Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1828
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-18T18:12:18.861293Z
status: published
lens: g_space
sector: health
headline: Machine learning predicts major adverse kidney and cardiovascular events in heart failure with reduced ejection fraction
dek: Background Heart failure with reduced ejection fraction (HFrEF) links to adverse kidney and cardiovascular events. However, current machine learning (ML)-based algorithms do not include comprehensive echocardiographic features or predict adverse kidney outcomes. Methods We built ML models to predict 3-year all-cause mortality, major adverse cardiovascular events (MACE), and major adverse kidney events (MAKE) in retrospective HFrEF cohorts. Models were developed from 1,486 patients at Taipei Veterans General Hosp…
gain_title: Ensemble models using detailed clinical and echocardiographic features predicted 3-year mortality and cardiorenal events in HFrEF better than conventional scores, supporting earlier detection of high-risk patients.
problem_title: (none)
trace_subject: (none)
gain_reading: Ensemble models using detailed clinical and echocardiographic features predicted 3-year mortality and cardiorenal events in HFrEF better than conventional scores, supporting earlier detection of high-risk patients.
gain_evidence: Ensemble ML models with detailed clinical and echocardiographic features surpassed conventional risk scores
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers built machine learning models to forecast three-year death, major adverse cardiovascular events and major adverse kidney events in people with heart failure with reduced ejection fraction. They trained on 1,486 patients from Taipei Veterans General Hospital and tested on 246 patients from a post-acute-care program in Taichung.

The work matters because existing machine learning tools for heart failure omit comprehensive echocardiographic features and do not predict kidney outcomes. The authors conclude that their ensemble models beat traditional scores and could allow earlier detection, but the study is retrospective and the abstract provided gives no evidence of effects on patient care.
limitation: The models were built and validated using retrospective cohorts, and the external validation group was small at 246 patients, so prospective performance is not shown.
tag: Evidence-backed gain
key_points: Models predicted 3-year all-cause mortality, MACE and MAKE in HFrEF patients. | Development used 1,486 patients treated at Taipei Veterans General Hospital between 2011 and 2018. | External validation used 246 patients from a heart-failure post-acute-care program at Taichung Veterans General Hospital in 2020-2023. | Comparators were the MAGGIC score, logistic regression and the Kidney-Failure-Risk-Equation. | The ensemble model reached an AUROC of 0.863 for mortality.
rundown: The study addresses a gap: heart failure with reduced ejection fraction is linked to both kidney and cardiovascular adverse events, yet earlier machine learning algorithms lacked detailed echocardiographic inputs and did not forecast adverse kidney outcomes.

The team compared its models with the traditional MAGGIC score, logistic regression and the Kidney-Failure-Risk-Equation. For mortality, the ensemble approach had the highest AUROC at 0.863 with a 95% confidence interval of 0.814-0.913.

The authors conclude that combining detailed clinical and echocardiographic features allows early identification of major adverse cardiorenal events in this patient group.
sources:
- peer_reviewed | Kidney Research and Clinical Practice | https://doi.org/10.23876/j.krcp.26.020 | 2026-10-06
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

7f4bfe8a330cd0e04167da88f199b8c19fee9e56eefda17ca9cf42eb61f7e08e

previous

0000000000000000000000000000000000000000000000000000000000000000