• 제목/요약/키워드: small sample size

검색결과 737건 처리시간 0.036초

텍스트 분류 기반 기계학습의 정신과 진단 예측 적용 (Application of Text-Classification Based Machine Learning in Predicting Psychiatric Diagnosis)

  • 백두현;황민규;이민지;우성일;한상우;이연정;황재욱
    • 생물정신의학
    • /
    • 제27권1호
    • /
    • pp.18-26
    • /
    • 2020
  • Objectives The aim was to find effective vectorization and classification models to predict a psychiatric diagnosis from text-based medical records. Methods Electronic medical records (n = 494) of present illness were collected retrospectively in inpatient admission notes with three diagnoses of major depressive disorder, type 1 bipolar disorder, and schizophrenia. Data were split into 400 training data and 94 independent validation data. Data were vectorized by two different models such as term frequency-inverse document frequency (TF-IDF) and Doc2vec. Machine learning models for classification including stochastic gradient descent, logistic regression, support vector classification, and deep learning (DL) were applied to predict three psychiatric diagnoses. Five-fold cross-validation was used to find an effective model. Metrics such as accuracy, precision, recall, and F1-score were measured for comparison between the models. Results Five-fold cross-validation in training data showed DL model with Doc2vec was the most effective model to predict the diagnosis (accuracy = 0.87, F1-score = 0.87). However, these metrics have been reduced in independent test data set with final working DL models (accuracy = 0.79, F1-score = 0.79), while the model of logistic regression and support vector machine with Doc2vec showed slightly better performance (accuracy = 0.80, F1-score = 0.80) than the DL models with Doc2vec and others with TF-IDF. Conclusions The current results suggest that the vectorization may have more impact on the performance of classification than the machine learning model. However, data set had a number of limitations including small sample size, imbalance among the category, and its generalizability. With this regard, the need for research with multi-sites and large samples is suggested to improve the machine learning models.

답(沓) 이용도(利用度) 제고(提高)를 위(爲)한 조사(調査) 연구(硏究) (Investigation on the Efficient Utilization of Paddies in Korea)

  • 최범열;김영래;김문규;최창열;조재성;김달웅;김충수
    • 농업과학연구
    • /
    • 제2권1호
    • /
    • pp.151-177
    • /
    • 1975
  • 충남지방(忠南地方)의 답이용도(沓利用度) 제고(提高)를 위(爲)한 기초정보(基礎情報)를 얻기 위(爲)하여 충남(忠南)의 입지적(立地的)인 조건(條件)과 아울러 답이용현황(沓利用現況) 및 답이용도(沓利用度)의 제고(提高)를 조해(阻害)하는 요인(要人)을 분석(分析)하고 보다 생산성(生産性) 높은 작업(作業) 및 경종체계를 모색(摸索)하였던바 결과(結果)를 요약(要約)하면 다음과 같다. 1. 답전후작(沓前後作)을 기피(忌避)하는 일차적(一次的)인 원인(原因)은 논의 배수불량(排水不良)이었고 이차적(二次的)인 원인(原因)은 이앙기(移秧期)에 지장(支障)을 초래(招來)한다는 점(点)이었고 삼차적(三次的)인 원인(原因)은 노동력(勞動力) 부족(不足)이었다. 2. 답전후작(沓前後作)을 수행(遂行)하는 농가(農家)는 대부분영세농(大部分零細農)으로서 자가식량(自家食糧)을 대보를 위(爲)한 소규모재배(小規模栽培)가 많았으며 주작목(主作目)은 보리였다. 3. 충남(忠南)의 기조적(氣條的)인 조건(條件)을 고려(考慮)할 때 대체로 답전후작물(沓前後作物)의 수확기(收穫期)는 6월(月)10일(日)이 한계(限界)이며 수도(水稻)의 이앙기(移秧期)는 6월(月)25일(日)이 한계(限界)이다. 4. 답전후작(沓前後作)으로 보리를 파종(播種)할 경우(境遇) 경운전(耕耘前) 로타리살파(撒播)가 가장 능률적(能率的)이었을 뿐아니라 적은 경비(經費)가 소요(所要)되었다. 5 소득(所得)과 아울러 수확(收穫) 및 이앙작업소요기간(移秧作業所要期間)을 고려(考慮)할 경우(境遇) 일반적(一般的)으로 조생(早生)통일+올보리의 작부방식(作付方式)이 유리(有利)하였다.

