RESUMO
There is a growing body of literature supporting the utilization of machine learning (ML) to improve diagnosis and prognosis tools of cardiovascular disease. The current study was to investigate the impact that the ML framework may have on the sensitivity of predicting the presence or absence of congenital heart disease (CHD) using fetal echocardiography. A comprehensive fetal echocardiogram including 2D cardiac chamber quantification, valvar assessments, assessment of great vessel morphology, and Doppler-derived blood flow interrogation was recorded. The postnatal echocardiogram was used to ascertain the diagnosis of CHD. A random forest (RF) algorithm with a nested tenfold cross-validation was used to train models for assessing the presence of CHD. The study population was derived from a database of 3910 singleton fetuses with maternal age of 28.8 ± 5.2 years and gestational age at the time of fetal echocardiography of 22.0 weeks (IQR 21-24). The proportion of CHD was 14.1% for the studied cohort confirmed by post-natal echocardiograms. Our proposed RF-based framework provided a sensitivity of 0.85, a specificity of 0.88, a positive predictive value of 0.55 and a negative predictive value of 0.97 to detect the CHD with the mean of mean ROC curves of 0.94 and the mean of mean PR curves of 0.84. Additionally, six first features, including cardiac axis, peak velocity of blood flow across the pulmonic valve, cardiothoracic ratio, pulmonary valvar annulus diameter, right ventricular end-diastolic diameter, and aortic valvar annulus diameter, are essential features that play crucial roles in adding more predictive values to the model in detecting patients with CHD. ML using RF can provide increased sensitivity in prenatal CHD screening with very good performance. The incorporation of ML algorithms into fetal echocardiography may further standardize the assessment for CHD.
RESUMO
Background: Patient adherence to cardioprotective medications improves outcomes of acute coronary syndrome (ACS), but few adherence-enhancing interventions have been tested in low-income and middle-income countries. Objectives: We aimed to assess whether a pharmacist-led intervention enhances medication adherence in patients with ACS and reduces mortality and hospital readmission. Methods: We conducted a randomized controlled trial in Vietnam. Patients with ACS were recruited, randomized to the intervention or usual care prior to discharge, and followed 3 months after discharge. Intervention patients received educational and behavioral interventions by a pharmacist. Primary outcome was the proportion of adherent patients 1 month after discharge. Adherence was a combined measure of self-reported adherence (the 8-item Morisky Medication Adherence Scale) and obtaining repeat prescriptions on time. Secondary outcomes were (1) the proportion of patients adherent to medication; (2) rates of mortality and hospital readmission; and (3) change in quality of life from baseline assessed with the European Quality of Life Questionnaire - 5 Dimensions - 3 Levels at 3 months after discharge. Logistic regression was used to analyze data. Registration: ClinicalTrials.gov (NCT02787941). Results: Overall, 166 patients (87 control, 79 intervention) were included (mean age 61.2 years, 73% male). In the analysis excluding patients from the intervention group who did not receive the intervention and excluding all patients who withdrew, were lost to follow-up, died or were readmitted to hospital, a greater proportion of patients were adherent in the intervention compared with the control at 1 month (90.0% vs. 76.5%; adjusted OR = 2.77; 95% CI, 1.01-7.62) and at 3 months after discharge (90.2% vs. 77.0%; adjusted OR = 3.68; 95% CI, 1.14-11.88). There was no significant difference in median change of EQ-5D-3L index values between intervention and control [0.000 (0.000; 0.275) vs. 0.234 (0.000; 0.379); p = 0.081]. Rates of mortality, readmission, or both were 0.8, 10.3, or 11.1%, respectively; with no significant differences between the 2 groups. Conclusion: Pharmacist-led interventions increased patient adherence to medication regimens by over 13% in the first 3 months after ACS hospital discharge, but not quality of life, mortality and readmission. These results are promising but should be tested in other settings prior to broader dissemination.