• Interdisciplinary Bio Central
• /
• 제5권2호
• /
• pp.3.1-3.7
• /
• 2013

Recently, the productivity of drug discovery has gradually decreased as the limitations of single-target-based drugs for various and complex diseases become exposed. To overcome these limitations, drug combinations have been proposed, and great efforts have been made to predict efficacious drug combinations by statistical methods using drug databases. However, previous methods which did not take into account biological networks are insufficient for elaborate predictions. Also, increased evidences to support the fact that drug effects are closely related to metabolic enzymes suggested the possibility for a new approach to the study drug combinations. Therefore, in this paper we suggest a novel approach for analyzing drug combinations using a metabolic network in a systematic manner. The influence of a drug on the metabolic network is described using the distance between the drug target and an enzyme. Target-enzyme distances are converted into influence scores, and from these scores we calculated the correlations between drugs. The result shows that the influence score derived from the targetenzyme distance reflects the mechanism of drug action onto the metabolic network properly. In an analysis of the correlation score distribution, efficacious drug combinations tended to have low correlation scores, and this tendency corresponded to the known properties of the drug combinations. These facts suggest that our approach is useful for prediction drug combinations with an advanced understanding of drug mechanisms.

• 대한의용생체공학회:의공학회지
• /
• 제16권2호
• /
• pp.157-166
• /
• 1995

The Electroencephalogram (EEG) and evoked potential (EP) t;ave widely been used for study of brain functions. The EEG and EP signals acquired from multi-channel electrodes placed on the head surface are often interfered by other relatively large physiological signals such as electromyogram (EMG) or electroculogram (EOG). Since these artifact-affected EEG signals degrade EEG mapping, the removal of the artifact-affected EEGs is one of the key elements in neuro-functional mapping. Conventionally this task has been carried out by human experts spending lots of examination time. In this paper a neural-network based classification is proposed to replace or to reduce human expert's efforts and time. From experiments, the neural-network based classification performs as good as human experts : variation of decisions between the neural network and human expert appears even smaller than that between human experts.

• Smart Structures and Systems
• /
• 제5권4호
• /
• pp.415-426
• /
• 2009

This study focuses on the concept of multi-scale wireless sensor networks for damage detection in civil infrastructure systems by first over viewing the general network philosophy and attributes in the areas of data acquisition, data reduction, assessment and decision making. The data acquisition aspect includes a scalable wireless sensor network acquiring acceleration and strain data, triggered using a Restricted Input Network Activation scheme (RINAS) that extends network lifetime and reduces the size of the requisite undamaged reference pool. Major emphasis is given in this study to data reduction and assessment aspects that enable a decentralized approach operating within the hardware and power constraints of wireless sensor networks to avoid issues associated with packet loss, synchronization and latency. After over viewing various models for data reduction, the concept of a data-driven Bivariate Regressive Adaptive INdex (BRAIN) for damage detection is presented. Subsequent examples using experimental and simulated data verify two major hypotheses related to the BRAIN concept: (i) data-driven damage metrics are more robust and reliable than their counterparts and (ii) the use of heterogeneous sensing enhances overall detection capability of such data-driven damage metrics.

• International Journal of Computer Science & Network Security
• /
• 제23권5호
• /
• pp.73-88
• /
• 2023

Brain tumors can also be an abnormal collection or accumulation of cells in the brain that can be life-threatening due to their ability to invade and metastasize to nearby tissues. Accurate diagnosis is critical to the success of treatment planning, and resonant imaging is the primary diagnostic imaging method used to diagnose brain tumors and their extent. Deep learning methods for computer vision applications have shown significant improvements in recent years, primarily due to the undeniable fact that there is a large amount of data on the market to teach models. Therefore, improvements within the model architecture perform better approximations in the monitored configuration. Tumor classification using these deep learning techniques has made great strides by providing reliable, annotated open data sets. Reduce computational effort and learn specific spatial and temporal relationships. This white paper describes transfer models such as the MobileNet model, VGG19 model, InceptionResNetV2 model, Inception model, and DenseNet201 model. The model uses three different optimizers, Adam, SGD, and RMSprop. Finally, the pre-trained MobileNet with RMSprop optimizer is the best model in this paper, with 0.995 accuracies, 0.99 sensitivity, and 1.00 specificity, while at the same time having the lowest computational cost.

