• Journal of the Korea Computer Graphics Society
• /
• v.27 no.4
• /
• pp.1-9
• /
• 2021

Segmenting the compartments of neurons, such as axons, dendrites, and cell bodies, is helpful in the analysis of neurological phenomena. Recently, there have been several studies to segment the compartments through deep learning. However, this approach has the potential to include errors in the results due to noise in data and differences between training data and actual data. Therefore, in order to use these for actual analysis, it is essential to proofread the results. The proofreading process requires a lot of effort and time because an expert must perform it manually. We propose an interactive neuron structure proofreading method that can more easily correct errors in the segmentation results of a deep learning. This method proofread the neuron structure based on the characteristics of the neuron with structural consistency, so that a high-accuracy proofreading result can be obtained with less interaction.

• Annals of Clinical Neurophysiology
• /
• v.18 no.1
• /
• pp.18-20
• /
• 2016

Amyotrophic lateral sclerosis (ALS) is a progressive motor neuron degenerative disease that clinically manifests both upper and lower motor neuron signs. However, it is unknown where and how the motor neuron degeneration begins, and conflicting hypotheses have been suggested. Recent advanced radiological techniques enable us to look into ALS neuropathology in vivo. Herein, we report a case with upper motor neuron-predominant ALS in whom the results of brain magnetic resonance imaging (MRI) and myelin water fraction MRI suggest axonal degeneration.

• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1406-1420
• /
• 2017

Artificial neural networks (ANNs) play an important role in the fields of function approximation, prediction, and classification. ANN performance is critically dependent on the input parameters, including the number of neurons in each layer, and the optimal values of weights and biases assigned to each neuron. In this study, we apply the particle swarm optimization method, a popular optimization algorithm for determining the optimal values of weights and biases for every neuron in different layers of the ANN. Several regression models, including general linear regression, Fourier regression, smoothing spline, and polynomial regression, are conducted to evaluate the proposed method's prediction power compared to multiple linear regression (MLR) methods. In addition, residual analysis is conducted to evaluate the optimized ANN accuracy for both training and test datasets. The experimental results demonstrate that the proposed method can effectively determine optimal values for neuron weights and biases, and high accuracy results are obtained for prediction applications. Evaluations of the proposed method reveal that it can be used for prediction and estimation purposes, with a high accuracy ratio, and the designed model provides a reliable technique for optimization. The simulation results show that the optimized ANN exhibits superior performance to MLR for prediction purposes.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.1
• /
• pp.52-58
• /
• 2005

We have developed a Group Method of Data Handling (GMDH) type algorithm for designing multi-layered neural networks. The algorithm is general enough that it will accept any number of inputs and any sized training set. Each neuron of the resulting network is a function of two of the inputs to the layer. The equation for each of the neurons is a quadratic polynomial. Several forms of the equation are tested for each neuron to make sure that only the best equation of two inputs is kept. All possible combinations of two inputs to each layer are also tested. By carefully testing each resulting neuron, we have developed an algorithm to keep only the best neurons at each level. The algorithm's goal is to create as accurate a network as possible while minimizing the size of the network. Software was developed to train and simulate networks using our algorithm. Several applications were modeled using our software, and the result was that our algorithm succeeded in developing small, accurate, multi-layer networks.

• Journal of Intelligence and Information Systems
• /
• v.11 no.2
• /
• pp.81-103
• /
• 2005

The Neuron structure in a nervous system consists of inhibitory neurons and excitory neurons. Both neurons are activated by agonistic neurons and inactivated by antagonist neurons. In this paper, we proposed a physiological fuzzy neural network by analyzing the physiological neuron structure in the nervous system. The proposed structure selectively activates the neurons which go through a state of excitement caused by agonistic neurons and also transmit the signal of these neurons to the output layers. The proposed physiological fuzzy neural networks based on the nervous system consists of a input player, and the hidden layer which classifies features of learning data, and output layer. The proposed fuzzy neural network is applied to recognize bronchial squamous cell carcinoma images and car plate images. The result of the experiments shows that the learning time, the convergence, and the recognition rate of the proposed physiological fuzzy neural networks outperform the conventional neural networks.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.357-358
• /
• 2023

