RESUMO
Many diverse phenomena in nature often inherently encode both short- and long-term temporal dependencies, which especially result from the direction of the flow of time. In this respect, we discovered experimental evidence suggesting that interrelations of these events are higher for closer time stamps. However, to be able for attention-based models to learn these regularities in short-term dependencies, it requires large amounts of data, which are often infeasible. This is because, while they are good at learning piece-wise temporal dependencies, attention-based models lack structures that encode biases in time series. As a resolution, we propose a simple and efficient method that enables attention layers to better encode the short-term temporal bias of these data sets by applying learnable, adaptive kernels directly to the attention matrices. We chose various prediction tasks for the experiments using Electronic Health Records (EHR) data sets since they are great examples with underlying long- and short-term temporal dependencies. Our experiments show exceptional classification results compared to best-performing models on most tasks and data sets.
RESUMO
Single neuron models are fundamental for computational modeling of the brain's neuronal networks, and understanding how ion channel dynamics mediate neural function. A challenge in defining such models is determining biophysically realistic channel distributions. Here, we present an efficient, highly parallel evolutionary algorithm for developing such models, named NeuroGPU-EA. NeuroGPU-EA uses CPUs and GPUs concurrently to simulate and evaluate neuron membrane potentials with respect to multiple stimuli. We demonstrate a logarithmic cost for scaling the stimuli used in the fitting procedure. NeuroGPU-EA outperforms the typically used CPU based evolutionary algorithm by a factor of 10 on a series of scaling benchmarks. We report observed performance bottlenecks and propose mitigation strategies. Finally, we also discuss the potential of this method for efficient simulation and evaluation of electrophysiological waveforms.
RESUMO
BACKGROUND: The membrane potential of individual neurons depends on a large number of interacting biophysical processes operating on spatial-temporal scales spanning several orders of magnitude. The multi-scale nature of these processes dictates that accurate prediction of membrane potentials in specific neurons requires the utilization of detailed simulations. Unfortunately, constraining parameters within biologically detailed neuron models can be difficult, leading to poor model fits. This obstacle can be overcome partially by numerical optimization or detailed exploration of parameter space. However, these processes, which currently rely on central processing unit (CPU) computation, often incur orders of magnitude increases in computing time for marginal improvements in model behavior. As a result, model quality is often compromised to accommodate compute resources. NEW METHOD: Here, we present a simulation environment, NeuroGPU, that takes advantage of the inherent parallelized structure of the graphics processing unit (GPU) to accelerate neuronal simulation. RESULTS & COMPARISON WITH EXISTING METHODS: NeuroGPU can simulate most biologically detailed models 10-200 times faster than NEURON simulation running on a single core and 5 times faster than GPU simulators (CoreNEURON). NeuroGPU is designed for model parameter tuning and best performs when the GPU is fully utilized by running multiple (> 100) instances of the same model with different parameters. When using multiple GPUs, NeuroGPU can reach to a speed-up of 800 fold compared to single core simulations, especially when simulating the same model morphology with different parameters. We demonstrate the power of NeuoGPU through large-scale parameter exploration to reveal the response landscape of a neuron. Finally, we accelerate numerical optimization of biophysically detailed neuron models to achieve highly accurate fitting of models to simulation and experimental data. CONCLUSIONS: Thus, NeuroGPU is the fastest available platform that enables rapid simulation of multi-compartment, biophysically detailed neuron models on commonly used computing systems accessible by many scientists.
