71W 77Y 7O~ AAAAV AAGIX AAHPQ AAIQE AAMOA AAQKA AARTV AASCR AAWTL AAXQO AAYEP ABASU ABBUW ABCQX ABDIG ABFRF ABOCM ABVCZ ABXVJ ABZAD ABZZY ACBKD ACDDN ACEWG ACGFO ACGFS ACILI ACLDA ACWDW ACWRI ACXJB ACXNZ ACZKN ADGGA ADHPY ADKSD ADNKB ADPDF ADSXY AE6 AEFWE AENEX AFBFQ AFDTB AGOPY AHOMT AHQNM AHVBC AIJEX AINUH AJCLO AJIOK AJNWD AJNYG AJZMW AKCTQ AKULP ALKUP ALMA_UNASSIGNED_HOLDINGS ALMTX AMJPA AMKUR AMNEI AOHHW AOQMC BQLVK CGR CS3 CUY CVF DIWNM DU5 E.X EBS ECM EEVPB EIF ERAAH EX3 F2K F2L F2M F2N F5P FCALG FD6 FIJ FL- GNXGY GQDEL GROUPED_DOAJ H0~ HLJTE HYE HZ~ H~9 IKREB IKYAY IN~ IPNFZ JK3 JK8 K8S KD2 KMI KQ8 L-C N9A NPM N~7 N~B O9- OAG OAH OB2 OHH OK1 OL1 OLB OLG OLH OLU OLV OLY OLZ OPUJH OUVQU OVD OVDNE OVEED OVIDH OVLEI OWV OWW OWZ OXXIT P2P RIG RLZ RPM RXW S4R S4S TEORI TSPGW UNMZH V2I VVN W3M WOQ WOW X3V X3W XYM YFH YOC ZFV ZY1 7X8 TAF
ID	FETCH-LOGICAL-c3500-bf1a5b83d20c4a59116f94326822bd8237df1ebdbadf69650faf38f7ede1f2d82
IEDL.DBID	7X8
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001608941400038&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1536-5964
IngestDate	Wed Nov 05 17:16:19 EST 2025 Tue Nov 25 01:40:45 EST 2025
IsDoiOpenAccess	false
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	42
Keywords	supervised machine learning risk bias systematic review prognostic model arthroplasty
Language	English
License	Copyright © 2025 the Author(s). Published by Wolters Kluwer Health, Inc.
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c3500-bf1a5b83d20c4a59116f94326822bd8237df1ebdbadf69650faf38f7ede1f2d82
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 ObjectType-Undefined-3
OpenAccessLink	https://ovidsp.ovid.com/ovidweb.cgi?T=JS&NEWS=n&CSC=Y&PAGE=fulltext&D=ovft&DO=10.1097/MD.0000000000045230
PMID	41189192
PQID	3268841277
PQPubID	23479
ParticipantIDs	proquest_miscellaneous_3268841277 pubmed_primary_41189192
PublicationCentury	2000
PublicationDate	2025-Oct-17 20251017
PublicationDateYYYYMMDD	2025-10-17
PublicationDate_xml	– month: 10 year: 2025 text: 2025-Oct-17 day: 17
PublicationDecade	2020
PublicationPlace	United States
PublicationPlace_xml	– name: United States
PublicationTitle	Medicine (Baltimore)
PublicationTitleAlternate	Medicine (Baltimore)
PublicationYear	2025
SSID	ssj0013724
Score	2.4729133
SecondaryResourceType	review_article
Snippet	As various machine learning (ML) algorithms have become more popular in orthopedic surgery, the research quality of these models requires further evaluation,...
SourceID	proquest pubmed
SourceType	Aggregation Database Index Database
StartPage	e45230
SubjectTerms	Arthroplasty, Replacement, Hip - statistics & numerical data Arthroplasty, Replacement, Knee - statistics & numerical data Bias Humans Prediction Algorithms Predictive Learning Models - standards Prognosis Risk Assessment - methods Risk Assessment - standards Software Validation Supervised Machine Learning - standards
Title	Supervised machine learning-based bias risk of prognostic models for total knee or hip arthroplasty patients: A systematic review
URI	https://www.ncbi.nlm.nih.gov/pubmed/41189192 https://www.proquest.com/docview/3268841277
Volume	104
WOSCitedRecordID	wos001608941400038&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings	1
isFullTextHit
isPrint
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1NS8QwEA3qinjx-_uDEbyGTdu0Sb2IqIuXXRZU2NuSNIkuYrvaXcGj_9xJ22W9CII9FAottOl08jLz-h4h55LFlkkXUyN0RLm0miqbChokzPDABkyp2mxC9HpyMEj7TcGtbGiVs5xYJWpTZL5G3kaYISUPQiEux2_Uu0b57mpjobFIWhFCGR_VYvCjiyBqU9s4SmicJnymOpSKdvemVi6sN-5ro79jzGqu6az_9y43yFqDMuGqDotNsmDzLbLSbfro2-Trfjr2SaK0Bl4rOqWFxj_iifqJzYAeqRI88RwKB57FlRde0hkq65wSEOvCpEDkDi-5tYBHz6MxYBx62wVE5JNPaCRbywu4grlgNNQ_y-yQx87tw_UdbcwYaBbFjFHtAhVrGZmQZVzFmCMTl3L_vGGojZe8MS6w2mhlXJIi7nPKRdIJa2zgQjxhlyzlRW73CZjMGBZqoWXKecYyaSyu750SDrNvLMQBOZsN7hCD3XcwVG6LaTmcD-8B2avf0HBcq3IMOS6VUsSrh3-4-oisht7H1zNTxDFpOfzU7QlZzj4mo_L9tIoi3Pf63W_botQZ
linkProvider	ProQuest
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Supervised+machine+learning-based+bias+risk+of+prognostic+models+for+total+knee+or+hip+arthroplasty+patients%3A+A+systematic+review&rft.jtitle=Medicine+%28Baltimore%29&rft.au=Zhang%2C+Hongxia&rft.au=Jiang%2C+Lina&rft.au=Zheng%2C+Jiang&rft.au=Li%2C+Chuanbo&rft.date=2025-10-17&rft.eissn=1536-5964&rft.volume=104&rft.issue=42&rft.spage=e45230&rft_id=info:doi/10.1097%2FMD.0000000000045230&rft_id=info%3Apmid%2F41189192&rft_id=info%3Apmid%2F41189192&rft.externalDocID=41189192
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1536-5964&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1536-5964&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1536-5964&client=summon