본 연구는 실제 행동과 운동 심상으로 팔과 다리 동작 인식을 위한 BCI 패러다임을 제안하고 유도성 분석을 한다. 이 페러다임은 각 팔과 양다리의 특정 움직임을 인식하기 위해 ERP를 기반 페러다임을 구성한다. BCI 페러다임은 왼팔, 오른팔, 양다리를 움직이는 영상 자극을 주며 이를 기반으로 왼팔, 오른팔, 양다리 움직임에 대한 인식을 한다. 거울뉴런은 실제 행동과 실제 행동을 보았을때와 운동심상을 통한 자극을 받았을 때 같은 뉴런이 활성화된다는 성질을 가지고 있다. 이러한 성질을 이용하여 운동심상만과 실제 행동을 동시에 학습할 경우를 유도성 분석을 진행한다. 또한 유도성 특징 분석을 통해 나타난 결과를 바탕으로 BCI 패러다임을 제안한다.

• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.104-112
• /
• 2019

This study proposed a robust detection algorithm. It detects face more stably with respect to changes in light and rotation forthe identification of a face shape. The proposed algorithm uses face shape asinput information in a single camera environment and divides only face area through preprocessing process. However, it is not easy to accurately recognize the face area that is sensitive to lighting changes and has a large degree of freedom, and the error range is large. In this paper, we separated the background and face area using the brightness difference of the two images to increase the recognition rate. The brightness difference between the two images means the difference between the images taken under the bright light and the images taken under the dark light. After separating only the face region, the face shape is recognized by using the self-organization feature map (SOFM) algorithm. SOFM first selects the first top neuron through the learning process. Second, the highest neuron is renewed by competing again between the highest neuron and neighboring neurons through the competition process. Third, the final top neuron is selected by repeating the learning process and the competition process. In addition, the competition will go through a three-step learning process to ensure that the top neurons are updated well among neurons. By using these SOFM neural network algorithms, we intend to implement a stable and robust real-time face shape recognition system in face shape recognition.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.1
• /
• pp.62-68
• /
• 2012

Steady-State Visually Evoked Potentials (SSVEP) are natural response signal associated with the visual stimuli with specific frequency. By using SSVEP, occipital lobe region is electrically activated as frequency form equivalent to stimuli frequency with bandwidth from 3.5Hz to 75Hz. In this paper, we propose an experimental paradigm for analyzing EEGs based on the properties of SSVEP. At first, an experiment is performed to extract frequency feature of EEGs that is measured from the image-based visual stimuli associated with specific objective with affordance and object-related affordance is measured by using mirror neuron system based on the frequency feature. And then, linear discriminant analysis (LDA) method is applied to perform the online classification of the objective pattern associated with the EEG-based affordance data. By using the SSVEP measurement experiment, we propose a Brain-Computer Interface (BCI) system for recognizing user's inherent intentions. The existing SSVEP application system, such as speller, is able to classify the EEG pattern based on grid image patterns and their variations. However, our proposed SSVEP-based BCI system performs object pattern classification based on the matters with a variety of shapes in input images and has higher generality than existing system.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.17 no.2
• /
• pp.210-215
• /
• 2017

We have developed the neuromorphic system that can work with the four-terminal Si-based synaptic devices and verified the operation of the system using simulation tool and printed-circuit-board (PCB). The symmetrical current mirrors connected to the n-channel and p-channel synaptic devices constitute the synaptic integration part to express the excitation and the inhibition mechanism of neurons, respectively. The number and the weight of the synaptic devices affect the amount of the current reproduced from the current mirror. The double-stage inverters controlling delay time and the NMOS with large threshold voltage ($V_T$) constitute the action-potential generation part. The generated action-potential is transmitted to next neuron and simultaneously returned to the back gate of the synaptic device for changing its weight based on spike-timing-dependent-plasticity (STDP).

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.