Assuntos
Algoritmos , Gráficos por Computador , Simulação por Computador , Potenciais da Membrana , Neurônios/fisiologiaRESUMO
Background: Subclinical atrial fibrillation (AF) is one of the pathogeneses of embolic stroke. Detection of occult AF and providing proper anticoagulant treatment is an important way to prevent stroke recurrence. The purpose of this study was to determine whether an artificial intelligence (AI) model can assess occult AF using 24-h Holter monitoring during normal sinus rhythm. Methods: This study is a retrospective cohort study that included those who underwent Holter monitoring. The primary outcome was identifying patients with AF analyzed with an AI model using 24-h Holter monitoring without AF documentation. We trained the AI using a Holter monitor, including supraventricular ectopy (SVE) events (setting 1) and excluding SVE events (setting 2). Additionally, we performed comparisons using the SVE burden recorded in Holter annotation data. Results: The area under the receiver operating characteristics curve (AUROC) of setting 1 was 0.85 (0.83-0.87) and that of setting 2 was 0.84 (0.82-0.86). The AUROC of the SVE burden with Holter annotation data was 0.73. According to the diurnal period, the AUROCs for daytime were 0.83 (0.78-0.88) for setting 1 and 0.83 (0.78-0.88) for setting 2, respectively, while those for nighttime were 0.85 (0.82-0.88) for setting 1 and 0.85 (0.80-0.90) for setting 2. Conclusion: We have demonstrated that an AI can identify occult paroxysmal AF using 24-h continuous ambulatory Holter monitoring during sinus rhythm. The performance of our AI model outperformed the use of SVE burden in the Holter exam to identify paroxysmal AF. According to the diurnal period, nighttime recordings showed more favorable performance compared to daytime recordings.
RESUMO
High ionic conductivity and good stability are major factors that influence the use of polymer electrolytes in electrochemical storage and conversion devices. In this study, we present polyurethane acrylate (PUA) membranes having enhanced ionic conductivity and swelling stability by double crosslinking the polyurethane (PU) and polyacrylate (PA) compartments. The crosslinking agent concentration was varied to control their mechanical properties, swelling stability, and ionic conductivity. Under optimum conditions, the electrolyte uptake of the double-crosslinked PUA membranes without notable defects was 245%. The maximum ionic conductivity of these membranes reached 9.6 mS/cm, which was higher than those with respect to most of the previously reported PUA- or PU-based polymer electrolytes.
RESUMO
A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.
RESUMO
High-performance devices based on conducting polymers (CPs) require the fabrication of a thick CP-coated electrode with high stability. Herein, we propose a method for enhancing the interfacial adhesion strength between a gold electrode and an electropolymerized polypyrrole (pPy) layer by introducing a polyethyleneimine (PEI) layer. Although this insulating layer hinders the initial growth of the pPy layer on the Au surface, it improves the adhesion by up to 250%. Therefore, a thick layer of pPy can be fabricated without delamination during drying. X-ray photoelectron spectroscopy shows that the PEI layer interacts with the Au surface via polar/ionic groups and van der Waals interactions. Scanning electron microscopy reveals that the cohesion of the pPy layer is stronger than the interfacial adhesion between the PEI layer and the pPy layer. Importantly, the electroactivities of pPy and its dopant are unaffected by the PEI layer, and the electrochemical storage capacity of the pPy layers on the PEI-coated Au electrodes increases with thickness, reaching ~530 mC/cm2. Negative potential sweep is detrimental to pPy layer adhesion: pPy layers on a bare Au electrode peel off instantly as the potential is swept from 0.2 to -0.7 V, and most of the charge stored in the layer becomes inaccessible. In contrast, pPy layers deposited on PEI coated Au electrode are mechanically stable and majority of the charge can be accessed, demonstrating that this method is also effective for enhancing electrochemical stability. Our simple approach can find utility in various applications involving CP-coated electrodes, where thickness-dependent performance must be improved without loss of stability.
RESUMO
Recent research has highlighted the potential use of "smart" films, such as graphene sheets, that would allow for the controlled release of a variety of therapeutic drugs. Taking full advantage of these versatile conducting sheets, we investigated the novel concept of applying graphene oxide (GO) and reduced graphene oxide (rGO) materials as both barrier and conducting layers that afford controlled entrapment and release of any molecules of interest. We fabricated multilayered nanofilm architectures using a hydrolytically degradable cationic poly(ß-amino ester) (PAE), a model protein antigen, ovalbumin (OVA) as a building block along with the GO and rGO. We successfully showed that these multilayer films are capable of blocking the initial burst release of OVA, and they can be triggered to precisely control the release upon the application of electrochemical potential. This new drug delivery platform will find its usefulness in various transdermal drug delivery devices where on-demand control of drug release from the surface is necessary.