수정DRAGON점수를 이용한 기계적동맥내혈전제거술의예후 예측

김근현,노명호,김용범 대한신경과학회 2015 대한신경과학회지 Vol.33 No.4

Background: The DRAGON score is reportedly useful for predicting the outcome of intravenous thrombolysis. This study tested whether the modified DRAGON (mDRAGON) score, in which the onset-to-treatment time is extended, can predict the long-term outcome of transient ischemic stroke (AIS) patients who are candidates for mechanical thrombectomy (MT). Methods: We assessed 40 AIS patients who were treated by MT alone or in combination with intravenous tissue plasminogen activator. The mDRAGON score is based on the following findings: hyperdense cerebral artery sign/early infarct signs on computed tomography (both=2, either=1, none=0), prestroke modified Rankin scale (mRS) score (>1=1), age (≥80 years=2, 65-79 years=1, <65 years=0), glucose level (>144 mg/dL=1), onset-to-treatment time (>230 minutes=1), and baseline National Institutes of Health Stroke Scale score (>15=3, 10-15=2, 5-9=1, 0-4=0). Receiver operating characteristics (ROC) curve analysis was used to compare the performance of the mDRAGON score with the performances of other stroke prediction scores. Results: Among 40 AIS patients treated with MT, the proportions with a good outcome (mRS score=0-2) in the groups with mDRAGON scores of 2 or 3, 4 or 5, 6 or 7, and 8-10 were 75%, 20%, 0%, and 0%, respectively, while the corresponding proportions with a poor outcome (mRS score=3-6) were 25%, 80%, 100%, and 100%. For the prediction of a good outcome at 3 months (mRS score=0-2), the area under the ROC curve of the mDRAGON scores was 0.87 (0.76-0.90). Conclusions: The mDRAGON score can be used to reliably predict the clinical outcome of AIS patients following endovascular treatment.

DMZ펀치볼둘레길 이용객 인식분석을 통한 활성화 방안: 설문조사와 텍스트마이닝 비교분석을 중심으로

김근현,이정희,이수광 (사)한국관광레저학회 2022 관광레저연구 Vol.34 No.8

The purpose of this study was to analyze the visitor's perception of the DMZ punch bowl dulle-gil through surveys and text mining, and to suggest trail activation plans by complementing the results between the two techniques. Survey data and text mining data were collected in 2020 and 2021. The survey consisted of 62 questions and 337 copies of data were analyzed using SPSS, 743 cases of Naver·Daum blog, cafe and news data were text mining analyzed using word cloud, n-gram and CONCOR. As a result, the data related to 'courses' showed that the preference for the Oyubatgil course was high in common for diverse reasons. According to the survey, it turned out that people are interested in the natural scenery of the trail they haven't previously visited. Words such as "time, distance, and intermediate" were derived high in text mining, indicating that difficulty was considered important when choosing a course. "High" indicated that difficulty was mainly considered when choosing a course.

양자수력학적 모델을 이용한 공진터널링 다이오드 전송 특성의 자기충족적 해석

金根賢,李鐘昌 弘益大學校 科學技術硏究所 2004 科學技術硏究論文集 Vol.15 No.-

The transport properties of resonant tunneling diodes were studied by using a self-consistent quantum hydrodynamic (QHD) model. This method utilized the smoothened potential model to take the non-normal incidence of electrons into account as well as the self-consistent Poisson's equation to count for the band bending effects. This method was applied to investigate the current-voltage characteristics of an Al_(0.4)Ga_(0.6)As/GaAs/Al_(0.4)Ga_(0.6)As resonant tunneling diode.

