• 한국해양공학회지
• /
• 제35권5호
• /
• pp.336-346
• /
• 2021

The prediction of wave conditions is crucial in the field of marine and ocean engineering. Hence, this study aims to predict the significant wave height through machine learning (ML), a soft computing method. The adopted metocean data, collected from 2012 to 2020, were obtained from the Korea Institute of Ocean Science and Technology. We adopted the feedforward neural network (FNN) and long-short term memory (LSTM) models to predict significant wave height. Input parameters for the input layer were selected by Pearson correlation coefficients. To obtain the optimized hyperparameter, we conducted a sensitivity study on the window size, node, layer, and activation function. Finally, the significant wave height was predicted using the FNN and LSTM models, by varying the three input parameters and three window sizes. Accordingly, FNN (W48) (i.e., FNN with window size 48) and LSTM (W48) (i.e., LSTM with window size 48) were superior outcomes. The most suitable model for predicting the significant wave height was FNN(W48) owing to its accuracy and calculation time. If the metocean data were further accumulated, the accuracy of the ML model would have improved, and it will be beneficial to predict added resistance by waves when conducting a sea trial test.

• Journal of Biosystems Engineering
• /
• 제43권4호
• /
• pp.352-361
• /
• 2018

Purpose: The objective of this study was to develop a multilayer perceptron (MLP) model to estimate solar radiation using a solar module. Methods: Data for the short-circuit current of a solar module and other environmental parameters were collected for a year. For MLP learning, 14,400 combinations of input variables, learning rates, activation functions, numbers of layers, and numbers of neurons were trained. The best MLP model employed the batch backpropagation algorithm with all input variables and two hidden layers. Results: The root-mean-squared error (RMSE) of each learning cycle and its average over three repetitions were calculated. The average RMSE of the best artificial neural network model was $48.13W{\cdot}m^{-2}$. This result was better than that obtained for the regression model, for which the RMSE was $66.67W{\cdot}m^{-2}$. Conclusions: It is possible to utilize a solar module as a power source and a sensor to measure solar radiation for an agricultural sensor node.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2014년도 춘계학술대회
• /
• pp.469-472
• /
• 2014

S-MAC은 패킷 교한을 조정하고 idle listening을 줄이기 위해 로컬 sleep-wake 스케쥴을 사용하는 CSMA와 TDMA의 하이브리드 방식이다. 이 기법에서는 모든 노드들이 동일한 우선순위를 가지고 있기 때문에 트래픽의 양이 많은 경우 지연시간이 증가된다. 본 논문에서는 실시간 음성 스트리밍과 같은 어플리케이션에 적합한 처리량과 진송지연을 제공할 수 있는 트래픽 적응적 MAC 프로토콜을 제안한다. 제안된 프로토콜에서는 실시간에 적합한 성능을 제공하기 위해 (m,k)-firm 스케쥴링 기법을 이용한 우선순위 개념을 사용한다. 성능 평가를 위해 다양한 노드 수에 대해 패킷 전송률과 노드의 평균지연시간을 시뮬레이션을 수행했다.

• 한국식품과학회지
• /
• 제38권6호
• /
• pp.829-834
• /
• 2006

HIV 감염 후 AIDS로의 진행과정에서 cytokines의 변화는 필수적인 결과로서 나타나게 된다. 본 연구에서는 AIDS 동물모델을 사용하여 Pyc Cytokines발현에 미치는 영향과 조절 매개체로서 $NF-{\kappa}B$의 관련성 및 작용 기작 확인에 초점을 두었다. LP-BM5 retrovirus에 감염된 C57BL/6 생쥐를 이용하여12주간에 걸친 실험에서Pyc 투여는 Th1과 Th2 cytokines의 혈중농도를 조절하여 murine AIDS로의 진행을 억제하는 데 역할을 하는 것으로 사료된다. 특히 Th2 cytokines의 mRNA 발현을 억제함에 따라 Th1 cytokines의 혈중농도는 상대적으로 증가한 것으로 사료된다. 이러한 결과는 Pyc가 Th2 cytokines을 선택적으로 transcription level에서 조절하고 있음을 의미하고 있다. 또한 Th2 cytokines mRNA 발현 조절은 $NF-{\kappa}B$의 활성화에 의한 것으로 추정된다. 본 연구는 AIDS 동물모델에서 $NF-{\kappa}B$의 조절을 통한 cytokines의 발현 조절의 가능성을 제시함과 동시에 Pyc의 역할에 대해 자료를 제시하고 있다. 또한 향후 murine AIDS 진행 억제제로서의 Pyc의 기작에 대한 일면을 보인 것에 의의를 두고자 한다.

