http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
지표면상을 전파하는 소음의 초과감쇠 산정방법에 관한 연구
오재응,김동규,임동규,Oh, J.E.,Kim, D.G.,Yim, T.K. 한국음향학회 1988 韓國音響學會誌 Vol.7 No.2
본 연구는 소음전파에 대한 옥외실험과 축적 음향모형실험에 의해서 지표면에 의한 초과감쇠 특징을 밝힌 것으로써, 옥외실험에 의한 소음전파감쇠는 음향출력이 큰 소형엔진을 사용하여, 거리감쇠로부터 산출한 실측의 초과감쇠와 Log(D/(Hs+Hr))의 관계를 확인했다. 그 결과 소음전파감쇠는 풍향, 주파수에 따라 다르며 직선회귀 된다는 것을 알 수 있었다. 그리고, 지표면상의 초과감쇠치는 통기저항을 이용해서 Log(D/Hs+Hr))을 파라미터로써 구할 수 있었고, 가장 적당한 통기저항$\sigma$는 실측의 초과감쇠치와 임의의 $\sigma$에 대한 $L=-20Log\mid1+(r_1/r_2)Qexp(ik, \bigtriangleup r)\mid$ 식의 평균자승 오차가 가장 적어질 때 결정된다. 모형의 지표로써 축척 1/40의 모형실험으로, 큰 무향실내에서 거리감쇠의 측정을 한 결과, 실측치와의 대응이 충분하다는 것을 확인했다. This study is to explain the characteristic of excess attenuation on the ground through the outdoors experiment about noise propagation and the reduced model experiment of acoustic. The outdoors experiment on the attenuation of noise propagation was tried with the small engine that had large acoustic output, and then it was conformed that there was relationship between the excess attenuation calculated by measurement from distance attenuation and Log(D/(Hs+Hr)). As a result, it was found that the attenuation of noise propogation depended upon the direction of the wind and frequency and was regressed in a straight line. And the numerical values of excess attenuation on the ground could be calculated by regarding Log(D/(Hs+Hr)) as a parameter with an airing resistance $\sigma$. It was found that when the mean square error between the excess attenuation calculated by measurement and the value calculated by a fomula $L=-20Log\mid1+(r_1/r_2)Qexp(ik, \bigtriangleup r)\mid$ about optional $\sigma$ was least, the optimal decision of u was made. As the characteristic of model is the model experiment on a reduced scale of 1 to 40, It was conformed that it corresponds enough with the measurement value with measuring the distance attenuation in the large anecoic chamber.
박유환(Y .H .Park),정춘혜(C .H .Chung),조영신(Y .S .Cho),김동규(D .G .Kim),손현화(H .H .Son),서영선(Y .S .Seo),이미자(M .J .Lee) 대한내과학회 2000 대한내과학회지 Vol.58 No.5
N/A Background : The role of oncogenes and tumor suppressor genes in the pathogenesis of gastric carcinoma has recently received considerable attention. The tumor suppressor gene, p53 locus on chromosome 17p, ceases a mitosis in G1/S check point after DNA demage. Abnormality of the p53 tumor suppressor gene plays an important role in alteration of cells and possibly leads to cancer development. The authors investigated the correlation between expression of p53 and prognosis in advanced gastric carcinoma. Methods : Expression of tumor suppresor gene p53 was investigated immunohistochemically in the primary lesion of 34 patients with advanced gastric carcinoma using paraffin embedded surgical specimens and the relationship of p53 immunopositivity with the clinicopathologic variables(Age, Sex, TNM staging, Lauren classification, Borrmann classification), 5-year survivals and life curves were analyzed. Kaplan and Meier's method is used as a life curve and log crank method is used for the analysis of prognostic factor. Results : Total p53 positive rate was 53% (18 of 34) of all cases. p53 immunopositivity was not associated with other clinicopathologic variables. The 5-year survivals were 44% and 11% for patients with p53 negative and positive gastric carcinomas, respectively. Patents with the expression of p53 has predominantly poor results in comparison of life curve(p<0.05) Conclusion : These findings suggest that p53 gene alteration be associated with poor prognosis of advanced gastric adenocarcinoma(Korean J Med 58:568-574, 2000)
동ㆍ서양인 얼굴 데이터를 활용한 딥러닝 기반 안면마비 인식 정확도 평가
김동규(D. G. Kim),서상규(S. K. Seo),이현주(H. J. Lee),태기식(K. S. Tae) 한국재활복지공학회 2023 재활복지공학회논문지 Vol.17 No.1
안면마비의 연간 발병률은 100,000명 당 20명으로 통계적으로 흔한 질병이다. 그러나 많은 사람들이 적절한 치료를 받지 못해 다양한 후유증을 일으키고 있다. 안면마비에 대한 정확하고 빠른 진단은 중재 및 긍정적인 예후를 위해 반드시 필요하다. 본 연구에서는 동양인과 서양인을 구별하여 안면마비를 정확히 예측할 수 있는 자가진단 딥러닝 모델을 개발하고 검증하였다. 이를 위해 얼굴 탐지(face detecting)와 얼굴 랜드마크(face landmark)를 통해 데이터 전처리와 모델 훈련을 수행하였다. 동양인으로만 구성된 데이터를 사용하여 모델을 학습시킨 결과, 동양인의 안면마비 예측 정확도는 96.2%였지만 서양인의 예측 정확도는 68.7%로 나타났다. 서양인으로만 구성된 데이터를 사용하여 모델을 학습시킨 경우에는, 서양인의 안면 마비 예측 정확도 95.5%, 동양인의 예측 정확도는 80.5%였다. 이러한 결과를 기반으로, 동서양 혼합 모델을 사용한 딥러닝 학습을 하여 동서양 얼굴 인식 모두에게서 91.2% 이상의 높은 예측 정확도를 도출하였다. 추후 충분한 데이터 확보와 함께 딥러닝 모델을 이용하면 다양한 인종에서 안면 마비를 정확히 예측할 수 있는 시스템이 마련될 수 있는 것으로 사료된다. The annual incidence rate of facial paralysis is 20 per 100,000 people, making it a statistically common disease. However, many people suffer from various sequelae due to lack of appropriate treatment. Accurate and prompt diagnosis of facial paralysis is essential for intervention and positive prognosis. In this study, a self-diagnostic deep learning model was developed and validated to accurately predict facial paralysis, distinguishing between East Asian and Western populations. Data preprocessing and model training were performed using face detection and face landmarking. When the model was trained using data consisting of only East Asian individuals, the accuracy of facial paralysis prediction for East Asians was 96.2%, but it was 68.7% for Westerners. Conversely, when the model was trained using data consisting of only Westerners, the accuracy of facial paralysis prediction for Westerners was 95.5%, while it was 80.5% for East Asians. Based on these results, deep learning training using a mixed model of both East Asian and Western populations yielded high prediction accuracy of over 91.2% for both populations. With sufficient data, it is expected that a system that accurately predicts facial paralysis in various races can be established using deep learning models in the future.