http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Does the traditional snakebite severity score correctly classify envenomated patients?
강승호,문정미,전병조 대한응급의학회 2016 Clinical and Experimental Emergency Medicine Vol.3 No.1
Objective This study aims to help set domestic guidelines for administration of antivenom to envenomated patients after snakebites. Methods This retrospective observational case series comprised 128 patients with snake envenomation. The patients were divided into two groups according to the need for additional antivenom after the initial treatment based on the traditional snakebite severity grading scale. One group successfully recovered after the initial treatment and did not need any additional antivenom (n=85) and the other needed an additional administration of antivenom (n=43). Results The group requiring additional administration of antivenom showed a higher local effect score and a traditional snakebite severity grade at presentation, a shorter prothrombin and activated partial prothrombin time, a higher frequency of rhabdomyolysis and disseminated intravascular coagulopathy, and longer hospitalization than the group that did not need additional antivenom. The most common cause for additional administration was the progression of local symptoms. The independent factor that was associated with the need for additional antivenom was the local effect pain score (odds ratio, 2.477; 95% confidence interval, 1.309 to 4.689). The optimal cut-off value of the local effect pain score was 1.5 with 62.8% sensitivity and 71.8% specificity. Conclusion When treating patients who are envenomated by a snake, and when using the traditional snakebite severity scale, the local effect pain score should be taken into account. If the score is more than 2, additional antivenom should be considered and the patient should be frequently assessed.
강승호,남주선,Kang, Seung-Ho,Nam, Ju-Sun 한국통계학회 2012 응용통계연구 Vol.25 No.1
Recent assessments of the biosimilarity of biologic products have received considerable global attention. A clinical trial should be conducted to assess the biosimilarity of a biosimilar product and a innovator biological product. In this paper we will describe several methods for the implementation of clinical trials and statistical analysis, a real international case and related international guidelines. 최근 들어 바이오시밀러에 대한 국내외 관심이 매우 증가하고 있다. 바이오시밀러가 오리지널 생물의약품과 효능과 안전성이 유사함을 보이기 위해서는 최종적으로 임상시험을 수행하여야 한다. 본 논문에서는 이러한 임상시험의 수행과 통계적 분석에 필요한 여러 방법들과 외국의 사례 그리고 관련된 가이드라인들을 살펴볼 것이다.
강승호 한국사회과학연구회 2004 동향과 전망 Vol.- No.60
A comparison of the hollowing out of industries in Korea, Japan and Taiwan 한국ㆍ일본ㆍ대만의 산업공동화 비교
실시간 공격 탐지를 위한 Pearson 상관계수 기반 특징 집합 선택 방법
강승호,정인선,임형석 한국융합보안학회 2018 융합보안 논문지 Vol.18 No.5
The performance of a network intrusion detection system using the machine learning method depends heavily on the composition and the size of the feature set. The detection accuracy, such as the detection rate or the false positi ve rate, of the system relies on the feature composition. And the time it takes to train and detect depends on the siz e of the feature set. Therefore, in order to enable the system to detect intrusions in real-time, the feature set to be used should have a small size as well as an appropriate composition. In this paper, we show that the size of the feat ure set can be further reduced without decreasing the detection rate through using Pearson correlation coefficient bet ween features along with the multi-objective genetic algorithm which was used to shorten the size of the feature set in previous work. For the evaluation of the proposed method, the experiments to classify 10 kinds of attacks and be nign traffic are performed against NSL_KDD data set. 기계학습을 이용하는 침입 탐지 시스템의 성능은 특징 집합의 구성과 크기에 크게 좌우된다. 탐지율과 같은 시스템의 탐지 정확도는 특징 집합의 구성에, 학습 및 탐지 시간은 특징 집합의 크기에 의존한다. 따라서 즉각적인 대응이 필수인 침입 탐지 시스템의 실시간 탐지가 가능하도록 하려면, 특징 집합은 크기가 작으면서도 적절한 특징들로 구성하여야 한 다. 본 논문은 실시간 탐지를 위한 특징 집합 선택 문제를 해결하기 위해 사용했던 기존의 다목적 유전자 알고리즘에 특징 간의 Pearson 상관계수를 함께 사용하면 탐지율을 거의 낮추지 않으면서도 특징 집합의 크기를 줄일 수 있음을 보 인다. 제안한 방법의 성능평가를 위해 NSL_KDD 데이터를 사용하여 10가지 공격 유형과 정상적인 트래픽을 구별하도 록 인공신경망을 설계, 구현하여 실험한다.