Zhang, Pei

AbstractIntroductionThe efficacy of cardiac resynchronization therapy (CRT) has been widely studied in the medical literature; however, about 30% of candidates fail to respond to this treatment strategy. Smart computational approaches based on clinical data can help expose hidden patterns useful for identifying CRT responders.MethodsWe retrospectively analyzed the electronic health records of 1664 patients who underwent CRT procedures from January 1, 2002 to December 31, 2017. An ensemble of ensemble (EoE) machine learning (ML) system composed of a supervised and an unsupervised ML layers was developed to generate a prediction model for CRT response.ResultsWe compared the performance of EoE against traditional ML methods and the state‐of‐the‐art convolutional neural network (CNN) model trained on raw electrocardiographic (ECG) waveforms. We observed that the models exhibited improvement in performance as more features were incrementally used for training. Using the most comprehensive set of predictors, the performance of the EoE model in terms of the area under the receiver operating characteristic curve and F1‐score were 0.76 and 0.73, respectively. Direct application of the CNN model on the raw ECG waveforms did not generate promising results.ConclusionThe proposed CRT risk calculator effectively discriminates which heart failure (HF) patient is likely to respond to CRT significantly better than using clinical guidelines and traditional ML methods, thus suggesting that the tool can enhanced care management of HF patients by helping to identify high‐risk patients.

AbstractIntroductionThe best management strategy for patients with atrial fibrillation (AF) with heart failure (HF) and preserved left ventricular ejection fraction (LVEF) is unknown.Methods and ResultsThis cohort study was conducted in Olmsted County, Minnesota, with resources of the Rochester Epidemiology Project. Patients with incident AF occurring between 2000 and 2014 with a prior or concurrent HF were included. Patients with LVEF ≥ 50% were designated as HF and preserved ejection fraction (HFpEF) and those with LVEF < 50% were designated as HF and reduced ejection fraction (HFrEF). Rhythm control in the first year after AF diagnosis was defined as prescriptions for an antiarrhythmic drug, catheter ablation, or maze procedure. The primary endpoint was all‐cause mortality. The secondary endpoints were cardiovascular death, cardiovascular hospitalization, and stroke or transient ischemic attack. Of 859 patients (age, 77.2 ± 12.1 years; 49.2%, female), 447 had HFpEF‐AF, and 412 had HFrEF‐AF. There was no difference in all‐cause mortality (10‐year mortality, 83% vs 79%; p = .54) or secondary endpoints between the HFpEF‐AF and HFrEF‐AF, respectively. Compared with the rate control strategy, rhythm control in HFpEF‐AF patients (n = 40, 15.9%) offered no survival benefits (adjusted HR, 0.70; 95% CI, 0.42–1.16; p = .16), whereas rhythm control in HFrEF‐AF patients (n = 52, 22.5%) decrease cardiovascular mortality (HR, 0.38; 95% CI, 0.17–0.86; p = .02).ConclusionsPatients with HFpEF‐AF and HFrEF‐AF had similar poor prognoses. Rhythm control strategy was seldom adopted in community care in patients with HF and AF. A rhythm control strategy may provide survival benefit for patients with HFrEF‐AF and the benefit of rhythm control in patients with HFpEF‐AF warrants further study.