• 한국미생물학회:학술대회논문집
• /
• 한국미생물학회 2008년도 International Meeting of the Microbiological Society of Korea
• /
• pp.23-25
• /
• 2008

Serum complement proteins comprise an important system that is responsible for several innate and adaptive immune defence mechanisms. There were three well described pathways known to lead to the generation of a C3 convertase, which catalyses the proteolysis of complement component C3, and leads to the formation of C3 opsonins (C3b, iC3b and C3d) that fix to bacteria. A pivotal step in the complement pathway is the assembly of a C3 convertase, which digests the C3 complement component to form microbial-binding C3 fragments recognized by leukocytes. The spleen clears microorganisms from the blood. Individuals lacking this organ are more susceptible to Streptococcus pneumoniae. Innate resistance to S. pneumoniae has previously been shown to involve complement components C3 and C4, however this resistance has only a partial requirement for mediators of these three pathways, such as immunoglobulin, factor B and mannose-binding lectin. Therefore it was likely that spleen and complement system provide resistance against blood-borne S. pneumoniae infection through unknown mechanism. To better understand the mechanisms involved, we studied Specific intracellular adhesion molecule-grabbing nonintegrin (SIGN)-R1. SIGN-R1, is a C-type lectin that is expressed at high levels by spleen marginal-zone macrophages and lymph-node macrophages. SIGN-R1 has previously been shown to be the main receptor for bacterial dextrans, as well as for the capsular pneumococcal polysaccharide (CPS) of S. pneumoniae. We examined the specific role of this receptor in the activation of complement. Using a monoclonal antibody that selectively downregulates SIGN-R1 expression in vivo, we show that in response to S. pneumoniae or CPS, SIGN-R1 mediates the immediate proteolysis of C3 and fixation of C3 opsonins to S. pneumoniae or to marginal-zone macrophages that had taken up CPS. These data indicate that SIGN-R1 is largely responsible for the rapid C3 convertase formation induced by S. pneumoniae in the spleen of mice. Also, we found that SIGN-R1 directly binds C1q and that C3 fixation by SIGN-R1 requires C1q and C4 but not factor B or immunoglobulin. Traditionally C3 convertase can be formed by the classical C1q- and immunoglobulin-dependent pathway, the alternative factor-B-dependent pathway and the soluble mannose-binding lectin pathway. Furthermore Conditional SIGN-R1 knockout mice developed deficits in C3 catabolism when given S. pneumoniae or its capsular polysaccharide intravenously. There were marked reductions in proteolysis of serum C3, deposition of C3 on organisms within SIGN-$R1^+$ spleen macrophages, and formation of C3 ligands. The transmembrane lectin SIGN-R1 therefore contributes to innate resistance by an unusual C3 activation pathway. We propose that in the SIGN-R1 mediated complement activation pathway, after binding to polysaccharide, SIGN-R1 captures C1q. SIGN-R1 can then, in association with several other complement proteins including C4, lead to the formation of a C3 convertase and fixation of C3. Therefore, this new pathway for C3 fixation by SIGN-R1, which is unusual as it is a classical C1q-dependent pathway that does not require immuno globulin, contributes to innate immune resistance to certain encapsulated microorganisms.

