Artificial intelligence algorithm for predicting mortality of patients with acute heart failure

Abstract

Aims This study aimed to develop and validate deep-learning-based artificial intelligence algorithm for predicting mortality of AHF (DAHF). Methods and results 12,654 dataset from 2165 patients with AHF in two hospitals were used as train data for DAHF development, and 4759 dataset from 4759 patients with AHF in 10 hospitals enrolled to the Korean AHF registry were used as performance test data. The endpoints were in-hospital, 12-month, and 36-month mortality. We compared the DAHF performance with the Get with the Guidelines–Heart Failure (GWTG-HF) score, Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) score, and other machine-learning models by using the test data. Area under the receiver operating characteristic curve of the DAHF were 0.880 (95% confidence interval, 0.876–0.884) for predicting in-hospital mortality; these results significantly outperformed those of the GWTG-HF (0.728 [0.720–0.737]) and other machine-learning models. For predicting 12- and 36-month endpoints, DAHF (0.782 and 0.813) significantly outperformed MAGGIC score (0.718 and 0.729). During the 36-month follow-up, the high-risk group, defined by the DAHF, had a significantly higher mortality rate than the low-risk group( p <0.001). Conclusion DAHF predicted the in-hospital and long-term mortality of patients with AHF more accurately than the existing risk scores and other machine-learning models.

Identification of risk features for complication in Gaucher’s disease patients: a machine learning analysis of the Spanish registry of Gaucher disease

“…Large-scale data (big data) in our case when referring to a rare disease have been adapted to the number of cases available; this kind of analysis is a new tool that has recently been incorporated into biomedical activity; machine learning is the study of computer algorithms that improve automatically through experience and it involves a wide series of algorithms, classification and regression models such as decision trees being some of them [ 26 – 29 ]. This methodology is especially useful for obtaining pooled information on the diversity of outcomes and identifying prognostic factors potentially related to disease complications [ 35 ]. In rare disease research, this is of particular interest due to the scarcity and the spread of the data among the different centers [ 36 ].…”

Short Keynote Paper: Mainstreaming Personalized Healthcare-Transforming Healthcare through new era of Artificial Intelligence

“…Today, AI is already used in healthcare [26] for example to decrease false-positive results in screening for breast cancer [27], [28], reduce medical transcription costs [29], improve physician workflow while relieving and helping to prevent burnout [30], robotic surgery resulting in shorter length of hospitalization and loss of blood [31] and predicting mortality rates of patients with acute heart failure [32].…”

Neural network analysis of mortality risk predictors in patients after acute coronary syndrome

“…Поэтому, в своей работе мы использовали относительно неглубокую сеть. Указанное обстоятельство несколько снижает чувствительность созданной прогностической модели [9].…”

“…Дальнейшее увеличение обучаемой выборки позволит выявить наиболее значимые признаки, исследовать признаки на наличие выбросов. С помощью искусственного увеличения тренировочного корпуса, путем различных аугментаций можно нарастить мощность нейронной сети за счёт добавления свёрточных и рекуррентных слоев, что позволит анализировать выявленные закономерности во времени [9][10][11].…”