Search

In this study, stochastic resonance (SR) exploits the inherent stochastic characteristics of the OTS threshold voltage to enhance the inference performance of neural networks. First, the threshold switching of the OTS device is characterized, and a signal detection using an OTS device is proposed. Next, we investigate the impact of stochasticity in the threshold voltage on detecting weak signals in the SR system. Finally, by evaluating the inference performance of the artificial neural network, we confirm that the inherent stochasticity can effectively restore the degraded MNIST image in poor visibility conditions in the OTS device. As a result, the recognition accuracy was improved from 10.28% to 95.78% when the stochasticity characteristic was reflected. These results show that stochasticity in the device can improve system performance.

Background: Congenital diaphragmatic hernia (CDH) is a rare disease often requiring mechanical ventilation after birth. In severe cases, extracorporeal membrane oxygenation (ECMO) may be needed. This study analyzed the outcomes of patients with CDH treated with ECMO and investigated factors related to in-hospital mortality. Methods: Among 254 newborns diagnosed with CDH between 2008 and 2020, 51 patients needed ECMO support. At Asan Medical Center, a multidisciplinary team approach has been applied for managing newborns with CDH since 2018. Outcomes were compared between hospital survivors and nonsurvivors. Results: ECMO was established at a median of 17 hours after birth. The mean birth weight was 3.1±0.5 kg. Twenty-three patients (23/51, 45.1%) were weaned from ECMO, and 16 patients (16/51, 31.4%) survived to discharge. The ECMO mode was veno-venous in 24 patients (47.1%) and veno-arterial in 27 patients (52.9%). Most cannulations (50/51, 98%) were accomplished through a transverse cervical incision. No significant between-group differences in baseline characteristics and prenatal indices were observed. The oxygenation index (1 hour before: 90.0 vs. 51.0, p=0.005) and blood lactate level (peak: 7.9 vs. 5.2 mmol/L, p=0.023) before ECMO were higher in nonsurvivors. Major bleeding during ECMO more frequently occurred in nonsurvivors (57.1% vs. 12.5%, p=0.007). In the multivariate analysis, the oxygenation index measured at 1 hour before ECMO initiation was identified as a significant risk factor for in-hospital mortality (odds ratio, 1.02; 95% confidence interval, 1.01–1.04; p=0.05). Conclusion: The survival of neonates after ECMO for CDH is suboptimal. Timely application of ECMO is crucial for better survival outcomes.

Coronary artery bypass grafting (CABG) is rarely performed in infants because of its technical difficulty and unclear long-term results. A 90-day-old male infant weighing 3.5 kg who underwent an arterial switch operation (ASO) for transposition of the great arteries developed left coronary artery insufficiency despite augmentation and reimplantation of the left coronary button. On-pump beating heart CABG was performed using an internal mammary artery graft to revascularize the left anterior descending artery. Postoperative computed tomography angiography revealed that the graft was patent. At 7 months postoperatively, the patient weighed 8.5 kg, and echocardiography revealed good ventricular function. CABG can be an alternative treatment for post-ASO coronary complications in early infancy.

Hardware neural network (HNN) based on analog synapse array excels in accelerating parallel computations. To implement an energy-efficient HNN with high accuracy, high-precision synaptic devices and fully-parallel array operations are essential. However, existing resistive memory (RRAM) devices can represent only a finite number of conductance states. Recently, there have been attempts to compensate device nonidealities using multiple devices per weight. While there is a benefit, it is difficult to apply the existing parallel updating scheme to the synaptic units, which significantly increases updating process’s cost in terms of computation speed, energy, and complexity. Here, we propose an RRAM-based hybrid synaptic unit consisting of a “big” synapse and a “small” synapse, and a related training method. Unlike previous attempts, array-wise fully-parallel learning is possible with our proposed architecture with a simple array selection logic. To experimentally verify the hybrid synapse, we exploit Mo/TiOx RRAM, which shows promising synaptic properties and areal dependency of conductance precision. By realizing the intrinsic gain via proportionally scaled device area, we show that the big and small synapse can be implemented at the device-level without modifications to the operational scheme. Through neural network simulations, we confirm that RRAM-based hybrid synapse with the proposed learning method achieves maximum accuracy of 97 %, comparable to floating-point implementation (97.92%) of the software even with only 50 conductance states in each device. Our results promise training efficiency and inference accuracy by using existing RRAM devices.

Abstract We report a simple method for electrospinning polymers onto flexible, insulating substrates by controlling the wettability of the substrate surface. Water molecules were adsorbed onto the surface of a hydrophilic polymer substrate by increasing the local humidity around the substrate. The adsorbed water was used as the ground electrode for electrospinning. The electrospun fibers were deposited only onto hydrophilic areas of the substrate, allowing for patterning through wettability control. Direct writing of polymer fiber was also possible through near-field electrospinning onto a hydrophilic surface.

