• Journal of the Korean Society of Physical Medicine
• /
• v.12 no.3
• /
• pp.33-42
• /
• 2017

PURPOSE: To compare postural vertical training with and without visual feedback for improving functional recovery in post-stroke hemiparesis patients with pusher syndrome. METHODS: This study used a single-subject research with alternating design with multiple baselines. Three patients with hemiparetic post-stroke diagnosed with pusher syndrome were selected from the inpatients at the department of physical therapy of a local rehabilitation hospital. For subjective postural vertical (SPV) training with and without visual feedback, an alternating treatment was used. The subjects were randomly selected using the sequence of the two training methods upon starting the intervention, and then the training was alternated. SPV training was performed twice a day, once in the morning and again in the afternoon. Scale for contraversive pushing (SCP), postural assessment scale for stroke, and Barthel index score were used to determine the intervention-related changes. RESULTS: Compared to the average score at baseline, the average SCP score for the SPV training without visual feedback decreased from 5.3 to 2.8, from 4.6 to 3, and from 3.5 to 2.7 for subjects 1, 2, and 3, respectively. However, the average score for the SPV training with visual feedback decreased from 5.3 to 3.1, from 4.6 to 3.5, and from 3.5 to 3.3 for subjects 1, 2, and 3, respectively. CONCLUSION: Postural vertical training without visual feedback may be more beneficial than postural vertical training with visual feedback for improving pushing behavior and functional activity in stroke patients with pusher syndrome.

• Journal of Animal Science and Technology
• /
• v.52 no.6
• /
• pp.475-480
• /
• 2010

The bovine fatty acid binding protein 4 (FABP4) plays an important role to uptake intracellular fatty acid. It has been previously reported as a positional candidate gene for marbling score in livestock. The re-sequencing of FABP4 gene detected a polymorphic AT repeated sequence in intron II of FABP4 gene. Allelic distribution for this microsatellite marker was examined in other cattle breeds. A total of 8 alleles were detected with diverse repeat units (14 to 21 AT repeat) in Hanwoo and 7 breeds. Of the 8 alleles, the predominant alleles were $[AT]_{16}$, $[AT]_{18}$ and $[AT]_{19}$ in the Hanwoo and 7 cattle breeds. The linear mixed model for genotypic effect (3237AT) on carcass traits showed a significant effect on marbling score (MAR P=0.025) and live weight (LWT; P=0.04) in the 583 Hanwoo cattle population. Live weight (LW) was highest in the homozygous $(AT)_{17}$ genotype ($557.5{\pm}6.94$) and lowest in the heterozygous $(AT)_{16/17}$ genotype ($521.7{\pm}7.70$). On the other hand, the homozygous $(AT)_{17}$ genotype ($3.0{\pm}0.15$) has the highest effect on marbling score and the lowest effect was in homozygous (AT)$_{18}$ genotype ($2.2{\pm}0.15$). The marbling score difference between both groups was 0.8 which is around two times higher than SNP genotype effect on marbling score in Limousin $\times$ Wagyu crosses.

• Microbiology and Biotechnology Letters
• /
• v.25 no.6
• /
• pp.560-565
• /
• 1997

DNA fragment assembly is a major concem in shot-gun DNA sequencing project. It is to reconstruct a consensus DNA sequence from a collection of random oritented fragments. We developed a computer program that is useful for DNA fragment assembly. Inputs to the program are DNA fragment sequences including IUB-IUPAC bases. The program produces the most probable reconstruction ot the original DNA sequence as a text format or a PostScript format. The program consists of four phases: the first phase quickly eliminates fragment pairs that can not possibly overlap. In the second phase, the quality of overlap between each pair is calculated to a score. In the third phase, overlap pairs are sorted by their scores and consistency of the overlaps is checked. The last phase determines consensus sequences and displays them. The performance of fragment assembly program was tested on a set of DNA fragment sequences which were generated from long DNA sequences of GenBank by a fragmentation program.

