On-Chip Training Spiking Neural Networks Using Approximated Backpropagation With Analog Synaptic Devices.

Hardware-based spiking neural networks (SNNs) inspired by a biological nervous system are regarded as an innovative computing system with very low power consumption and massively parallel operation. To train SNNs with supervision, we propose an efficient on-chip training scheme approximating backpropagation algorithm suitable for hardware implementation. We show that the accuracy of the proposed scheme for SNNs is close to that of conventional artificial neural networks (ANNs) by using the stochastic characteristics of neurons. In a hardware configuration, gated Schottky diodes (GSDs) are used as synaptic devices, which have a saturated current with respect to the input voltage. We design the SNN system by using the proposed on-chip training scheme with the GSDs, which can update their conductance in parallel to speed up the overall system. The performance of the on-chip training SNN system is validated through MNIST data set classification based on network size and total time step. The SNN systems achieve accuracy of 97.83% with 1 hidden layer and 98.44% with 4 hidden layers in fully connected neural networks. We then evaluate the effect of non-linearity and asymmetry of conductance response for long-term potentiation (LTP) and long-term depression (LTD) on the performance of the on-chip training SNN system. In addition, the impact of device variations on the performance of the on-chip training SNN system is evaluated.

Implementation of Synaptic Device Using Various High-k Gate Dielectric Stacks.

We investigate the characteristics of short-term and long-term synaptic plasticity in a Si-based fieldeffect transistor (FET)-type memory device. An Al2O3/HfO2/Si3N4/SiO2 gate dielectric stack is used to realize short-term and long-term plasticity (STP/LTP). Si3N4 and HfO2 layers are designed to charge trap layer for synaptic device. The mechanism of STP and LTP operation is analyzed by considering the device response to the potentiation and depression pulses and retention measurement of the memory functionality. To investigate the STP operation, paired pulse facilitation (PPF) measurement is performed. The retention characteristic is also studied to validate the LTP property of the device. By investigating a device with an Al2O3/HfO2/Si3N4 stack as a control device, it is shown that the Al2O3/HfO2/Si3N4/SiO2 stack device is suitable for a synaptic device in neuromorphic systems.

Emerging memory technologies for neuromorphic computing.

In this paper, we reviewed the recent trends on neuromorphic computing using emerging memory technologies. Two representative learning algorithms used to implement a hardware-based neural network are described as a bio-inspired learning algorithm and software-based learning algorithm, in particular back-propagation. The requirements of the synaptic device to apply each algorithm were analyzed. Then, we reviewed the research trends of synaptic devices to implement an artificial neural network.

Development of a Dysphagia Screening Test for Preterm Infants (DST-PI).

OBJECTIVE: To explore both the early prediction and diagnosis of dysphagia in preterm infants as an important developmental aspect as well as the prevention of respiratory complications, we developed the simple and-easy-to-apply Dysphagia Screening Test for Preterm Infants (DST-PI) to predict supraglottic penetration and subglottic aspiration. METHODS: Fifty-two infants were enrolled in a videofluoroscopic swallowing study (VFSS) due to clinical suspicions of dysphagia. Thirteen items related to supraglottic penetration or subglottic aspiration were initially selected from previous studies for the DST-PI. Finally, 7 items were determined by linear logistic regression analysis. Cutoff values, sensitivity, specificity, and the area under the ROC curve (AUC) of the DST-PI for predicting supraglottic penetration or subglottic aspiration were calculated using a ROC curve. For inter-rater reliability, the kappa coefficient was calculated. RESULTS: Seven items were selected: 'gestational age,' 'history of apnea,' 'history of cyanosis during feeding,' 'swallowing pattern,' 'coughs during or after feeding,' 'decreased oxygen saturation within 3 minutes of feeding,' and 'voice change after feeding.' The Spearman correlation coefficient between the DST-PI and the penetration-aspiration scale (PAS) was 0.807 (p<0.001). The sensitivity and specificity at different cutoff values for detecting supraglottic penetration and subglottic aspiration were 96.6% and 76.9% at 3.25, and 88.9% and 75.8% at 6.25, respectively. CONCLUSION: The DST-PI is a valid and reliable dysphagia screening test for supraglottic penetration or subglottic aspiration in preterm infants that is easy to apply in a clinical context.

Persistent cauda equina syndrome after caudal epidural injection under severe spinal stenosis: a case report.

Caudal epidural injection (CEI) is one of the most common treatments for low-back pain with sciatica. CEI rarely leads to neurologic complications. We report a case of persistent cauda equina syndrome after CEI. A 44-year-old male patient with severe L4 and L5 spinal ste-nosis underwent CEI for low-back pain and sciatica. The CEI solution consisted of bupivacaine, hyaluronidase, triamcinolone acetonide, and normal saline. He experienced motor weakness and sensory loss in both lower extremities and neurogenic bladder for more than 1 year after the procedure. His ankle dorsiflexors, big-toe extensors, and ankle plantar flexors on both sides were checked and categorized as motor-power Medical Research Council grade 0. His bilateral ankle-jerk reflection was absent. An electrophysiological study showed lumbosacral polyradiculopathy affecting both sides of the L5 and S1 nerve roots. A urodynamic study revealed hypoactive neurogenic bladder affecting both sacral roots.

Lemierre syndrome: a case of postanginal sepsis.

Lemierre syndrome is a rare disease that's characterized by internal jugular vein thrombosis and septic emboli. These symptoms typically develop after acute oropharyngeal infection by Fusobacterium necrophorum. Although this syndrome is less frequently seen in modern times due to the availability of antibiotics, physicians must be aware of the syndrome in order to initiate prompt antibiotics therapy, including coverage of the anerobic organisms. We discuss here the case of an 18-year-old female with Lemierre syndrome and we review the relevant literature on this syndrome.

Effect of pinitol on glucose metabolism and adipocytokines in uncontrolled type 2 diabetes.

Pinitol (3-O-methyl-D-chiro-inositol) was identified in putative insulin mediator fractions that have hypoglycemic activity, and appears to mimic the act effects of insulin by acting downstream in the insulin signaling pathway. We evaluated the effect of pinitol therapy in type 2 diabetic patients who were poorly controlled with hypoglycemic drugs, such as sulfonylurea, metformin and/or insulin. Twenty type 2 diabetic patients were enrolled in our study. Fasting glucose, fasting c-peptide, total cholesterol, triglyceride, and HDL- and LDL-cholesterol were checked before and after a 12-week pinitol treatment (20 mg kg(-1)day(-1)). All subjects continued their current medications during the study. Adipocytokines, such as adiponectin, leptin, free fatty acids, and c-reactive protein (CRP) were checked before and after pinitol treatment. After pinitol treatment, fasting glucose, post-prandial glucose levels, and hemoglobin A1c were significantly decreased (P<0.05). Fasting serum adiponectin, leptin, free fatty acid, and CRP levels did not change after pinitol treatment. In the unresponsive group, serum c-peptide levels were higher than in the responsive group. Twelve weeks of pinitol treatment altered glucose metabolism, but not lipid profiles or adipocytokine levels, in type 2 diabetic patients. Additional research is needed to define the physiological and potential therapeutic effects of pinitol administration.