다크투어리즘 스토리텔링이 관광지 러브마크, 행동의도에 미치는 영향: 관광지 러브마크 매개효과를 중심으로

김근현,위주연,이수광 관광경영학회 2021 관광경영연구 Vol.102 No.-

The purpose of this study was to examine the effect of relationship between dark tourism storytelling, love marks from tourism, and intention of action : the mediating effect of tourism love marks. In addition, this study was proposed by the mediating effect of love mark on the relationship between dark tourism storytelling and intention of action. The empirical analysis was analyzed using the SPSS WIN Ver. 22.0 statistical package program after the questinnaire was collected, and the survey was conducted from January 20 to February 10 in 2021, and the survey method used a self-written online survey method targeting tourists who visited the DMZ in the past. A total of 181 questionnaire was collected. Looking at the analysis results, In summary of relationship between the storytelling, lovemarks and behavior intention behind Dark Tourism, ‘interest’ and ‘educational factors’ are likely to have greater impact on the creation of tourist’s lovemarks on the tourist sites, compared to other factors such as ‘comprehension’. It is attributable to the poor delivery of tour guides in sharing the history, or spatial constraints considering that the guide is provided outside. In order the overcome such constraints, the storyteller should pick plain words that are less ambiguous, and increased support should be provided such as competent staffs who are capable of grabbing visitors’ attention at all times and more sophisticated equipments like microphones. The study suggests that ‘reputation’, one of the factors behind lovemarks, having had little impact on ‘intention to revisit’ or ‘intention to recommend’is due to the visitor’s perception of DMZ, associating the place with death. In other words, the place itself has a somewhat negative image, which led to the visitors not recommending it to others. In order to address such difficulties, positive outcomes should be proven by promoting artistic or cultural advantages of historic sites, and holding musicals, plays or exhibitions of tragic history are all case in points. In the end, such endeavors could improve its overall image, providing opportunities to become proficient in history. As a result, interest in Dark Tourism, including visiting DMZ, would begin to grow if they are exposed more frequently in our daily lives, which means more recommendations and revisits would also take place. In addition to that, if the Dark Tourism were combined with a wide range of areas, rather than a single area, public awareness could be further improved.

텍스트마이닝을 활용한 코로나19 전후 제주올레 게시글 분석

김근현,이수광 (사)한국관광레저학회 2023 관광레저연구 Vol.35 No.1

The purpose of this study is to collect posts about Jeju Olle Trail uploaded to blogs and cafes (Naver, Daum) and to understand the perception and the behavior of Jeju Olle Trail users. After collecting posts with the keyword "Jeju Olle" using Textom, CONCOR analysis was performed using word cloud, N-gram, 2-way word tree, and Ucinet (ver.6.0). As a result of word cloud analysis, static words such as 'cafe, restaurant, drive course, hotel' showed high frequency in 2019, and dynamic words such as 'walk, road, pension' showed high frequency in 2020 and 2021. In particular, it was characterized by a high frequency of words related to small lodging facilities such as pensions. As a result of N-gram and 2-way word tree analysis, it was found that the frequency of crowded places has decreased, while the frequency of little-known places with fewer people has increased. As a result of the CONCOR analysis, words such as 'accommodation recommendation, one of the best' grouped together, which shows the behavior of people visiting places recommended by other people, before COVID-19, whereas lesser-known places such as 'Onpyeong-ri, Samdali, and Nandr' grouped together after COVID-19. In conclusion, the study is intended to provide basic data for tourism policies to revitalize Jeju Olle Trail in the post-COVID-19 er

