• 문화기술의 융합
• /
• 제6권3호
• /
• pp.463-468
• /
• 2020

인공지능 연구는 다양한 분야에 적용되며 발전하고 있다. 본 논문에서는 차세대 인공지능 연구 분야인 SNN(Spiking Neural Networks) 형태의 인공지능 구현 방식을 사용하여 신경망을 구축하고, 그 신경망에서 뉴런의 개수가 신경망의 성능에 어떠한 영향을 미치는지를 분석한다. 또한 신경망 학습량을 증가시키면서 신경망의 성능이 어떻게 바뀌는지를 분석한다. 해당 연구 결과를 통해 각 분야에서 사용되는 SNN 기반의 신경망을 최적화 할 수 있을 것이다.

• 한국연초학회지
• /
• 제29권2호
• /
• pp.118-124
• /
• 2007

GRLs(Guidance Residue Levels) of agricultural chemicals for tobacco are recommended by the CORESTA Agro-Chemical Advisory Committee guide. In the GRLs list, organochlorine group is one of pesticides commonly used on tobacco cultivation. In this model study, the quantitative correlation in the transfer rate of pesticide residue into tobacco smoke by spiking of organochlorine pesticides to cigarette and pyrolysates were investigated. The spiking concentration referred to the range of GRLs list and the organochlorine pesticides in mainstream smoke were analyzed by GC-MS. For the understanding of the composition variation versus temperature, the behavior of pesticides was investigated by pyrolysis-gas chromatography-mass spectrometry(Py-GC-MS). In this study, the transfer rate of pesticide residue into tobacco smoke at four different spiking concentration and the composition of pyrolysates were analyzed differently. At $10\;{\mu}g/cig$ spiking concentrations, the organochlorine pesticides were transferred into tobacco smoke in $0.02\;{\sim}\;10.19\;%$ each of component and the most of pesticides were pyrolyzed during smoking. It was found that the decomposition compounds from organochlorine pesticides were mainly composed of oxygenated and nitrogenous compounds. This study could estimate that the transfer rate of pesticides into tobacco smoke is very small amount.

• Tuberculosis and Respiratory Diseases
• /
• 제41권3호
• /
• pp.302-306
• /
• 1994

A 33-year old male was admitted due to continuous high spiking fever for 2 months via local clinic. He had been diagnosed pulmonary tuberculosis at local clinic. However, spiking fever had not been controlled by anti-tuberculous medications. Chest PA showed confluent consolidation on right upper & mid-lung field. 5 anti-tuberculous regimens(Streptomycin, Isoniazid, Rifampin, Ethambutol, Pyrazinamaide) were administered initially and steroid therapy was followed for relieving toxic symptoms Very slowly resolved chest X-ray lesion and continuous fever suggested the possibility of misdiagnosis. After 60th hospital day, the chest X-ray lesion was resolved gradually and fever subsided almost completely. He was discharged on 76th hospital day with anti-tuberculous drugs and steroid(prednisolon), without any other problems except sustained mild fever.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권9호
• /
• pp.2846-2866
• /
• 2022

Caenorhabditis elegans exhibits sophisticated chemotaxis behavior through two parallel strategies, klinokinesis and klinotaxis, executed entirely by a small nervous circuit. It is therefore suitable for inspiring fast and energy-efficient solutions for autonomous navigation. As a random search strategy, the Lévy walk is optimal for diverse animals when foraging without external chemical cues. In this study, by combining these biological strategies for the first time, we propose a spiking neural network model for search and contour tracking of specific concentrations of environmental variables. Specifically, we first design a klinotaxis module using spiking neurons. This module works in conjunction with a klinokinesis module, allowing rapid searches for the concentration setpoint and subsequent contour tracking with small deviations. Second, we build a random exploration module. It generates a Lévy walk in the absence of concentration gradients, increasing the chance of encountering gradients. Third, considering local extrema traps, we develop a termination module combined with an escape module to initiate or terminate the escape in a timely manner. Experimental results demonstrate that the proposed model integrating these modules can switch strategies autonomously according to the information from a single sensor and control steering through output spikes, enabling the model worm to efficiently navigate across various scenarios.

• The Korean Journal of Physiology and Pharmacology
• /
• 제7권6호
• /
• pp.317-323
• /
• 2003

Interneuron diversity is one of the key factors to hinder understanding the mechanism of cortical neural network functions even with their important roles. We characterized inhibitory interneurons in layer II/III of the rat primary visual cortex, using patch-clamp recording and confocal reconstruction, and classified inhibitory interneurons into fast spiking (FS), late spiking (LS), burst spiking (BS), and regular spiking non-pyramidal (RSNP) neurons according to their electrophysiological characteristics. Global parameters to identify inhibitory interneurons were resting membrane potential (＞-70 mV) and action potential (AP) width (＜0.9 msec at half amplitude). FS could be differentiated from LS, based on smaller amplitude of the AP (＜∼50 mV) and shorter peak-to-trough time (P-T time) of the afterhyperpolarization (＜4 msec). In addition to the shorter AP width, RSNP had the higher input resistance (＞200 $M{Omega}$) and the shorter P-T time (＜20 msec) than those of regular spiking pyramidal neurons. Confocal reconstruction of recorded cells revealed characteristic morphology of each subtype of inhibitory interneurons. Thus, our results provide at least four subtypes of inhibitory interneurons in layer II/III of the rat primary visual cortex and a classification scheme of inhibitory interneurons.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1992-1995
• /
• 2002

This paper presents a novel 4-D chaotic spiking oscillator. The oscillator can generate hyperchaos characterized by two positive Lyapunov exponents. Us-ing a simple test circuit, typical phenomena can be verified in the laboratory.