• The Journal of Korean Physical Therapy
• /
• 제13권3호
• /
• pp.791-797
• /
• 2001

After brain injury, patients show a wide range in the degree of recovery. By a variety of mechanisms, the human brain is constantly undergoing plastic changes. Spontaneous recovery from brain injury in the chronic stage omes about because of plasticity. The brain regions are altered. resulting in functionally modified cortical network. This review cnsidered the neural plasticity from cellular and molecular mechanisms of synapse formation to behavioural recovery from brain injury in elderly humans. The stimuli required to elicit plasticity are thought to be activity-dependent elements. especially exercise and learning. Knowledge about the physiology of brain plasticity has led to the development of methods for rehabilitation.

• 로봇학회논문지
• /
• 제1권2호
• /
• pp.163-171
• /
• 2006

This paper introduces the research progress on the artificial brain in the Telerobotics and Control Laboratory at KAIST. This series of studies is based on the assumption that it will be possible to develop an artificial intelligence by copying the mechanisms of the animal brain. Two important brain mechanisms are considered: spike-timing dependent plasticity and dopaminergic plasticity. Each mechanism is implemented in two coding paradigms: spike-codes and rate-codes. Spike-timing dependent plasticity is essential for self-organization in the brain. Dopamine neurons deliver reward signals and modify the synaptic efficacies in order to maximize the predicted reward. This paper addresses how artificial intelligence can emerge by the synergy between self-organization and reinforcement learning. For implementation issues, the rate codes of the brain mechanisms are developed to calculate the neuron dynamics efficiently.

• 응용통계연구
• /
• 제36권5호
• /
• pp.361-380
• /
• 2023

Magnetoencephalography (MEG)는 뉴론 활동에 신경 세포들간 전류 흐름에 의해 유도된 자기장을 측정하는 비침습 뇌영상 기술이다. 기능적 뇌활동은 뇌영역간 또는 뉴런들의 연결로 기능적 연결로 수행된다. MEG 데이터는 상관성, 시공간성을 가지며 다중 다층적 동적 네트워크인 특징을 갖는다. 이러한 복잡성 때문에 MEG 네트워크에 대한 연구는 아직 많지 않은 편이다. 본 연구에서는 MEG 네트워크 모형과 분석법을 소개하고 실제 MEG 데이터 분석에 활용되어 해석된 경우를 요약하고 앞으로 MEG 네트워크 모형 개발 연구의 필요성을 설명하고자 한다. 그러므로 통계적 네트워크 분석이 뇌과학에서 신경학적 질병을 포함하여 뇌기능에 대한 이해에 중요한 역할을 할 수 있음을 알리고자 한다.

• 스마트미디어저널
• /
• 제9권3호
• /
• pp.9-17
• /
• 2020

Hippocampus is a complex brain structure embedded deep into the temporal lobe. Studies have shown that this structure gets affected by neurological and psychiatric disorders and it is a significant landmark for diagnosing neurodegenerative diseases. Hippocampus features play very significant roles in region-of-interest based analysis for disease diagnosis and prognosis. In this study, we have attempted to learn the embeddings of this important biomarker. As conventional metric learning methods for feature embedding is known to lacking in capturing semantic similarity among the data under study, we have trained deep Siamese convolutional neural network for learning metric of the hippocampus. We have exploited Gwangju Alzheimer's and Related Dementia cohort data set in our study. The input to the network was pairs of three-view patches (TVPs) of size 32 × 32 × 3. The positive samples were taken from the vicinity of a specified landmark for the hippocampus and negative samples were taken from random locations of the brain excluding hippocampi regions. We have achieved 98.72% accuracy in verifying hippocampus TVPs.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1990년도 하계학술대회 논문집
• /
• pp.143-146
• /
• 1990

The human brain has the most powerful capabilities in thinking, interpreting, remembering, and problem-solving. Artificial neural network is appeared by scientists who have tried to simulate such a human brain. The artificial neural network has the capability of learning, massive parallelism capability and robustness for disturbance which are necessary for power system application. In this paper, We reviewed the typical topologies and learning algorithms of artifical neural networks which can be used for pattern classification. And we surveyed for the applications of artifical neural network to the power system.

• 한국멀티미디어학회논문지
• /
• 제24권4호
• /
• pp.519-527
• /
• 2021

Alzheimer's disease is one of the challenges to tackle in the coming aging era and is attempting to diagnose and predict through various biomarkers. While the application of various deep learning-based technologies as powerful imaging technologies has recently expanded across the medical industry, empirical design is not easy because there are various deep earning neural networks architecture and categorical hyperparameters that rely on problems and data to solve. In this paper, we show the possibility of optimizing a deep learning neural network structure and hyperparameters for Alzheimer's disease classification in amyloid brain images in a representative deep earning neural networks architecture using genetic algorithms. It was observed that the optimal deep learning neural network structure and hyperparameter were chosen as the values of the experiment were converging.