빅데이터 분석기법을 활용한 10년간의 공공데이터 관련 정책 연구:정권별 비교분석을 중심으로

김근현,정성훈,양재동,위주연 중앙대학교 국가정책연구소 2023 국가정책연구 Vol.37 No.4

본 연구는 빅데이터 분석기법을 활용하여 10년간의 각 정권별(박근혜, 문재인, 윤석열) 공공데이터 관련 정책을 살펴보기 위해 수행되었으며, 방대한 양의 데이터를 수집하기에 장점이 있는 텍스트마이닝을 활용하였다. 분석방법은 Python과 R을 활용하였으며, 공공데이터 관련 언론보도 3만여 개의 데이터를 수집하였다. 분석기법으로는 워드클라우드, N그램, 감성분석, 토픽모델링, CONCOR분석 5가지를 활용하였다. 분석결과 박근혜 정부에서는 공공데이터 ‘개방’에 주안점을 두었고 ‘민간 데이터’와 관련된 단어빈도가 높게 도출되었다. 문재인 정부에서는 ‘활용’에 주안점을 둔 정책과 ‘인공지능’에 대한 단어빈도가 높게 도출되었다. 윤석열 정부에서는 공공데이터 ‘활용’에 주안점을 둔 정책과 ‘플랫폼’에 대한 단어빈도가 높게 도출됨을 확인하였다. 감성분석 결과 ‘재난, 범죄, 질병’ 등 부정적 단어들이 높게 도출되었는데, 이는 공공데이터를 활용함으로써 사회적 문제를 해결하고자 하는 목적에서 기인 된 단어라 판단된다. 토픽모델링 분석결과 ‘관리 및 평가’ 토픽의 비중 감소를 통해 품질을 향상시키고 양질의 평가 기준 마련 필요성과 ‘인프라 구축’의 중요도를 확인하였다. 마지막으로 CONCOR 분석결과 박근혜 정부에서는 ‘경진대회, 혁신, 활용’ 등의 군집이 높았고, 문재인 정부에서는 ‘뉴딜, 청년일자리, 시스템’ 등의 군집이 높았으며, 윤석열 정부에서는 ‘ 플랫폼, 지자체, 창업’ 등에 대한 군집이 높게 도출되었음을 확인하였다. 본 연구의 결과를 통해 공공데이터 이해관계자들과 학술적 목적을 가진 연구자들에게 기초자료를 제공하고자 한다. This study was conducted to examine the public data-related policies of each regime (Park Geun-hye, Moon Jae-in, and Yoon Seok-yeol) over a 10-year period using big data analysis techniques, and utilized text mining, which has the advantage of collecting a large amount of data. The analysis method utilized Python and R, and collected data from more than 30,000 public data-related media reports. Five analysis techniques were used: word cloud, N-gram, sentiment analysis, topic modeling, and CONCOR analysis. As a result of the analysis, the Park Geun-hye government focused on ‘opening’ public data, and the frequency of words related to ‘private data’ was high. In the Moon Jae-in government, policies focused on ‘utilization’ and the frequency of words related to ‘artificial intelligence’ was high. Under Yoon Seok-yeol’s government, policies focused on “utilization” of public data and the word “platform” were found to be highly frequent. The sentiment analysis showed a high frequency of negative words such as ‘disaster, crime, and disease’, which are judged to be words that originate from the purpose of solving social problems by utilizing open data. The topic modeling analysis revealed the need to improve quality by reducing the proportion of ‘management and evaluation’ topics and to establish quality evaluation criteria, and the importance of ‘infrastructure construction’. Finally, the CONCOR analysis confirmed that the Park Geun-hye government had high clusters of ‘contest, innovation, and utilization’, the Moon Jae-in government had high clusters of ‘new deal, youth jobs, and system’, and the Yoon Seok-yeol government had high clusters of ‘platform, local government, and startup’. Through the results of this study, we hope to provide basic data for open data stakeholders and researchers with academic purposes.

Te Usefulness of Proximal Radial Motor Conduction in Acute Compressive Radial Neuropathy

김근현,박기덕,정필욱,문희수,김용범,윤원태,박형준,서범천 대한신경과학회 2015 Journal of Clinical Neurology Vol.11 No.2

Background and Purpose Te objective of this study was to determine diagnostic andprognostic values of proximal radial motor conduction in acute compressive radial neuropathy. Methods Tirty-nine consecutive cases of acute compressive radial neuropathy with radialconduction studies–including stimulation at Erb’s point–performed within 14 days fromclinical onset were reviewed. Te radial conduction data of 39 control subjects were used asreference data. Results Tirty-one men and eight women (age, 45.2±12.7 years, mean±SD) were enrolled. All 33 patients in whom clinical follow-up data were available experienced complete recovery, with a recovery time of 46.8±34.3 days. Partial conduction block was found frequently(17 patients) on radial conduction studies. Te decrease in the compound muscle action potential area between the arm and Erb’s point was an independent predictor for recovery time. Conclusions Proximal radial motor conduction appears to be a useful method for the earlydetection and prediction of prognosis of acute compressive radial neuropathy.