Electric fields are widely used for controlling liquids in various research fields. To control a liquid, an alternating current (AC) electric field can offer unique advantages over a direct current (DC) electric field, such as fast and programmable flows and reduced side effects, namely the generation of gas bubbles. Here, we demonstrate one-directional flow along carbon nanotube nanowires under an AC electric field, with no additional equipment or frequency matching. This phenomenon has the following characteristics: First, the flow rates of the transported liquid were changed by altering the frequency showing Gaussian behaviour. Second, a particular frequency generated maximum liquid flow. Third, flow rates with an AC electric field (approximately nanolitre per minute) were much faster than those of a DC electric field (approximately picolitre per minute). Fourth, the flow rates could be controlled by changing the applied voltage, frequency, ion concentration of the solution and offset voltage. Our finding of microfluidic control using an AC electric field could provide a new method for controlling liquids in various research fields.

Electroosmotic flow (EOF) is one of the most important techniques in a microfluidic system. Many microfluidic devices are made from a combination of different materials, and thus asymmetric electrochemical boundary conditions should be applied for the reasonable analysis of the EOF. In this study, the EOF of power-law fluids in a slit microchannel with different zeta potentials at the top and bottom walls are studied analytically. The flow is assumed to be steady, fully developed, and unidirectional with no applied pressure. The continuity equation, the Cauchy momentum equation, and the linearized Poisson-Boltzmann equation are solved for the velocity field. The exact solutions of the velocity distribution are obtained in terms of the Appell’s first hypergeometric functions. The velocity distributions are investigated and discussed as a function of the fluid behavior index, Debye length, and the difference in the zeta potential between the top and bottom.

Electroosmotic flow (EOF) is widely used in microfluidic systems and chemical analysis. It is driven by an electric force inside microchannel with highly charged boundary conditions. In practical applications, electrochemical boundary conditions are often inhomogeneous because different materials as walls are commonly utilized in routine fabrication methods. In the present study, we focus on the analytic solutions of the EOF generated in a planar microchannel with asymmetric electrochemical boundary conditions for non-Newtonian fluids. The velocity profile and flow rate are approximated by employing the power-law model of fluids in the Cauchy momentum equation. The hydrodynamic features of the EOF under asymmetric zeta potentials are scrutinized as a function of the fluid behavior index of the power-law fluid, thickness of Debye length, and zeta potential ratios between planes. The approximate solutions of the power-law model are comparable to the numerically obtained solutions when the Debye length is small and the fluid behavior index is close to unity. This study provides insights into the electrical control of non-Newtonian fluids, such as biological materials of blood, saliva, and DNA solution, in lab-on-a-chip devices.

In this paper, a systematic study of a fully developed electroosmotic flow of power-law fluids in a rectangular microchannel bounded by walls with different zeta potentials is described. Because the upper and lower layers of most microchannels are made of different materials, it is necessary to study the flow characteristics for cases in which the microchannels have different zeta potentials at each wall. The electrical potential and momentum equations were solved numerically using a finite element analysis. The velocity profiles and flow rates were studied parametrically by varying the fluid behavior index, channel aspect ratio, and electrochemical properties of the liquid and the bounding walls. The calculated volumetric flow rates in a rectangular microchannel were compared with those between two infinite parallel plates.

Objective: To examine the effect of mobile neurofeedback training on the clinical symptoms, attention abilities, and execution functions of children with attention deficit hyperactivity disorder (ADHD). Methods: The participants were 74 children with ADHD aged 8−15 years who visited the Department of Child and Adolescent Psychiatry at Seoul National University Children’s Hospital. The participants were randomly assigned to the mobile neurofeedback (n = 35) or control (sham; n = 39) group. Neurofeedback training was administered using a mobile app (equipped with a headset with a 2-channel electroencephalogram [EEG] sensor) for 30 min/day, 3 days/week, for 3 months. Children with ADHD were individually administered various neuropsychological tests, including the continuous performance test, Children’s Color Trails Test-1 and 2, and Stroop Color and Word Tests. The effects of mobile neurofeedback were evaluated at baseline and at 3 and 6 months after treatment initiation. Results: Following treatment, both mobile neurofeedback-only and sham-only groups showed significant improvements in attention and response inhibition. In the visual continuous performance test, omission errors decreased to the normal range in the mobile neurofeedback-only group after training, suggesting that mobile neurofeedback effectively reduced inattention in children with ADHD. In the advanced test of attention, auditory response times decreased in the mobile neurofeedback + medication group after training, but increased in the sham+medication group. Overall, there were no significant between-group differences in other performance outcomes. Conclusion: Mobile neurofeedback may have potential as an additional therapeutic option alongside medication for children with ADHD. [ABSTRACT FROM AUTHOR]

This study examines the determinants and consequences of seasoned equity offering (SEO) issuers' decisions to bias their management earnings forecasts before SEOs in Japan, where management forecasts are mandatory. We identify firm characteristics associated with incentives to inflate management forecasts before SEOs: (i) the use of fund to pay debt, (ii) the economic significance of SEO proceeds, and (iii) financial distress. We also find that the effect of those firm characteristics on forecast error can be reduced by strong bank relationships. In addition, we demonstrate that optimism in management forecasts prior to SEOs is associated with post-SEO underperformance. [ABSTRACT FROM AUTHOR]