• Journal of Korean Academy of Nursing
• /
• v.37 no.3
• /
• pp.343-352
• /
• 2007

Purpose: The purpose of this study was to examine the effects of an inpatient pulmonary rehabilitation program on dyspnea, exercise capacity, and health related quality of life in inpatients with chronic lung disease. Method: This quasi experimental study was designed with a nonequivalent control group pre-post test time series. Twenty three patients were assigned to the experimental group and nineteen to the control group. The inpatient pulmonary rehabilitation program was composed of upper and lower extremity exercise, breathing retraining, inspiratory muscle training, education, relaxation and telephone contacts. This program consisted of 4 sessions with inpatients and 4 weeks at home after discharge. The control group was given a home based pulmonary rehabilitation program at the time of discharge. The outcomes were measured by the Borg score, 6MWD and the Chronic Respiratory Disease Questionnaire(CRQ). Results: There was a statistically significant difference in dyspnea between the experimental group and control group, but not among time sequence, or interaction between groups and time sequence. Also significant improvements in exercise capacity and health related quality of life were found only in the experimental group. Conclusions: An Inpatient pulmonary rehabilitation program may be a useful intervention to reduce dyspnea, and increase exercise capacity and health related quality of life for chronic lung disease patients.

• Journal of information and communication convergence engineering
• /
• v.20 no.4
• /
• pp.309-316
• /
• 2022

This investigation is aimed at automatic text summarization on large-scale Vietnamese datasets. Vietnamese articles were collected from newspaper websites and plain text was extracted to build the dataset, that included 1,101,101 documents. Next, a new single-document extractive text summarization model was proposed to evaluate this dataset. In this summary model, the k-means algorithm is used to cluster the sentences of the input document using different text representations, such as BoW (bag-of-words), TF-IDF (term frequency - inverse document frequency), Word2Vec (Word-to-vector), Glove, and FastText. The summary algorithm then uses the trained k-means model to rank the candidate sentences and create a summary with the highest-ranked sentences. The empirical results of the F1-score achieved 51.91% ROUGE-1, 18.77% ROUGE-2 and 29.72% ROUGE-L, compared to 52.33% ROUGE-1, 16.17% ROUGE-2, and 33.09% ROUGE-L performed using a competitive abstractive model. The advantage of the proposed model is that it can perform well with O(n,k,p) = O(n^(k+2/p)) + O(nlog₂n) + O(np) + O(nk²) + O(k) time complexity.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.967-970
• /
• 2020

In a previous work we have done, we presented a monitoring system to automatically detect some dogs' behaviors from videos. However, the input video data used by that system was pre-trimmed to ensure it contained a dog only. In a real-life situation, the monitoring system would continuously receive video data, including frames that are empty and ones that contain people. In this paper, we propose a YOLOv4-based system for automatic object detection and trimming of dog videos. Sequences of frames trimmed from the video data received from the camera are analyzed to detect dogs and people frame by frame using a YOLOv4 model, and then records of the occurrences of dogs and people are generated. The records of each sequence are then analyzed through a rule-based decision tree to classify the sequence, forward it if it contains a dog only or ignore it otherwise. The results of the experiments on long untrimmed videos show that our proposed method manages an excellent detection performance reaching 0.97 in average of precision, recall and f-1 score at a detection rate of approximately 30 fps, guaranteeing with that real-time processing.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.167-171
• /
• 2020

