Deep-learning-based risk stratification for mortality of patients with acute myocardial infarction

Abstract

ObjectiveConventional risk stratification models for mortality of acute myocardial infarction (AMI) have potential limitations. This study aimed to develop and validate deep-learning-based risk stratification for the mortality of patients with AMI (DAMI).MethodsThe data of 22,875 AMI patients from the Korean working group of the myocardial infarction (KorMI) registry were exclusively divided into 12,152 derivation data of 36 hospitals and 10,723 validation data of 23 hospitals. The predictor variables were the initial demographic and laboratory data. The endpoints were in-hospital mortality and 12-months mortality. We compared the DAMI performance with the global registry of acute coronary event (GRACE) score, acute coronary treatment and intervention outcomes network (ACTION) score, and the thrombolysis in myocardial infarction (TIMI) score using the validation data.ResultsIn-hospital mortality for the study subjects was 4.4% and 6-month mortality after survival upon discharge was 2.2%. The areas under the receiver operating characteristic curves (AUCs) of the DAMI were 0.905 [95% confidence interval 0.902–0.909] and 0.870 [0.865–0.876] for the ST elevation myocardial infarction (STEMI) and non ST elevation myocardial infarction (NSTEMI) patients, respectively; these results significantly outperformed those of the GRACE (0.851 [0.846–0.856], 0.810 [0.803–0.819]), ACTION (0.852 [0.847–0.857], 0.806 [0.799–0.814] and TIMI score (0.781 [0.775–0.787], 0.593[0.585–0.603]). DAMI predicted 30.9% of patients more accurately than the GRACE score. As secondary outcome, during the 6-month follow-up, the high risk group, defined by the DAMI, has a significantly higher mortality rate than the low risk group (17.1% vs. 0.5%, p < 0.001).ConclusionsThe DAMI predicted in-hospital mortality and 12-month mortality of AMI patients more accurately than the existing risk scores and other machine-learning methods.

Machine learning enhances the performance of short and long-term mortality prediction model in non-ST-segment elevation myocardial infarction

“…The ML algorithms outperformed the traditional risk score methods when the predictors were the same, but the difference was similar in STEMI, and the best working algorithms varied according to the predictors and outcomes. Some studies suggested applying ML algorithms to enhance the performance of the prognosis prediction model for patients with AMI 11 , 17 . A recent study reported that deep learning (AUC: 0.905) could outperform the GRACE score (AUC: 0.851) in predicting the in-hospital mortality of AMI patients 11 .…”

“…GRACE and ACTION-GWTG presented a common model for ST-segment elevation myocardial infarction (STEMI) and non-ST-segment elevation myocardial infarction (NSTEMI), whereas TIMI suggested two distinct risk stratifications. Although these models were validated and are commonly accepted tools, concerns have been raised recently because most traditional risk stratifications were developed 20 years ago using randomized controlled trial (RCT) data before the introduction of drug-eluting stents and newer generation antiplatelets 11 . Moreover, the outcomes of the prediction models were limited to short-term mortality, such as in-hospital, 14-day, and 30-day mortality 3 , 12 , 13 .…”

“…Few attempts have been made to apply ML to risk prediction in patients with AMI, but the attempts made were inconsistent 15 , 16 . Recent research reported the possibility of performance enhancement using deep learning 11 , 17 . On the other hand, a direct comparison was not possible because far more predictors were included in the ML models than the traditional methods.…”

Identification and analysis of key genes associated with acute myocardial infarction by integrated bioinformatics methods

“…[ 2 , 3 ] However, AMI is still the leading threat to human health in CVDs. [ 4 , 5 ] In addition, previous studies have shown that the short-term prognosis of AMI is poor. [ 6 , 7 ] The cause of AMI is mainly related to the rupture of atherosclerotic plaque, leading to myocardial hypoxia, necrosis, and extensive myocardial damage.…”

Adversarial attack on deep learning-based dermatoscopic image recognition systems

“…By extension, our study suggests the vulnerability of all deep learning applications in the field of healthcare, including for example algorithms for the clinical management of patients with sepsis or risk stratification for mortality of patients with acute myocardial infarction. [ 23 , 24 ]…”