Cohort-Derived Machine Learning Models for Individual Prediction of Chronic Kidney Disease in People Living With Human Immunodeficiency Virus: A Prospective Multicenter Cohort Study.
HIV
chronic kidney disease
digital epidemiology
machine learning
prediction
Journal
The Journal of infectious diseases
ISSN: 1537-6613
Titre abrégé: J Infect Dis
Pays: United States
ID NLM: 0413675
Informations de publication
Date de publication:
13 10 2021
13 10 2021
Historique:
received:
18
12
2019
accepted:
05
05
2020
pubmed:
10
5
2020
medline:
25
2
2022
entrez:
10
5
2020
Statut:
ppublish
Résumé
It is unclear whether data-driven machine learning models, which are trained on large epidemiological cohorts, may improve prediction of comorbidities in people living with human immunodeficiency virus (HIV). In this proof-of-concept study, we included people living with HIV in the prospective Swiss HIV Cohort Study with a first estimated glomerular filtration rate (eGFR) >60 mL/minute/1.73 m2 after 1 January 2002. Our primary outcome was chronic kidney disease (CKD)-defined as confirmed decrease in eGFR ≤60 mL/minute/1.73 m2 over 3 months apart. We split the cohort data into a training set (80%), validation set (10%), and test set (10%), stratified for CKD status and follow-up length. Of 12 761 eligible individuals (median baseline eGFR, 103 mL/minute/1.73 m2), 1192 (9%) developed a CKD after a median of 8 years. We used 64 static and 502 time-changing variables: Across prediction horizons and algorithms and in contrast to expert-based standard models, most machine learning models achieved state-of-the-art predictive performances with areas under the receiver operating characteristic curve and precision recall curve ranging from 0.926 to 0.996 and from 0.631 to 0.956, respectively. In people living with HIV, we observed state-of-the-art performances in forecasting individual CKD onsets with different machine learning algorithms.
Sections du résumé
BACKGROUND
It is unclear whether data-driven machine learning models, which are trained on large epidemiological cohorts, may improve prediction of comorbidities in people living with human immunodeficiency virus (HIV).
METHODS
In this proof-of-concept study, we included people living with HIV in the prospective Swiss HIV Cohort Study with a first estimated glomerular filtration rate (eGFR) >60 mL/minute/1.73 m2 after 1 January 2002. Our primary outcome was chronic kidney disease (CKD)-defined as confirmed decrease in eGFR ≤60 mL/minute/1.73 m2 over 3 months apart. We split the cohort data into a training set (80%), validation set (10%), and test set (10%), stratified for CKD status and follow-up length.
RESULTS
Of 12 761 eligible individuals (median baseline eGFR, 103 mL/minute/1.73 m2), 1192 (9%) developed a CKD after a median of 8 years. We used 64 static and 502 time-changing variables: Across prediction horizons and algorithms and in contrast to expert-based standard models, most machine learning models achieved state-of-the-art predictive performances with areas under the receiver operating characteristic curve and precision recall curve ranging from 0.926 to 0.996 and from 0.631 to 0.956, respectively.
CONCLUSIONS
In people living with HIV, we observed state-of-the-art performances in forecasting individual CKD onsets with different machine learning algorithms.
Identifiants
pubmed: 32386061
pii: 5835004
doi: 10.1093/infdis/jiaa236
pmc: PMC8514185
doi:
Types de publication
Journal Article
Multicenter Study
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
1198-1208Subventions
Organisme : Swiss National Science Foundation
ID : 177499
Pays : Switzerland
Informations de copyright
© The Author(s) 2020. Published by Oxford University Press for the Infectious Diseases Society of America.
Références
Nat Biomed Eng. 2018 Oct;2(10):719-731
pubmed: 31015651
J Med Internet Res. 2016 Dec 16;18(12):e323
pubmed: 27986644
Infect Control Hosp Epidemiol. 2018 Dec;39(12):1457-1462
pubmed: 30394238
Antiviral Res. 2010 Dec;88(3):347-54
pubmed: 20887753
J Clin Epidemiol. 2008 Apr;61(4):344-9
pubmed: 18313558
AIDS. 2014 Jun 1;28(9):1289-95
pubmed: 24922479
Nat Med. 2019 Jan;25(1):44-56
pubmed: 30617339
AIDS. 2013 Jun 19;27(10):1573-81
pubmed: 23435293
AIDS. 2017 Jan 28;31(3):427-436
pubmed: 27831953
Clin Infect Dis. 2020 Feb 14;70(5):890-897
pubmed: 30953057
N Engl J Med. 2019 Apr 4;380(14):1347-1358
pubmed: 30943338
AIDS. 2018 Aug 24;32(13):1829-1835
pubmed: 29847332
AIDS Rev. 2016 Oct-Dec;18(4):184-192
pubmed: 27438578
Neural Comput. 1997 Nov 15;9(8):1735-80
pubmed: 9377276
BMC Nephrol. 2017 Feb 10;18(1):58
pubmed: 28183270
Ann Intern Med. 2009 May 5;150(9):604-12
pubmed: 19414839
Ann Intern Med. 2015 May 19;162(10):735-6
pubmed: 25984857
Curr Opin HIV AIDS. 2016 Sep;11(5):492-500
pubmed: 27254748
J Acquir Immune Defic Syndr. 2016 Sep 1;73(1):39-46
pubmed: 27028501
Int J Epidemiol. 2010 Oct;39(5):1179-89
pubmed: 19948780
Lancet HIV. 2016 Jan;3(1):e23-32
pubmed: 26762990
HIV Med. 2013 Apr;14(4):195-207
pubmed: 22998068
PLoS Med. 2015 Mar 31;12(3):e1001809
pubmed: 25826420