문서 요약은 길이가 긴 원본 문서에서 의미를 유지한 채 짧은 문서나 문장을 얻어내는 작업을 의미한다. 딥러닝을 이용한 자연어처리 기술들이 연구됨에 따라 end-to-end 방식의 자연어 생성 모델인 sequence-to-sequence 모델을 문서 요약 생성에 적용하는 방법들이 연구되었다. 본 논문에서는 여러 자연어처리 분야에서 높은 성능을 보이고 있는 BERT 모델을 이용한 자연어 생성 모델에 복사 메커니즘과 강화 학습을 추가한 문서 요약 모델을 제안한다. 복사 메커니즘은 입력 문장의 단어들을 출력 문장에 복사하는 기술로 학습데이터에서 학습되기 힘든 고유 명사 등의 단어들에 대한 성능을 높이는 방법이다. 강화 학습은 정답 단어의 확률을 높이기 위해 학습하는 지도 학습 방법과는 달리 연속적인 단어 생성으로 얻어진 전체 문장의 보상 점수를 높이는 방향으로 학습하여 생성되는 단어 자체보다는 최종 생성된 문장이 더 중요한 자연어 생성 문제에 효과적일 수 있다. 실험결과 기존의 BERT 생성 모델 보다 복사 메커니즘과 강화 학습을 적용한 모델의 Rouge score가 더 높음을 확인 하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.615-617
• /
• 2023

In recent years, the growing popularity of dogs due to the benefits they bring their owners has contributed to the increase of the number of dogs raised. For owners, it is their responsibility to ensure their dogs' health and safety. However, it is challenging for them to continuously monitor their dogs' activities, which are important to understand and guarantee their wellbeing. In this work, we introduce a camera-based monitoring system to help owners automatically monitor their dogs' activities. The system receives sequences of RGB images and uses YOLOv7 to detect the dog presence, and then applies post-processing to perform dog-centered image cropping on each input sequence. The optical flow is extracted from each sequence, and both sequences of RGB and flow are input to a two-stream EfficientNet to extract their respective features. Finally, the features are concatenated, and a bi-directional LSTM is utilized to retrieve temporal features and recognize the activity. The experiments prove that our system achieves a good performance with the F-1 score exceeding 0.90 for all activities and reaching 0.963 on average.

• Asian-Australasian Journal of Animal Sciences
• /
• v.19 no.11
• /
• pp.1529-1535
• /
• 2006

Leptin has been investigated as a candidate gene for fat characteristics in beef cattle. Previously, we have reported 57 sequence variants discovered in Korean cattle (Bos Taurus coreanae). In this study, we examined the association between polymorphisms of leptin and carcass traits (cold carcass weight (CWT) and marbling score (Marb)) in Korean cattle. Among 57 polymorphisms, 11 common polymorphic sites were genotyped in our beef cattle (n = 437). Statistical analysis revealed that one single nucleotide polymorphism in coding exon (c.+411T>C (A137A)) showed a significant association with the yield trait, CWT. The C-bearing genotypes (CC or CT) of c.+411T>C (A137A) showed the higher CWT (p = 0.006). c.+150C>G (S50S) also showed a significant association with the quality trait, Marb (p = 0.01). Our findings suggest that polymorphisms in leptin might be one of the important genetic factors that influence carcass yield and quality in beef cattle, especially in CWT and Marb.

• Journal of Korean Public Health Nursing
• /
• v.31 no.2
• /
• pp.220-233
• /
• 2017

Purpose: This study compared the body composition and health conservation of low and high physical activity groups in professional retired elderly. Methods: The participants were 132 professional retired elders using a university of education for the elderly and senior welfare centers in D and K city. Data were collected through interviews from November to December in 2015. The body composition was measured using an Inbody 230 and the health conservation was measured using Sung's (2005) Health Conservation Scale. The low and high physical activity groups were divided by CHAMPS (The Community Healthy Activities Model Program for Seniors Questionnaire). The collected data were analyzed using a $x^2$ test, independentt-test, andbinary logistic regression using the SPSS/WIN 20.0 program. Results: The body composition (body mass index, body fat percentage, and fitness score) and health conservation score were higher in the high physical activity group than in the low physical activity group. The result shows that the body weight (OR=10.62) is the highest risk factor influencing the physical activity in professional career-retired elders. Conclusion: These results suggest that to induce physical activity level among professional retired elders, nursing interventions are needed to increase the health conservation and body composition, particularly in controlling the body weight.