빅데이터를 활용한 10년간의 숲길 관련 블로그ㆍ카페 이용자 인식분석

김근현(Kim, Geun-Hyeon),이수광(Lee, Su-Gwang) 한국호텔관광학회 2021 호텔관광연구 Vol.23 No.3

국가숲길에 대한 국민과 이용객 인식조사: 지리산둘레길과 대관령숲길을 중심으로

이수광,김근현,Lee, Sugwang,Kim, Geun Hyeon 한국산림과학회 2022 한국산림과학회지 Vol.111 No.1

본 연구는 국가숲길과 관련된 이해관계자가 정책을 수립하고 추진하는 데 필요한 기초자료 제공을 목적으로 일반 국민 1,200명과 국가숲길(지리산둘레길, 대관령숲길)을 이용한 경험이 있는 이용객 800명을 대상으로 국가숲길에 대한 인식분석을 실시하였다. 국가숲길 인지도는 일반국민과 이용객 모두 비슷한 수준으로 낮게 나타났으나 국가숲길 제도의 필요성, 이용의사와 추천의사는 유의한 수준으로 이용객이 보다 높게 나타났다. 국가숲길 이용목적, 선택이유, 관심사항, 문제점, 필요규제, 기대역할에 응답한 1순위는 일반국민과 이용객 간에 유사하게 나타났다. 일반국민과 이용객의 관심사항과 희망하는 활동사항 간에는 성별과 연령에 따라 유의적으로 차이가 있었으나 경관이 가장 중요한 요인으로 나타났다. 따라서 국가숲길에 올 수 있도록 하는 유인인자로 경관을 우선시하여 주요 조망점들에 대한 세부적인 특성 조사를 통해 일반국민과 이용객이 원하는 정보가 제공될 수 있는 체계가 마련되어야 한다. 이러한 결과는 국가숲길과 관련된 이해관계자에게 국가숲길 정책수립을 위한 기초자료로 활용될 것으로 판단된다. The purpose of this study was to provide the basic data necessary for stakeholders to establish and promote policies related to the national forest trail. Awareness analysis was conducted on 800 visitors to the national forest trail, specifically to the Jirisan trail and Daegwallyeong forest trail, as well as 1,200 members of the public. Awareness of the national forest trail was low and at a similar level for both visitors and the general public; however, compared with the general public, the visitors had a higher need for the national forest trail system and were willing to visit and recommend the trail. The most common answers in response to the purpose of visit, reason for choosing the national trail, matters of interest, problems, necessary regulation, and role expectations were similar among the visitors and general public. Based on gender and age, there was a significant difference in the matters of interest and desired activity, but "scenery" was the most crucial factor. Therefore, after a comprehensive survey on the major view points, given that "scenery" was identified as an attractor, a system should be developed to identify and provide the information desired by visitors and the general public. These results are expected to be employed as basic data for stakeholders in decision making related to the national forest trail.

숲길 이용난이도 산출 연구

이수광,김근현 한국산림휴양복지학회 2022 한국산림휴양학회지 Vol.26 No.1

This study was conducted to present assessment criteria and application standards that classify difficulty levels of forest trail. Assessment criteria and application standards were derived from reviews of Korean and foreign references, and AHP analysis was performed to set the suitability and weight of the derived indicators(draft). The derived indicators(draft)were applied on 18 routes(49 sections)to produce improvement plans, and on 4 routes(9 sections) to assess suitability. As a result, five assessment criteria (grade, distance, tread condition, width, and sign) and application standards were created. Subsequently, 5 levels (from level 1 “very easy” to level 5 “expert”) were invented based on the total score (15 to 51) which was calculated by multiplying application standard for each criteria by weight. After applying those on the fields, the average was calculated to be 24.2±8.9 points (11∼43 points), while the sum of “very easy”, “easy”, and “intermediate” accounts for more than 90%. Since there was no level lower than expected from the field survey or “expert” section, the standard of the existing calculation formula was adjusted to be equally spread. Therefore, a new formula was created after suitability assessment. After re-applying the new formula on the assessment result of the field application, it was found that “intermediate” decreased by 9%, “advanced” increased by 9%, and “expert” increased by 3%. The results of this study are expected to be employed as basic data for assessing difficulty level of forest trail. 본 연구는 숲길 이용난이도 등급을 구분하는 평가항목 및 등급 구분기준을 제시하고자 수행되었다. 국내·외문헌조사를 통해 평가항목 및 적용기준을 도출하였으며, 도출된 지표(안)의 적합성과 가중치 설정을 위해 AHP 기법을 적용하였다. 도출된 지표(안)을 18개 노선(49개 구간)에 현장적용하여 개선안을 도출하였으며, 개선안을 4개노선(9개 구간)에 적용하여 적합성을 평가하고 확정하였다. 그 결과 구간경사도, 구간거리, 노면상태, 노면폭, 안내표지 유무 등 5개의 평가항목과 적용기준을 도출하였으며, 각 항목별 적용기준에 가중치를 곱한 값을 더한 총점(15 점~51점)을 분할하여 1등급(매우쉬움), 2등급(쉬움), 3등급(보통), 4등급(어려움), 5등급(매우어려움)으로 이용난이도 등급을 구분하였다. 현장적용 결과 평균 24.2±8.9점(11~43점)으로 도출되었으며, 매우쉬움, 쉬움, 보통의 합계가90% 이상으로 나타났다. 현장조사 의견보다 낮은 등급으로 도출되거나 매우어려움 구간이 부재함에 따라 기존식의 점수합계 기준을 균등분할로 조정하였고, 적합성을 평가하여 개선식을 도출하였다. 개선식을 현장적용 평가결과에 재적용한 결과 보통 약 9% 감소, 어려움 약 9% 증가, 매우어려움이 약 3% 증가하였다. 본 연구결과는 숲길 이용난이도 등급 산출 시 기초자료로 활용될 것으로 기대된다.