OBJECTIVE: To employ machine learning (ML) and stacked ensemble learning (EL) methods in predicting short- and long-term mortality in Asian patients diagnosed with NSTEMI/UA and to identify the associated features, subsequently evaluating these findings against established risk scores.
METHODS: We analyzed data from the National Cardiovascular Disease Database for Malaysia (2006-2019), representing a diverse NSTEMI/UA Asian cohort. Algorithm development utilized in-hospital records of 9,518 patients, 30-day data from 7,133 patients, and 1-year data from 7,031 patients. This study utilized 39 features, including demographic, cardiovascular risk, medication, and clinical features. In the development of the stacked EL model, four base learner algorithms were employed: eXtreme Gradient Boosting (XGB), Support Vector Machine (SVM), Naive Bayes (NB), and Random Forest (RF), with the Generalized Linear Model (GLM) serving as the meta learner. Significant features were chosen and ranked using ML feature importance with backward elimination. The predictive performance of the algorithms was assessed using the area under the curve (AUC) as a metric. Validation of the algorithms was conducted against the TIMI for NSTEMI/UA using a separate validation dataset, and the net reclassification index (NRI) was subsequently determined.
RESULTS: Using both complete and reduced features, the algorithm performance achieved an AUC ranging from 0.73 to 0.89. The top-performing ML algorithm consistently surpassed the TIMI risk score for in-hospital, 30-day, and 1-year predictions (with AUC values of 0.88, 0.88, and 0.81, respectively, all p < 0.001), while the TIMI scores registered significantly lower at 0.55, 0.54, and 0.61. This suggests the TIMI score tends to underestimate patient mortality risk. The net reclassification index (NRI) of the best ML algorithm for NSTEMI/UA patients across these periods yielded an NRI between 40-60% (p < 0.001) relative to the TIMI NSTEMI/UA risk score. Key features identified for both short- and long-term mortality included age, Killip class, heart rate, and Low-Molecular-Weight Heparin (LMWH) administration.
CONCLUSIONS: In a broad multi-ethnic population, ML approaches outperformed conventional TIMI scoring in classifying patients with NSTEMI and UA. ML allows for the precise identification of unique characteristics within individual Asian populations, improving the accuracy of mortality predictions. Continuous development, testing, and validation of these ML algorithms holds the promise of enhanced risk stratification, thereby revolutionizing future management strategies and patient outcomes.
METHODS: Out of 105 patients with IHD, 76 completed self-administration of HeartQoL at the clinic followed by at home within a 2-week interval. In retest, patients responded using non-interview methods (phone messaging, email, fax, and post). Phone interviewing was reserved for non-respondents to reminder.
RESULTS: Reliability of HeartQoL was good (intraclass correlation coefficients = 0.78-0.82), was supported in the Bland-Altman plot, and was comparable to five studies on MacNew of similar retest interval (MacNew-English = 0.70-0.75; translated MacNew = 0.72-0.91). Applicability of its standard error of measurement (0.20-0.25) and smallest detectable change (0.55-0.70) will depend on availability of normative data in future.
CONCLUSION: The reliability of HeartQoL is comparable to its parent instrument, the MacNew. The HeartQoL is a potentially reliable core IHD-specific HRQoL instrument in measuring group change.
METHODS: EPICOR Asia (Long-tErm follow-uP of antithrombotic management patterns In acute CORonary syndrome patients in Asia) (NCT01361386) is a prospective, multinational, observational study of patients discharged after hospitalization for an ACS, with 2-year follow-up. The aim is to describe short- and long-term (up to 2 years post-index event) AMPs in patients hospitalized for ACS and to record clinical outcomes, healthcare resource use, and self-reported health status. Pre- and in-hospital management, AMPs, and associated outcomes, with particular focus on ischemic and bleeding events, will be recorded during the 2-year follow up.
RESULTS: Between June 2011 and May 2012, 13 005 patients were enrolled. From these, 12 922 patients surviving an ACS (6616 with STEMI, 2570 with NSTEMI, and 3736 with UA) were eligible for inclusion from 219 hospitals across 8 countries and regions in Asia: China (n = 8214), Hong Kong (n = 177), India (n = 2468), Malaysia (n = 100), Singapore (n = 93), South Korea (n = 705), Thailand (n = 957), and Vietnam (n = 208).
CONCLUSIONS: EPICOR Asia will provide information regarding clinical management and AMPs for ACS patients in Asia. Impact of AMPs on clinical outcomes, healthcare resource use, and self-reported health status both during hospitalization and up to 2 years after discharge will also be described.
Methods: In this cross-sectional study, data from 147 ACS patients aged less than 45 years were analysed.
Results: The mean age was 39.1 (4.9) years, the male to female ratio was 3:1; 21.2% of patients presented with unstable angina, 58.5% had non-ST elevation myocardial infarction and 20.4% had ST elevation myocardial infarction. The most frequent risk factor of ACS was dyslipidaemia (65.3%), followed by hypertension (43.5%). In total, 49.7% of patients had inpatient complication(s), with the most common being heart failure (35.4%), followed by arrhythmia (20.4%). The significant factors associated with ACS complications were current smoking [adjusted odds ratio (AOR) 4.03; 95% confidence interval (CI): 1.33, 12.23;P-value = 0.014], diabetic mellitus [AOR 3.03; 95% CI: 1.19, 7.71;P-value = 0.020], treatments of fondaparinux [AOR 0.18; 95% CI: 0.08, 0.39;P-value < 0.001] and oral nitrates [AOR 0.18; 95% CI: 0.08, 0.42;P-value < 0.001].
Conclusions: Smoking status and diabetes mellitus were modifiable risk factors while pharmacological treatment was an important protective factor for ACS complications in young patients.