• 대한한방소아과학회지
• /
• 제24권1호
• /
• pp.9-35
• /
• 2010

Objectives The purpose of this study is to investigate the effect of SPDJTK(SoPungDoJeokTangKami) and concurrent administration of AJ(Atopy cream, Jawoongo)+SPDJTK on atopic dermatitis-like skin lesions by using in NC/Nga atopic dermatitis mouse induced by BMAC-induced mice. Methods Clinical skin score, hematology and Serum total IgE and IgG1 of NC/Nga atopic dermatitis mice were evaluated. Moreover, the cytokine level, total cell number, Immunohistochemical staining and Histological features of axillary lymph node(ALN), draining lymph node(DLN), peripheral blood mononuclear cells(PBMCs) and dorsal skin tissue were used in NC/Nga mice. Results Orally administrated SPDJTK with concurrent administration of SPDJTK and AJ decreased the clinical skin score, total cell number of WBC, eosinophils in blood, serum total IgE & IgG1, IL-5, IL-13, IFN-$\gamma$. Also, total cell number of ALN and dorsal skin tissue, absolute cell number of CD4+, CD8+, CD3+CD69+, CD3+CCR3+, CCR3+, CD4+CXCR5+ in ALN, absolute cell number of CD3+CCR3+, CCR3+ in DLN, granulocytes in PBMCs, activation cell number of CD3+CD69+, CCR3+, total cell number of CD3+ T cell in dorsal skin tissue were significantly decreased. Furthermore, thickness of epidermis, infiltrated inflammatory immune cell and mast cell in dermis, amount of Eotaxin2 mRNA, CCR3 mRNA in dorsal skin tissue, gene expression of IL-5, IL-13 mRNA in ALN, CD4+ Th cell in dorsal skin tissue and CCR3+ eosinophils in ALN were all significantly decreased. However, total number of DLN, absolute number of CD3e+ T cell and CD19+ B cell, absolute number of CD4+, number of Th cell in DLN and gene expression of foxp3 mRNA were significantly increased significantly. Conclusions Concurrent administration of SPDJTK and AJ on atopic dermatitis in NC/Nga atopic dermatitis mouse was very effective treatment for atopic dermatitis.

• 생명과학회지
• /
• 제26권9호
• /
• pp.1027-1032
• /
• 2016

항원은 병원체로부터 유래한 질병인자다. 생명체는 항원에 대항하는 방어계인 면역계를 가지고 있다. 항원은 식세포작용, 항체, 보체 활성화, NK세포 혹은 MHC 분자를 통한 세포독성 T세포와 같은 방법을 통해서 처리된다. 림프절은 스트로마세포와 3차원 네트워크를 통해서 구성되어 있다. Fibroblastic reticular cells (FRC)는 림프절 T zone에서 T세포와 상호작용한다. FRC는 세포외 기질 생산과 homing 케모카인을 생산하여 감염에 대비한다. 하지만, FRC가 항원처리과정에 관련되어있다는 보고는 없다. 본 연구는 FRC의 항원처리 관련성에 대한 연구이다. 이를 위해 FRC는 대식세포, T세포, LPS, 그리고 TNFα와 같은 다양한 감염상황에 노출시켜 연구를 진행하였다. FRC가 대식세포 및 T세포와 공배양 했을 때 FRC가 형태적 변화와 FRC간 빈 공간 형성이 관찰 되었다. MMP 활성은 Y27632와 T세포에 의해 조절 되었다. 더욱이, 염증물질인 TNFα를 FRC에 처리 후 마이크로어레이를 통한 결과에서 부착분자와 MHC I antigen transporter의 발현을 조절하는 것으로 나타났다. FRC 단일층에 LPS와 대식 세포를 공배양 했을 때 NO 생성력이 크게 향상되었다. GFP antigen을 FRC와 대식세포 공배양군에 처리 했을 때 항원 흡수율이 증가되었다. 이러 결과는 FRC가 항원처리에 관여하고 있다는 것을 의미하며 이는 림프절이 항원처리과정에 연관되어 있다는 것을 제시한다.

