What does this research mean for the field?

A machine learning model (ASrisk) can predict the risk of aortic stenosis (AS) three years prior to diagnosis in the general population. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

The aim is to develop and validate a machine learning model for predicting aortic stenosis risk over three years using routine medical records.

March 13, 2026Open Access

Three-year risk prediction of aortic stenosis using routine medical records: derivation and validation in 919 954 individuals from two cohorts

Key Result

The ASrisk model predicted a significant increase in odds for AS diagnosis and aortic valve area reduction in individuals with ASrisk > 0.95, achieving an 11-fold enrichment for diagnosis.

Key Points

The aim is to develop and validate a machine learning model for predicting aortic stenosis risk over three years using routine medical records.
Developed a machine learning model for aortic stenosis risk prediction.
Used serum biomarkers and vital sign measurements for risk assessment.
Validated the model in two cohorts: Mount Sinai Data Warehouse and UK Biobank.
Odds of AS sequelae increased with higher ASrisk deciles; OR of 1.63 in MSDW.
Higher ASrisk linked to echocardiographic findings in undiagnosed individuals (e.g., OR 2.50 for aortic valve area).
Individuals with ASrisk > 0.95 had an 11-fold increase in AS diagnosis and an average aortic valve area reduction of 0.42 cm².

Structured PICO

Can a machine learning model using routine medical records predict the 3-year risk of aortic stenosis in the general population?

Population

919,954 individuals from two cohorts (Mount Sinai Data Warehouse, n=429,996 with 2,179 AS cases; UK Biobank, n=489,958 with 5,066 AS cases)

Intervention

Machine learning model (ASrisk) using serum biomarkers and vital sign measurements

Outcome

3-year prediction of aortic stenosis (AS) risk

A machine learning model utilizing routine serum biomarkers and vital signs can accurately predict the risk of developing aortic stenosis up to 3 years prior to clinical diagnosis.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Abstract Aims All-cause mortality ranges between 33% and 42% for individuals with untreated moderate to severe aortic stenosis (AS). Transcatheter aortic valve replacement makes this a treatable condition, if identified early. Machine learning-based tools show great promise to predict cardiovascular outcomes. Methods and results We developed and validated a machine learning model for 3-year prediction of AS risk (ASrisk) using serum biomarkers and vital sign measurements. We then evaluated the tool’s capacity to identify diagnoses of AS sequelae, echocardiographic outcomes in individuals not diagnosed with AS, as well as enrichment and 3-year aortic valve area reduction in individuals with high ASrisk. Among 919 954 participants, 429 996 were from the Mount Sinai Data Warehouse (MSDW) 2179 (0.5%) AS cases and 489 958 were from the UK Biobank 5066 (1%) AS cases. Odds ratio (OR) of AS sequelae increased quantitatively with ascending deciles of ASrisk OR 1.63 (95% CI 1.60–1.67) in MSDW. Increasing ASrisk by 1 SD resulted in higher odds of echocardiographic findings in undiagnosed individuals OR 1.88 (95% CI 1.71–2.06) for Doppler velocity index, OR 2.50 (95% CI 2.36–2.64) for aortic valve area, and OR 2.61 (95% CI 1.89–2.71) for mean gradient. Three years after risk assessment, individuals with ASrisk 0.95 show an 11-fold enrichment for AS diagnosis in both cohorts and an average reduction in aortic valve area of 0.42 cm2. Conclusion ASrisk can predict risk of AS 3 years ahead of diagnosis in the general population.