Left Bundle Branch Area Pacing Versus Conventional Pacing in Patients with Advanced Atrioventricular Conduction Abnormalities: a Prospective Cohort Study.

BACKGROUND: Left bundle branch area pacing (LBBAP) is an emerging pacing method, which may prevent the deleterious effects of right ventricular pacing. The aim of this study is to compare the effects of LBBAP with right ventricular septal pacing (RVSP) in patients with advanced atrioventricular conduction abnormalities and preserved left ventricular ejection fraction. METHODS: The effect of pacing was evaluated by echocardiographic indices of dyssynchrony, including global myocardial work efficiency (GWE) and peak systolic dispersion (PSD). The primary endpoint was GWE postprocedural, at 3, 6 and 12 months after the procedure. RESULTS: Twenty patients received LBBAP and 18 RVSP. Complete follow-up was accomplished in 37 patients (97.4%), due to the death of a patient (RVSP arm), from non-related cause. GWE was significantly increased in the group of LBBAP compared to RVSP at all timepoints (90.8% in LBBAP vs 85.8% in RVSP group at 12 months, p=0.01). PSD was numerically lower in the LBBAP arm at all timepoints, yet not statistically significant (56.4 msec in LBBP vs 65.1 msec in RVSP arm at 12 months, p=0.178). The implantation time was increased (median 93 min in LBBAP vs 45 min in RVSP group, p<0.01), along with fluoroscopy time and dose area product (DAP), in the arm of LBBAP. There were no severe perioperative acute complications in either group. CONCLUSIONS: LBBAP is an emerging and safe technique for patients with a pacing indication. Despite the longer procedural and fluoroscopy time, as well as higher DAP, LBBAP seems to offer better left ventricular synchrony compared to RVSP, according to GWE measurements.

Deep Kernel Principal Component Analysis for multi-level feature learning.

Principal Component Analysis (PCA) and its nonlinear extension Kernel PCA (KPCA) are widely used across science and industry for data analysis and dimensionality reduction. Modern deep learning tools have achieved great empirical success, but a framework for deep principal component analysis is still lacking. Here we develop a deep kernel PCA methodology (DKPCA) to extract multiple levels of the most informative components of the data. Our scheme can effectively identify new hierarchical variables, called deep principal components, capturing the main characteristics of high-dimensional data through a simple and interpretable numerical optimization. We couple the principal components of multiple KPCA levels, theoretically showing that DKPCA creates both forward and backward dependency across levels, which has not been explored in kernel methods and yet is crucial to extract more informative features. Various experimental evaluations on multiple data types show that DKPCA finds more efficient and disentangled representations with higher explained variance in fewer principal components, compared to the shallow KPCA. We demonstrate that our method allows for effective hierarchical data exploration, with the ability to separate the key generative factors of the input data both for large datasets and when few training samples are available. Overall, DKPCA can facilitate the extraction of useful patterns from high-dimensional data by learning more informative features organized in different levels, giving diversified aspects to explore the variation factors in the data, while maintaining a simple mathematical formulation.

Safety and efficacy of left bundle branch area pacing compared with right ventricular pacing in patients with bradyarrhythmia and conduction system disorders: Systematic review and meta-analysis.

BACKGROUND: Right Ventricular Pacing (RVP) may have detrimental effects in ventricular function. Left Bundle Branch Area Pacing (LBBAP) is a new pacing strategy that appears to have better results. The aim of this systematic review and meta-analysis is to compare the safety and efficacy of LBBAP vs RVP in patients with bradyarrhythmia and conduction system disorders. METHODS: MEDLINE, EMBASE and Pubmed databases were searched for studies comparing LBBAP with RVP. Outcomes were all-cause mortality, atrial fibrillation (AF) occurrence, heart failure hospitalizations (HFH) and complications. QRS duration, mechanical synchrony and LVEF changes were also assessed. Pairwise meta-analysis was conducted using random and fixed effects models. RESULTS: Twenty-five trials with 4250 patients (2127 LBBAP) were included in the analysis. LBBAP was associated with lower risk for HFH (RR:0.33, CI 95%:0.21 to 0.50; p < 0.001), all-cause mortality (RR:0.52 CI 95%:0.34 to 0.80; p = 0.003), and AF occurrence (RR:0.43 CI 95%:0.27 to 0.68; p < 0.001) than RVP. Lead related complications were not different between the two groups (p = 0.780). QRSd was shorter in the LBBAP group at follow-up (WMD: -32.20 msec, CI 95%: -40.70 to -23.71; p < 0.001) and LBBAP achieved better intraventricular mechanical synchrony than RVP (SMD: -1.77, CI 95%: -2.45 to -1.09; p < 0.001). LBBAP had similar pacing thresholds (p = 0.860) and higher R wave amplitudes (p = 0.009) than RVP. CONCLUSIONS: LBBAP has better clinical outcomes, preserves ventricular electrical and mechanical synchrony and has excellent pacing parameters, with no difference in complications compared to RVP.

