Artificial intelligence algorithm for predicting mortality of patients with acute heart failure
Key result
The deep-learning-based DAHF algorithm predicted in-hospital mortality with an AUC of 0.880 (95% CI 0.876-0.884), significantly outperforming the GWTG-HF score.
Why the study?
The study aimed to develop and validate a deep-learning-based artificial intelligence algorithm (DAHF) for predicting mortality in patients with acute heart failure.
Does a deep-learning-based artificial intelligence algorithm (DAHF) improve mortality prediction compared to existing risk scores in patients with acute heart failure?
Comparison
DAHF algorithm vs GWTG-HF score, MAGGIC score, and other machine-learning models
Design
Development and validation cohort study
Follow-up
36-month follow-up
Authors
Discussion
Loading...
Member takes
Implication
May aid mortality risk stratification in acute HF cohorts; leaves open prospective validation before clinical use.
Study Design
Type
Observational (n=6,924)
Multicenter
Yes
Structured PICO
Does a deep-learning-based artificial intelligence algorithm (DAHF) improve mortality prediction compared to existing risk scores in patients with acute heart failure?
P
Population
6,924 patients with acute heart failure across multiple hospitals, followed for up to 36 months to develop and validate a deep-learning mortality prediction algorithm.
E
Exposure
Deep-learning-based artificial intelligence algorithm (DAHF) for mortality prediction
C
Comparator
Get with the Guidelines-Heart Failure (GWTG-HF) score, Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) score, and other machine-learning models
O
Outcome
In-hospital, 12-month, and 36-month mortalityhard clinical
Main Result
Effect estimate: AUC 0.880 (95% CI 0.876-0.884)
A novel deep-learning algorithm (DAHF) provides superior prediction of in-hospital and long-term mortality in acute heart failure compared to traditional risk scores.
Cite This Study
Kwon et al. (2019) conducted an observational in Acute heart failure (n=6,924). Deep-learning-based artificial intelligence algorithm (DAHF) vs. GWTG-HF score, MAGGIC score, and other machine-learning models was evaluated on In-hospital mortality (AUC 0.880, 95% CI 0.876-0.884). The deep-learning-based DAHF algorithm predicted in-hospital mortality with an AUC of 0.880 (95% CI 0.876-0.884), significantly outperforming the GWTG-HF score.