Improving classification of myocardial infarction with machine learning in a diverse population.
Phenotype classification with electronic health record (EHR) data is increasingly performed with machine learning (ML); however, their performance in diverse population remains understudied. We compared an international classification of diseases (ICD)–based algorithm with an ML phenotyping pipeline...
| Publicado en: | American Journal of Epidemiology Vol. 195; no. 3; pp. 841 - 850 |
|---|---|
| Autores principales: | , , , , , , , , , |
| Formato: | pictorial research tables/charts Journal Article |
| Publicado: |
Oxford University Press / USA
Mar2026
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=192099109&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 192099109 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 00029262 1X1 jtl: American Journal of Epidemiology issn: 00029262 maglogo: N pubinfo: dt: Mar2026 vid: 195 iid: 3 pid: 622 pub: Oxford University Press / USA artinfo: ui: 192099109 192099109 192099109 10.1093/aje/kwaf223 192099109 ppf: 841 ppct: 9 formats: tig: atl: Improving classification of myocardial infarction with machine learning in a diverse population. aug: au: Chen, Alicia W Hong, Chuan Ho, Yuk Lam Link, Nicholas Honerlaw, Jacqueline P Tanukonda, Vidisha Orkaby, Ariela R Qazi, Saadia Melley, Connor Galloway, Ashley affil: Massachusetts Veterans Epidemiology Research and Information Center (MAVERIC), VA Boston Healthcare System, Boston, MA, United States sug: subj: Myocardial Infarction Classification Machine Learning Algorithms Evaluation Phenotype Electronic Health Records Veterans Health Services Human International Classification of Diseases Self Report African Americans Aged Descriptive Statistics Male Female White Persons Natural Language Processing Sensitivity and Specificity Data Analysis Software Funding Source Aged: 65+ years Male Female ab: Phenotype classification with electronic health record (EHR) data is increasingly performed with machine learning (ML); however, their performance in diverse population remains understudied. We compared an international classification of diseases (ICD)–based algorithm with an ML phenotyping pipeline to classify myocardial infarction (MI) in a general and self-reported Black population. We determined the impact of differential performance by replicating a published MI risk factor study with MI defined by the ICD or ML algorithms. Individuals followed in the Veterans Health Administration (VHA) EHR with data from 2002 to 2019 were examined: 11 523 175 Veterans; mean age, 67.5 years; 93.8% male; 14.3% Black; 79.1% White. MI was classified using a published rule-based ICD algorithm and an ML pipeline, PheCAP, which incorporates natural language processing. Algorithms were trained and validated against n = 403 Veterans randomly selected and chart reviewed for MI (gold standard), oversampled for self-reported Black. Among chart-reviewed Veterans, the ICD algorithm had high positive predicted value (PPV) and low sensitivity (all race, PPV: 0.97, sensitivity: 0.17; Black Veterans, PPV: 0.94, sensitivity: 0.24). PheCAP MI had good PPV and higher sensitivity (all race, PPV: 0.90, sensitivity: 0.66; Black, PPV: 0.81, sensitivity: 0.79). Applying PheCAP MI to the entire VHA population to classify MI provided increased power to replicate findings from the published MI risk factor study compared to the ICD algorithm. pubtype: Academic Journal doctype: pictorial research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|