Right ventricular lead sensing latency in pacemaker therapy.

Background: Pacemaker implantation involves intraoperative testing of ventricular sensing using a device called a pacing system analyzer (PSA). The value obtained is expected to correspond to those taken by the pacemaker after its implantation. This study determined the latency period for sensing intracardiac electrogram (EGM) by the right ventricular (RV) lead. Methods: Patients without significant heart disease and underlying intrinsic atrioventricular (AV) conduction underwent Medtronic or Abbott dual-chamber pacemaker implantation with the RV lead positioned on the mid-septum. Real-time sensing data were obtained through PSA and after pacemaker implantation to evaluate latency as the time interval Q-VS between the onset of QRS on surface electrocardiogram and the sensed EGM by the RV lead. Results: Of 157 patients, 105 had narrow QRS (<120 ms) and 52 had wide QRS of complete right bundle branch block (RBBB). Both narrow-QRS and RBBB patients had longer sensing latency through PSA (50.9 ± 24.2 and 67.8 ± 32.9 ms, respectively) than through pacemaker (18.2 ± 12.8 and 31.2 ± 14.8 ms, respectively, both p < 0.001). RBBB patients had longer sensing latency compared with narrow QRS patients, either through PSA or through pacemaker (p < 0.001). The sensing latency of Medtronic recipients was longer than those of Abbott in narrow-QRS (p < 0.05), but not in RBBB. Conclusion: We demonstrated longer RV lead sensing latency (1) through PSA than through pacemaker, (2) in RBBB than in narrow-QRS, and (3) in Medtronic pacemakers compared with Abbott pacemakers. Knowledge of sensing latency helps the optimization of the AV delay.

Unsupervised learning of disentangled representations in deep restricted kernel machines with orthogonality constraints.

We introduce Constr-DRKM, a deep kernel method for the unsupervised learning of disentangled data representations. We propose augmenting the original deep restricted kernel machine formulation for kernel PCA by orthogonality constraints on the latent variables to promote disentanglement and to make it possible to carry out optimization without first defining a stabilized objective. After discussing a number of algorithms for end-to-end training, we quantitatively evaluate the proposed method's effectiveness in disentangled feature learning. We demonstrate on four benchmark datasets that this approach performs similarly overall to ß-VAE on several disentanglement metrics when few training points are available while being less sensitive to randomness and hyperparameter selection than ß-VAE. We also present a deterministic initialization of Constr-DRKM's training algorithm that significantly improves the reproducibility of the results. Finally, we empirically evaluate and discuss the role of the number of layers in the proposed methodology, examining the influence of each principal component in every layer and showing that components in lower layers act as local feature detectors capturing the broad trends of the data distribution, while components in deeper layers use the representation learned by previous layers and more accurately reproduce higher-level features.

Potential Relationship Between Lifestyle Changes and Incidence of Hospital Admissions for Acute Coronary Syndrome During the COVID-19 Lockdown.

Aims: To evaluate the impact of lockdown during the COVID-19 pandemic on lifestyle changes of the general population, and on admissions for acute coronary syndrome (ACS). Methods and Results: All ACS admissions during the COVID-19 lockdown (10 March to 4 May, 2020), in 3 municipalities (3 spoke, and 1 hub hospital), in Southwestern Greece (411,576 inhabitants), were prospectively recorded and compared to the equivalent periods during 2018, and 2019. A telephone survey of 1014 participants was conducted to explore the lifestyle habits of citizens aged ≥35-years-old before and during lockdown. The median ACS incidence rate decreased from 19.0 cases per week in 2018 and 21.5 in 2019 down to 13.0 in 2020 (RR: 0.66 during the Covid-19 lockdown; 95%CI: 0.53-0.82; P = 0.0002). This was driven by a significant reduction of admissions for Non-ST elevation myocardial infarction (NSTEMI) (RR: 0.68; 95%CI: 0.52-0.88; P = 0.0037), mainly in patients with a lower burden of cardiovascular risk factors, as we noticed an inverse association between the reduction of the incidence of ACS during the Covid-19 lockdown period and the number of registered patient risk factors. There was no difference in the rates of STEMI and population-based all-cause mortality across the examined time periods. The telephone survey demonstrated reduction of passive smoking, working hours, alcohol, junk food and salt consumption, and an increase in sleeping hours, mainly in participants with a lower burden of cardiovascular risk factors. Conclusions: A significant decline in ACS admissions during the COVID-19 lockdown was noted, affecting mainly NSTEMI patients with a lower burden of cardiovascular risk factors. This was accompanied by significant lifestyle changes. Thus, it is tempting to speculate that to some extend the latter might be associated with the observed decline in ACS admissions.

