• The Journal of Pediatrics of Korean Medicine
• /
• v.30 no.2
• /
• pp.82-95
• /
• 2016

Objectives The purpose of this study is to investigate recent clinical studies on effect of herbal medicine treatment for recurrent respiratory tract infection in children. Methods We searched some clinical studies about recurrent respiratory tract infection in children from the China Academic Journal (CAJ) of China National Knowledge Infrastructure (CNKI) (January 2011 to December 2015). Results 50 papers were selected from 168 studies. The herbal decoction was main herbal medicine treatment for recurrent respiratory tract infection in children. Commonly used herbs were Atractylodis Rhizoma Alba, Glycyrrhizae Radix, Astragali Radix, Saposhnikovia Radix and Pseudostellariae Radix. Yupingfeng keli was the most frequently used herbal compound. Total effective rate was 66.4-100%, experimental group was significantly higher than control group in 45 papers (P<0.05). Immune index (in 22 papers) and curative effect of TCM syndrome (in 7 papers) were significantly higher than those of control group after treatment (p<0.05). Conclusions Herbal medicine has been shown as an effective treatment on recurrent respiratory tract infection in children. These research results can be utilized in other clinical studies and in treatment of recurrent respiratory tract infection for children.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.7
• /
• pp.259-266
• /
• 2018

In this paper, we propose a method to classify a document using a Recurrent Neural Network by extracting features considering word sense and contexts. Word2vec method is adopted to include the order and meaning of the words expressing the word in the document as a vector. Doc2vec is applied for considering the context to extract the feature of the document. RNN classifier, which includes the output of the previous node as the input of the next node, is used as the document classification method. RNN classifier presents good performance for document classification because it is suitable for sequence data among neural network classifiers. We applied GRU (Gated Recurrent Unit) model which solves the vanishing gradient problem of RNN. It also reduces computation speed. We used one Hangul document set and two English document sets for the experiments and GRU based document classifier improves performance by about 3.5% compared to CNN based document classifier.

• Journal of Ocean Engineering and Technology
• /
• v.35 no.4
• /
• pp.287-295
• /
• 2021

This paper describes a recurrent neural network (RNN) for the fault classification of a blade pitch system of a spar-type floating wind turbine. An artificial neural network (ANN) can effectively recognize multiple faults of a system and build a training model with training data for decision-making. The ANN comprises an encoder and a decoder. The encoder uses a gated recurrent unit, which is a recurrent neural network, for dimensionality reduction of the input data. The decoder uses a multilayer perceptron (MLP) for diagnosis decision-making. To create data, we use a wind turbine simulator that enables fully coupled nonlinear time-domain numerical simulations of offshore wind turbines considering six fault types including biases and fixed outputs in pitch sensors and excessive friction, slit lock, incorrect voltage, and short circuits in actuators. The input data are time-series data collected by two sensors and two control inputs under the condition that of one fault of the six types occurs. A gated recurrent unit (GRU) that is one of the RNNs classifies the suggested faults of the blade pitch system. The performance of fault classification based on the gate recurrent unit is evaluated by a test procedure, and the results indicate that the proposed scheme works effectively. The proposed ANN shows a 1.4% improvement in its performance compared to an MLP-based approach.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.357-363
• /
• 2023

In underwater signal processing, separating individual signals from mixed signals has long been a challenge due to low signal quality. The common method using Short-time Fourier transform for spectrogram analysis has faced criticism for its complex parameter optimization and loss of phase data. We propose a Triple-path Recurrent Neural Network, based on the Dual-path Recurrent Neural Network's success in long time series signal processing, to handle three-dimensional tensors from multi-channel sensor input signals. By dividing input signals into short chunks and creating a 3D tensor, the method accounts for relationships within and between chunks and channels, enabling local and global feature learning. The proposed technique demonstrates improved Root Mean Square Error and Scale Invariant Signal to Noise Ratio compared to the existing method.

