Aim      To compare the accuracy of predicting an in-hospital fatal outcome for models based on current machine-learning technologies in patients with ischemic heart disease (IHD) after coronary bypass (CB) surgery.Material and methods  A retrospective analysis of 866 electronic medical records was performed for patients (685 men and 181 women) who have had a CB surgery for IHD in 2008–2018. Results of clinical, laboratory, and instrumental evaluations obtained prior to the CB surgery were analyzed. Patients were divided into two groups: group 1 included 35 (4 %) patients who died within the first 20 days of CB, and group 2 consisted of 831 (96 %) patients with a beneficial outcome of the surgery. Predictors of the in-hospital fatal outcome were identified by a multistep selection procedure with analysis of statistical hypotheses and calculation of weight coefficients. For construction of models and verification of predictors, machine-learning methods were used, including the multifactorial logistic regression (LR), random forest (RF), and artificial neural networks (ANN). Model accuracy was evaluated by three metrics: area under the ROC curve (AUC), sensitivity, and specificity. Cross validation of the models was performed on test samples, and the control validation was performed on a cohort of patients with IHD after CB, whose data were not used in development of the models.Results The following 7 risk factors for in-hospital fatal outcome with the greatest predictive potential were isolated from the EuroSCORE II scale: ejection fraction (EF) <30 %, EF 30-50 %, age of patients with recent MI, damage of peripheral arterial circulation, urgency of CB, functional class III-IV chronic heart failure, and 5 additional predictors, including heart rate, systolic blood pressure, presence of aortic stenosis, posterior left ventricular (LV) wall relative thickness index (RTI), and LV relative mass index (LVRMI). The models developed by the authors using LR, RF and ANN methods had higher AUC values and sensitivity compared to the classical EuroSCORE II scale. The ANN models including the RTI and LVRMI predictors demonstrated a maximum level of prognostic accuracy, which was illustrated by values of the quality metrics, AUC 93 %, sensitivity 90 %, and specificity 96 %. The predictive robustness of the models was confirmed by results of the control validation.Conclusion      The use of current machine-learning technologies allowed developing a novel algorithm for selection of predictors and highly accurate models for predicting an in-hospital fatal outcome after CB.

Introduction. Postoperative atrial fibrillation (POAF) is one of the most common complications of coronary artery bypass grafting (CABG) and occurs in 25–65% of patients.Aim. The study aimed to assess the predictive potential of preoperative risk factors for POAF in patients with coronary artery disease (CAD) after CABG based on machine learning (ML) methods.Material and Methods. An observational retrospective study was carried out based on data from 866 electronic case histories of CAD patients with a median age of 63 years and a 95% confidence interval [63; 64], who underwent isolated CABG on cardiopulmonary bypass. Patients were assigned to two groups: group 1 comprised 147 (18%) patients with newly registered atrial fibrillation (AF) paroxysms; group 2 included 648 (81.3%) patients without cardiac arrhythmia. The preoperative clinical and functional status was assessed using 100 factors. We used statistical analysis methods (Chi-square, Fisher, Mann – Whitney, and univariate logistic regression (LR) tests) and ML tests (multivariate LR and stochastic gradient boosting (SGB)) for data processing and analysis. The models’ accuracy was assessed by three quality metrics: area under the ROC-curve (AUC), sensitivity, and specificity. The cross-validation procedure was performed at least 1000 times on randomly selected data.Results. The processing and analysis of preoperative patient status indicators using ML methods allowed to identify 10 predictors that were linearly and nonlinearly related to the development of POAF. The most significant predictors were the anteroposterior dimension of the left atrium, tricuspid valve insufficiency, ejection fraction <40%, duration of the P–R interval, and chronic heart failure of functional class III–IV. The accuracy of the best predictive multifactorial model of LR was 0.61 in AUC, 0.49 in specificity, and 0.72 in sensitivity. The values of similar quality metrics for the best model based on SGB were 0.64, 0.6, and 0.68, respectively.Conclusion. The use of SGB made it possible to verify the nonlinearly related predictors of POAF. The prospects for further research on this problem require the use of modern medical care methods that allow taking into account the individual characteristics of patients when developing predictive models.

No abstract

No abstract