Optimal Versus Approximate Channel Selection Methods for EEG Decoding With Application to Topology-Constrained Neuro-Sensor Networks.

Channel selection or electrode placement for neural decoding is a commonly encountered problem in electroencephalography (EEG). Since evaluating all possible channel combinations is usually infeasible, one usually has to settle for heuristic methods or convex approximations without optimality guarantees. To date, it remains unclear how large the gap is between the selection made by these approximate methods and the truly optimal selection. The goal of this paper is to quantify this optimality gap for several state-of-the-art channel selection methods in the context of least-squares based neural decoding. To this end, we reformulate the channel selection problem as a mixed-integer quadratic program (MIQP), which allows the use of efficient MIQP solvers to find the optimal channel combination in a feasible computation time for up to 100 candidate channels. As this reveals the exact solution to the combinatorial problem, it allows to quantify the performance losses when using state-of-the-art sub-optimal (yet faster) channel selection methods. In a context of auditory attention decoding, we find that a greedy channel selection based on the utility metric does not show a significant optimality gap compared to optimal channel selection, whereas other state-of-the-art greedy or l1 -norm penalized methods do show a significant loss in performance. Furthermore, we demonstrate that the MIQP formulation also provides a natural way to incorporate topology constraints in the selection, e.g., for electrode placement in neuro-sensor networks with galvanic separation constraints. Furthermore, a combination of this utility-based greedy selection with an MIQP solver allows to perform a topology constrained electrode placement, even in large scale problems with more than 100 candidate positions.

Pyogenic Spondylodiscitis due to Streptococcus constellatus in an Immunocompromised Male Patient: A Case Report and Review of the Literature.

Pyogenic spondylodiscitis is a primary infection of the intervertebral disc and is a rare entity. Here, we describe the case of a 64-year-old male patient, a professional breeder, who attended the Emergency Department with sciatica and back pain that was worsening for a week. The patient had no history of surgery or trauma. The patient had poor oral hygiene. Magnetic resonance imaging (MRI) scan showed lumbar spondylodiscitis, and blood cultures revealed Streptococcus constellatus. The patient was initially treated with vancomycin but due to renal failure deterioration, the treatment was changed to daptomycin for 8 weeks. During hospitalization, he endured renal injury and nosocomial respiratory tract infection. The patient was discharged with no further complications. Follow-up revealed improvement of neurological signs. In our case, it seems that poor oral hygiene was the cause of bacteremia, which underlies the importance of a good oral health status in immunocompromised patients not only to prevent but also to successfully eliminate any dental source of infection. S. constellatus is an extremely rare pathogen and to our knowledge only two other cases of pyogenic spondylodiscitis are reported in the literature. Early diagnosis is very important for the prognosis of these patients.

Robust model predictive control for optimal continuous drug administration.

In this paper the model predictive control (MPC) technology is used for tackling the optimal drug administration problem. The important advantage of MPC compared to other control technologies is that it explicitly takes into account the constraints of the system. In particular, for drug treatments of living organisms, MPC can guarantee satisfaction of the minimum toxic concentration (MTC) constraints. A whole-body physiologically-based pharmacokinetic (PBPK) model serves as the dynamic prediction model of the system after it is formulated as a discrete-time state-space model. Only plasma measurements are assumed to be measured on-line. The rest of the states (drug concentrations in other organs and tissues) are estimated in real time by designing an artificial observer. The complete system (observer and MPC controller) is able to drive the drug concentration to the desired levels at the organs of interest, while satisfying the imposed constraints, even in the presence of modelling errors, disturbances and noise. A case study on a PBPK model with 7 compartments, constraints on 5 tissues and a variable drug concentration set-point illustrates the efficiency of the methodology in drug dosing control applications. The proposed methodology is also tested in an uncertain setting and proves successful in presence of modelling errors and inaccurate measurements.

Variable selection in nonlinear modeling based on RBF networks and evolutionary computation.

In this paper a novel variable selection method based on Radial Basis Function (RBF) neural networks and genetic algorithms is presented. The fuzzy means algorithm is utilized as the training method for the RBF networks, due to its inherent speed, the deterministic approach of selecting the hidden node centers and the fact that it involves only a single tuning parameter. The trade-off between the accuracy and parsimony of the produced model is handled by using Final Prediction Error criterion, based on the RBF training and validation errors, as a fitness function of the proposed genetic algorithm. The tuning parameter required by the fuzzy means algorithm is treated as a free variable by the genetic algorithm. The proposed method was tested in benchmark data sets stemming from the scientific communities of time-series prediction and medicinal chemistry and produced promising results.