Previous studies on biventricular (BIV) pacing and cardiac resynchronization therapy-defibrillator (CRT-D) efficacy have used arbitrarily chosen BIV pacing percentages, and no study has employed implantable cardioverter defibrillator (ICD) patients as a control group. Using Kaplan–Meier plots, we estimated the threshold of BIV pacing percentage needed for CRT-D to be superior to ICD on the end-point of heart failure (HF) or death in 1219 left bundle branch block (LBBB) patients in the MADIT-CRT trial. Patients were censored at the time of crossover. In multivariable Cox analyses, no difference was seen in the risk of HF/death between ICD and CRT-D patients with BIV pacing ≤90% [HR = 0.78 (0.47–1.30), P = 0.344], and with increasing BIV pacing the risk of HF/death was decreased [CRT-D BIV 91–96% vs. ICD: HR = 0.63 (0.42–0.94), P = 0.024 and CRT-D BIV ≥97% vs. ICD: HR = 0.32 (0.23–0.44), P < 0.001]. The risk of death alone was reduced by 52% in CRT-D patients with BIV ≥97% (HR = 0.48, P < 0.016), when compared with ICD patients. Within the CRT-D group, for every 1 percentage point increase in BIV pacing, the risk of HF/death and death alone significantly decreased by 6 and 10%, respectively. Increasing BIV pacing percentage was associated with significant reductions in left ventricular volume. In patients with LBBB, who were in sinus rhythm at enrolment, BIV pacing exceeding 90% was associated with a benefit of CRT-D in HF/death when compared with ICD patients. Furthermore, BIV pacing ≥97% was associated with an even further reduction in HF/death, a significant 52% reduction in death alone, and increased reverse remodelling. Clinical trials.gov identifier: NCT00180271.
Photoplethysmography (PPG) is a ubiquitous physiological measurement that detects beat-to-beat pulsatile blood volume changes and hence has a potential for monitoring cardiovascular conditions, particularly in ambulatory settings. A PPG dataset that is created for a particular use case is often imbalanced, due to a low prevalence of the pathological condition it targets to predict and the paroxysmal nature of the condition as well. To tackle this problem, we propose log-spectral matching GAN (LSM-GAN), a generative model that can be used as a data augmentation technique to alleviate the class imbalance in a PPG dataset to train a classifier. LSM-GAN utilizes a novel generator that generates a synthetic signal without a up-sampling process of input white noises, as well as adds the mismatch between real and synthetic signals in frequency domain to the conventional adversarial loss. In this study, experiments are designed focusing on examining how the influence of LSM-GAN as a data augmentation technique on one specific classification task - atrial fibrillation (AF) detection using PPG. We show that by taking spectral information into consideration, LSM-GAN as a data augmentation solution can generate more realistic PPG signals.