Home-Based Cardiac Rehabilitation Among Patients Unwilling to Participate in Hospital-Based Programs.

PURPOSE: Asynchronous home-based cardiac rehabilitation (HBCR) is a viable alternative to center-based cardiac rehabilitation (CBCR). However, to achieve significant functional improvement, a high level of adherence and activity must be achieved. The effectiveness of HBCR among patients who actively avoid CBCR has not been effectively investigated. This study aimed to investigate the effectiveness of the HBCR program among patients unwilling to participate in CBCR. METHODS: A randomized prospective study enrolled 45 participants to a 6-mo HBCR program and the remaining 24 were allocated to regular care. Both groups were digitally monitored for physical activity (PA) and self-reported outcomes. Change in peak oxygen uptake (VO 2peak ), the primary study outcome, was measured by the cardiopulmonary exercise test, immediately before program start and 4 mo thereafter. RESULTS: The study included 69 patients, 81% men, aged 55.9 ±12 yr, enrolled in a 6-mo HBCR program to follow a myocardial infarction (25.4%) or coronary interventions (41.3%), heart failure hospitalization (29%), or heart transplantation (10%). Weekly aerobic exercise totaled a median of 193.2 (110.2-251.5) min (129% of set exercise goal), of which 112 (70-150) min was in the heart rate zone recommended by the exercise physiologist.After 4 mo, VO 2peak improved by 10.2% in the intervention group versus -2.7% in the control group (+2.46 ± 2.67 vs -0.72 ± 3.02 mL/kg/min; P < .001). CONCLUSION: The monthly PA of patients in the HBCR versus conventional CBCR group were well within guideline recommendations, showing a significant improvement in cardiorespiratory fitness. Risk level, age, and lack of motivation at the beginning of the program did not prevent achieving goals and maintaining adherence.

Israel's first national remote cardiac rehabilitation program: A retrospective analysis.

Aims: Cardiac rehabilitation is an essential component of secondary prevention consistently unexploited by most eligible patients. Accordingly, the remote cardiac rehabilitation program (RCRP) was developed to create optimal conditions for remote instruction and supervision for patients to enable successful completion of the program. Methods: This study comprised 306 patients with established coronary heart disease who underwent a 6-month RCRP. RCRP involves regular exercise, monitored by a smartwatch that relays data to the operations center and a mobile application on the patient's smartphone. A stress test was performed immediately before the RCRP and repeated after 3 months. The aims were to determine the effectiveness of the RCRP in improving aerobic capacity, and correlating the program goals and first-month activity, with attaining program goals during the last month. Results: Participants were mostly male (81.5%), aged 58 ± 11, enrolled in the main after a myocardial infarction or coronary interventions. Patients exercised aerobically for 183 min each week, 101 min (55% of total exercise) at the target heart rate. There was a significant improvement in exercise capacity, assessed by stress tests, metabolic equivalents which increased from 9.5 ± 3 to 11.4 ± 7(p < 0.001). Independent predictors of RCRP goals were older age and more minutes of aerobic exercise during the first program month (p < 0.05). Conclusion: Participants succeeded in performing guideline recommendations, resulting in a significant improvement in exercise capacity. Older age and increased volume of first month of exercise were significant factors associated with a greater likelihood to attain program goals.

Deep Autoregressive Models for the Efficient Variational Simulation of Many-Body Quantum Systems.

Artificial neural networks were recently shown to be an efficient representation of highly entangled many-body quantum states. In practical applications, neural-network states inherit numerical schemes used in variational Monte Carlo method, most notably the use of Markov-chain Monte Carlo (MCMC) sampling to estimate quantum expectations. The local stochastic sampling in MCMC caps the potential advantages of neural networks in two ways: (i) Its intrinsic computational cost sets stringent practical limits on the width and depth of the networks, and therefore limits their expressive capacity; (ii) its difficulty in generating precise and uncorrelated samples can result in estimations of observables that are very far from their true value. Inspired by the state-of-the-art generative models used in machine learning, we propose a specialized neural-network architecture that supports efficient and exact sampling, completely circumventing the need for Markov-chain sampling. We demonstrate our approach for two-dimensional interacting spin models, showcasing the ability to obtain accurate results on larger system sizes than those currently accessible to neural-network quantum states.

Quantum Entanglement in Deep Learning Architectures.

Modern deep learning has enabled unprecedented achievements in various domains. Nonetheless, employment of machine learning for wave function representations is focused on more traditional architectures such as restricted Boltzmann machines (RBMs) and fully connected neural networks. In this Letter, we establish that contemporary deep learning architectures, in the form of deep convolutional and recurrent networks, can efficiently represent highly entangled quantum systems. By constructing tensor network equivalents of these architectures, we identify an inherent reuse of information in the network operation as a key trait which distinguishes them from standard tensor network-based representations, and which enhances their entanglement capacity. Our results show that such architectures can support volume-law entanglement scaling, polynomially more efficiently than presently employed RBMs. Thus, beyond a quantification of the entanglement capacity of leading deep learning architectures, our analysis formally motivates a shift of trending neural-network-based wave function representations closer to the state-of-the-art in machine learning.