• Investigative Magnetic Resonance Imaging
• /
• v.8 no.1
• /
• pp.24-31
• /
• 2004

Purpose : To know the differences of proton MR spectroscopic features between recurrent rectal cancer and fibrosis in post-operative period, and to evaluate the possibility to discriminate recurrent rectal cancer from post-operative fibrosis by analysis of proton MR spectra. Materials and Methods : We evaluated the proton MR spectra from 25 soft tissue masses in perirectal area that developed in post-operative period after operation for the resection of rectal cancer. Our series included 11 cases of recurrent rectal cancer and 14 of fibrotic mass. All cases of recurrent rectal cancer and post-operative fibrosis were confirmed by biopsy. We evaluated the spectra with an attention to the differences of pattern of the curves between recurrent rectal cancer and post-operative fibrosis. The ratio of peak area of all peaks at 1.6-4.1ppm to lipid (0.9-1.6ppm) [P (1.6-4.1ppm/P (0.9-1.6ppm)] was calculated in recurrent rectal cancer and post-operative fibrosis groups, and compared the results between these groups. We also evaluated the sensitivity and specificity for discriminating recurrent rectal cancer from post-operative fibrosis by analysis of $^1H-MRS$. Results : Proton MR spectra of post-operative fibrosis showed significantly diminished amount of lipids compared with that of recurrent rectal cancer. The ratio of P (1.6-4.1ppm)/P (0.9-1.6ppm) in post-operative fibrosis was much higher than that of recurrent rectal cancer with statistical significance (p < .05) due to decreased peak area of lipids. Mean (standard deviations of P (1.6-4.1ppm)/P (0.9-1.6ppm) in post-operative fibrosis and recurrent rectal cancer group were $2.71{\pm}1.48\;and\;0.29{\pm}0.11$, respectively. With a cut-off value of 0.6 for discriminating recurrent rectal cancer from post-operative fibrosis, both the sensitivity and specificity were $100\%$ (11/11, and 14/14). Conclusion : Recurrent rectal cancer and post-operative fibrosis can be distinguished from each other by analysis of proton MR spectroscopic features, and $^1H-MRS$ can be a new method for differential diagnosis between recurrent rectal cancer and post-operative fibrosis.

• Kyungpook Mathematical Journal
• /
• v.6 no.2
• /
• pp.69-76
• /
• 1965

• 대한생식의학회:학술대회논문집
• /
• 2004.06a
• /
• pp.39-42
• /
• 2004

• 대한생식의학회:학술대회논문집
• /
• 2006.06a
• /
• pp.114-116
• /
• 2006

• Pediatric Infection and Vaccine
• /
• v.15 no.2
• /
• pp.188-194
• /
• 2008

Purpose : The purpose of this study is to investigate the clinical characteristics of recurrent Kawasaki disease (KD). Methods : From January 2004 to December 2007, the medical records of 20 children with recurrent KD in Severance Children's Hospital were retrospectively reviewed. The clinical characteristics, laboratory findings, treatment and complications of these patients were compared between the initial episode and the second episode. Results : At the initial episode of the recurrent KD group, the gender ratio was 1.2:1 (male:female) and the mean age was $37.2{\pm}19.9$ months. The interval between the two episodes in the recurrent KD group was 3.3 months. The febrile period before admission was shorter for the second episode (P=0.034). The skin rash was less developed in the second episode. But there were no differences in the laboratory results and complications between the initial episode and the second episode. Three patients (15%) among those with a second episode failed to respond to the initial intravenous immunoglubulin treatment. On comparison between the initial episodes of the recurrent group and the nonrecurrent group, the erythrocyte sedimentation rate was higher in the first episode of the recurrent KD group. Conclusions : For recurrent KD, it tends to present more atypical features than the KD that occurs for the first time. Physicians should consider these characteristics when making the diagnosis and treating recurrent KD.

• Tuberculosis and Respiratory Diseases
• /
• v.64 no.5
• /
• pp.341-346
• /
• 2008

Background: Recurrent pulmonary tuberculosis (TB) can be due to relapse of the original infecting strain or due to reinfection with a new strain of Mycobacterium tuberculosis. We investigated the clinical characteristics and efficacy of short-term treatment (6 months) in patients with recurrent pulmonary TB. Methods: Twenty-nine patients with recurrent pulmonary TB were compared with control patients who received primary treatment for pulmonary TB with respect to drug sensitivity and outcomes of treatment. Results: Most patients with recurrent pulmonary TB (25 cases, 86.2%) recurred more than 2 years after the completion of previous treatment. Twenty-three patients (82.1%) with recurrent pulmonary TB were sensitive to all anti-tuberculous drugs and a ratio was similar to the drug sensitivities observed in control patients. The outcomes of short-term treatment in patients with drug-sensitive TB were not significantly different between the two groups. Conclusion: Recurrent pulmonary TB in the study area was likely due to reinfection with new strains. Thus the short-term treatment of patients with drug-sensitive recurrent pulmonary TB may be successful.