MétaCan
Menu
Back to cohort
Record W4410769795 · doi:10.1101/2025.05.26.25328366

Short-term and long-term outcome prediction for patients with coronary artery disease using machine learning and comprehensive multi-center patient data

2025· preprint· en· W4410769795 on OpenAlexafffundabout
Bryan Har, Bin Li, Danielle A. Southern, Robert C. Welsh, Benjamin D. Tyrrell, C. Aengus Murphy, Arjun Puri, Joon S. Lee

Bibliographic record

VenuemedRxiv · 2025
Typepreprint
Languageen
FieldHealth Professions
TopicArtificial Intelligence in Healthcare
Canadian institutionsAlberta Health ServicesRoyal Alexandra HospitalUniversity of OttawaLibin Cardiovascular Institute of AlbertaUniversity of AlbertaUniversity of Calgary
FundersCanadian Institutes of Health ResearchAlberta Innovates
KeywordsTerm (time)Outcome (game theory)Coronary artery diseaseCenter (category theory)MedicineDiseaseCardiologyInternal medicineComputer scienceArtificial intelligenceMathematics

Abstract

fetched live from OpenAlex

Background Revascularization decision-making for patients with coronary artery disease (CAD) can benefit from accurate patient outcome prediction. While previous studies have employed data-driven methods including machine learning (ML) to develop prediction models, they were mostly based on small patient cohorts with strict inclusion and exclusion criteria, limited feature sets, and only internal validation. Objectives To develop and externally validate ML-based models to predict a wide range of short- and long-term outcomes for patients with obstructive CAD using large-scale multi-center patient data. Methods Comprehensive data from patients with obstructive CAD who underwent coronary angiography at three hospitals in Alberta, Canada between 2009 and 2019 were extracted from the APPROACH Registry and linked administrative health databases. To predict all-cause mortality and major adverse cardiovascular events at 90 days, 1 year, 3 years, and 5 years, over 12,000 features were considered in an extensive ML framework that employed rigorous hyperparameter tuning, calibration, algorithmic bias assessment, and external validation. In addition to traditional ML models, we employed a generative transformer-based tabular foundation model, TabPFN. To increase the clinical utility of these prediction models, we also performed a secondary analysis that investigated the impact of the exclusion of angiography data on prediction performance. Results A total of 44,462 catheterizations from 38,767 unique patients were included in the study. The median areas under the receiver operating characteristic curves of the best models, mostly TabPFNs, in external validation ranged from 0.797 to 0.845 and 0.694 to 0.753 for mortality and MACE, respectively. CAD factors, angiography results, and patient history were the most influential feature groups. The algorithmic bias assessment focusing on patient sex showed that the models were mostly fair. The secondary analysis showed that prediction performance degraded slightly when angiography features were excluded. Conclusions The prediction performance reported in this study is state-of-the-art compared to previous studies. The large sample size, extensive feature set, external validation, and transformer architecture led to personalized models with robust performance. The models from this study have the potential to improve coronary revascularization decision-making and patient outcomes via accurate prognosis.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.050
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0000.000
Science and technology studies0.0010.000
Scholarly communication0.0000.000
Open science0.0000.002
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.241
GPT teacher head0.462
Teacher spread0.221 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations1
Published2025
Admission routes3
Has abstractyes

Explore more

Same venuemedRxivSame topicArtificial Intelligence in HealthcareFrench-language works237,207