• Annals of Clinical Neurophysiology
• /
• 제2권2호
• /
• pp.119-124
• /
• 2000

Background : We studied EEG changes during pilocarpine-induced status epilepticus(SE), a widely used model whose EEG characteristics have not been fully described previously. Methods : Male Sprague-Dawley rats weighing 250-350 grams were used as subjects. SE was induced 5-7 days after placement of chronic epidural electrodes, using 360-380 mg/Kg pilocarpine IP. Rats were observed with continuous EEG recording following pilocarpine injection until end of the SE episode. Results : SE occurred in 10/12 rats studied. SE began with a series of discrete seizures $11.1{\pm}3.93$ minutes after pilocarpine injection. $5.2{\pm}2.71$ seizures occurred over $10.9{\pm}4.62$ minutes, until the EEG converted to a waxing and waning pattern, during which the amplitude and frequency of epileptiform activity increased. After $1.4{\pm}1.82$ minutes, a pattern of continuous high amplitude rapid spiking was established. Continuous spiking continued for $3.4{\pm}0.48$ hours with a very gradual decline in amplitude and frequency, until periodic epileptiform discharges(PEDs) began to occur. The EEG consisted primarily of PEDs for another $7.4{\pm}3.09$ hours, until electrographic generalized seizures began to occur. These continued for $5.8{\pm}4.82$ hours until death. Duration of SE was $17.0{\pm}5.88$ hours. Flat periods were a prominent feature during all EEG patterns in this model. Conclusion : EEG features distinctive in pilocarpine SE(but not unique to it) include flat periods during all patterns and resumption of continuous spiking episodes after the onset of PEDs. The sequence of discrete seizures to waxing and waning to continuous spiking to PEDs was identical to that which has been described in humans and other animal models.

• 문화기술의 융합
• /
• 제8권5호
• /
• pp.613-618
• /
• 2022

SNN(Spiking Neural Networks) 기반의 인공지능 연구는 현재 유행하는 DNN(Deep Neural Networks) 기반의 인공지능의 한계를 극복할 수 있는 차세대 인공지능으로서 주목받고 있다. 본 논문에서는 SNN 형태의 입력을 뉴로모픽 컴퓨팅 시스템에서 구동시킬 수 있는 시스템 SW인 SNN 컴파일러의 구조에 대하여 설명한다. 또한 컴파일러 구현을 위하여 사용된 알고리즘을 소개하고 매핑 알고리즘의 동작 형태에 따라 뉴로모픽 컴퓨팅 시스템에서 수행시간이 어떻게 달라지는지에 대한 실험결과를 제시한다. 본문에서 제안한 매핑 알고리즘은 랜덤 매핑에 비해 최대 3.96배의 수행속도 향상이 있었다. 해당 연구 결과를 통해 SNN들을 다양한 뉴로모픽 하드웨어에서 적용할 수 있을 것이다.

• 대한용접접합학회:학술대회논문집
• /
• 대한용접접합학회 2010년도 춘계학술발표대회 초록집
• /
• pp.93-93
• /
• 2010

환경 규제 및 배출가스 규제에 의하여 차량 경량화를 위해 점차적으로 Al합금의 차체 및 부품적용 비율이 점차 확대되고 있다. 이에 따라 알루미늄의 레이저 용접 시 출력, 초점거리, 용접 속도 등 공정 변수의 상관관계와 용접 결함 현상에 의한 관심이 집중된 연구가 많이 발표되었으며, 알루미늄 5000계열의 경우 박판 용접 시 기공, 균열 등 과 같은 결함 현상을 방지하기 위하여, Twin spot laser, Laser-TIG hybrid 등과 같은 공법 적용을 제안되었다. 본 연구에서는 Yb:YAG laser welding 시 Mg 함량이 높은 AA5J32을 소재를 이용하여 박판 겹치기 용접 시 Back side spiking 결함 방지를 위한 레이저 빔 출력 파형을 설계하여 실험을 수행하였다. 또한 파형의 특성에 따라 나타나는 겹치기 용접부의 기계적 특성과 기공에 대해 알아 보고자 하였다.

• 센서학회지
• /
• 제30권1호
• /
• pp.20-24
• /
• 2021

The electrocardiogram (ECG) is one of the most extensively employed signals used to diagnose and predict cardiovascular diseases (CVDs). In recent years, several deep learning (DL) models have been proposed to improve detection accuracy. Among these, deep neural networks (DNNs) are the most popular, wherein the features are extracted automatically. Despite the increment in classification accuracy, DL models require exorbitant computational resources and power. This causes the mapping of DNNs to be slow; in addition, the mapping is challenging for a wearable device. Embedded systems have constrained power and memory resources. Therefore full-precision DNNs are not easily deployable on devices. To make the neural network faster and more power-efficient, spiking neural networks (SNNs) have been introduced for fewer operations and less complex hardware resources. However, the conventional SNN has low accuracy and high computational cost. Therefore, this paper proposes a new binarized SNN which modifies the synaptic weights of SNN constraining it to be binary (+1 and -1). In the simulation results, this paper compares the DL models and SNNs and evaluates which model is optimal for ECG classification. Although there is a slight compromise in accuracy, the latter proves to be energy-efficient.