• 한국융합학회논문지
• /
• 제9권5호
• /
• pp.33-41
• /
• 2018

본 연구는 심박변이도(HRV)와 인공신경망을 이용하여 강건하고 정확한 융복합 감정예측 모형인 EPNN (Emotion Prediction Neural Network)을 개발하는 것을 주요 연구목적으로 한다. 본 연구에서 제안하는 EPNN은 기존 유사연구와는 달리 은닉노드의 활성함수로서 하이퍼볼릭 탄젠트, 선형, 가우시안 함수를 융복합적으로 이용하여 모형의 정확도를 향상시킨다. 본 연구에서는 EPNN의 타당성을 검증하기 위하여 20명의 실험자를 대상으로 머니게임으로 감정을 유도한 후에 해당 실험자의 심박변이도 측정값을 입력자료로 사용하였다. 아울러 그들의 Valence와 Arousal을 EPNN의 출력값으로 사용하였다. 실험결과 Valence에 대한 F-Measure는 80%이고, Arousal의 경우 95%로 나타났다. 한편 EPNN의 타당성을 측정하기 위하여 기존 감정예측 연구에 사용된 경쟁모형인 인공신경망, 로지스틱 회귀분석, 서포트 벡터 머신, 랜덤 포레스트 모형과 성과를 비교하였다. 그 결과 본 연구에서 제안하는 EPNN이 더 우수한 감정예측 결과를 보였다. 본 연구의 결과는 향후 유비쿼터스 디지털 헬스 환경에서 사용되는 다양한 웨어러블 기기에 적용되어 사용자들의 일상생활 속에서 시시각각 변하는 감정을 정확히 예측하고 적절하게 관리하는데 적용될 수 있을 것이다.

• Animal cells and systems
• /
• 제1권1호
• /
• pp.157-164
• /
• 1997

Ly-6E.1 antigen was proposed as a regulatory molecule of T lymphocyte activation, a hematopoietic stem cell marker, a memory cell marker, and an adhesion molecule. Though there were several reports suggesting the presence of Ly-6 ligand, the characterization of the ligand was not yet performed, As an attempt to screen the expression of Ly-6E.1 ligand, we prepared a probe for detecting Ly-6E.1 ligand by producing a fusion protein between Ly-6E.1 and $hlgC_{r1}$, A mammalian cell expression vector with Ly-6E.$1/hlgC_{r1}$ chimeric cDNA was transfected in SP2/0-Ag14 myeloma cells, and stable transfectants were selected. The fusion protein was produced as a dimer and maintained the epitopes for monoclonal antibodies specific for Ly-6E.1 and for anti-human lgG antibody. The purified fusion protein through Gammabind G column was used for FACS analyses for the expression of Ly-6E.1 ligand. The fusion protein interacted with several cell lines originating from B cells, T cells, or monocytes. The fusion Protein also strongly stained bone marrow, lymph node, and spleen cells, but thymic cells weakly, if any. The staining was more obvious in C57BL/6 $(Ly-6^b)$ than Balb/c $(Ly-6^a)$ mice. These results suggest that the interaction of Ly-6E.1 with Ly-6E.1 ligand may function both in the stem cell environment and in the activation of mature lymphocytes. The fusion protein may be a valuable tool in characterization of biochemical properties of the Ly-6E.1 ligand and, further, in isolating its cDNA.

• 전기학회논문지
• /
• 제60권3호
• /
• pp.639-647
• /
• 2011

In this paper, we introduce a design methodology of data-centroid Radial Basis Function neural networks with extended polynomial function. The two underlying design mechanisms of such networks involve K-means clustering method and Particle Swarm Optimization(PSO). The proposed algorithm is based on K-means clustering method for efficient processing of data and the optimization of model was carried out using PSO. In this paper, as the connection weight of RBF neural networks, we are able to use four types of polynomials such as simplified, linear, quadratic, and modified quadratic. Using K-means clustering, the center values of Gaussian function as activation function are selected. And the PSO-based RBF neural networks results in a structurally optimized structure and comes with a higher level of flexibility than the one encountered in the conventional RBF neural networks. The PSO-based design procedure being applied at each node of RBF neural networks leads to the selection of preferred parameters with specific local characteristics (such as the number of input variables, a specific set of input variables, and the distribution constant value in activation function) available within the RBF neural networks. To evaluate the performance of the proposed data-centroid RBF neural network with extended polynomial function, the model is experimented with using the nonlinear process data(2-Dimensional synthetic data and Mackey-Glass time series process data) and the Machine Learning dataset(NOx emission process data in gas turbine plant, Automobile Miles per Gallon(MPG) data, and Boston housing data). For the characteristic analysis of the given entire dataset with non-linearity as well as the efficient construction and evaluation of the dynamic network model, the partition of the given entire dataset distinguishes between two cases of Division I(training dataset and testing dataset) and Division II(training dataset, validation dataset, and testing dataset). A comparative analysis shows that the proposed RBF neural networks produces model with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.