  • PDF

녹내장의 침치료 효과에 대한 체계적 문헌고찰 및 메타분석 (Acupuncture for glaucoma: A systematic review and meta-analysis of randomized controlled trials)

  • 이길희;정찬영;장석주;홍승욱
    • 한방안이비인후피부과학회지
    • /
    • 제33권3호
    • /
    • pp.45-68
    • /
    • 2020
  • Objectives : This study aims to evaluate the effectiveness and safety of manual and electroacupuncture on glaucoma. Method : We searched 11 electronic databases using index words to identify randomized clinical trials. Meta-analysis of weighted mean difference (WMD) or standardized mean difference (SMD) were used to evaluate the outcomes. Cochrane bias risk assessment tool was used to assess the risk of bias in each clinical study. The collected data was analyzed using RevMan software (ver. 5.3). Results : At the initial stage of data retrieval, 549 papers were searched. After reviewing 37 full texts, a total of 10 RCT studies (426patients, 715 eyes) were selected and 8 RCT studies (357 people, 617 eyes) were involved in meta-analysis. Meta-analysis of 8 RCTs showed that acupuncture alone was more effective in reducing intraocular pressure(IOP) than conventional treatment (WMD = -5.73, 95% CI: [-12.30, 0.83], P = 0.09, I2 = 97%) The combination of acupuncture or electroacupuncture with conventional treatment was also effective in lowering IOP (WMD = -1.84, 95% CI: [-2.31, -1.37], P <0.00001, I2 = 45%). It was estimated that the combination of acupuncture with conventional treatment was also effective for improving visual field (VF) (WMD = -2.17, 95% CI: [-4.32, -0.02], P = 0.05, I2 = 89%) but improvement in visual acuity (VA) was not significant (MD = 0.06, 95% CI: [-0.03, 0.15], P = 0.23, I2 = 0%). Subgroup analyzes were performed only for the studies that used open glaucoma as the study's disease and combination of acupuncture or electroacupuncture with conventional therapy would have an effect on lowering intraocular pressure (WMD = -1.68)., 95% CI: [-2.46, -0.90], P <0.0001, I2 = 29%). Conclusion : This study suggests that acupuncture treatment for glaucoma may be effective in reducing intraocular pressure and helpful in improving visual field defects. However, due to the small sample size, high risk of bias and high heterogeneity in the methodology, it is expected that further studies will be needed to verify the results. Further studies in large-scale samples based on a minimized biased methodology would be necessary.

비정규분포공정에서 메디안특수관리도 통용모형설정에 관한 실증적 연구(요약) (Median Control Chart for Nonnormally Distributed Processes)

  • 신용백
    • 산업경영시스템학회지
    • /
    • 제10권16호
    • /
    • pp.101-106
    • /
    • 1987
  • Statistical control charts are useful tools to monitor and control the manufacturing processes and are widely used in most Korean industries. Many Korean companies, however, do not always obtain desired results from the traditional control charts by Shewhart such as the $\bar{X}$-chart, $\bar{X}$-chart, $\bar{X}$-chart, etc. This is partly because the quality charterstics of the process are not distributed normally but are skewed due to the intermittent production, small lot size, etc. In Shewhart $\bar{X}$-chart. which is the most widely used one in Kora, such skewed distributions make the plots to be inclined below or above the central line or outside the control limits although no assignable causes can be found. To overcome such shortcomings in nonnormally distributed processes, a distribution-free type of confidence interval can be used, which should be based on order statistics. This thesis is concerned with the design of control chart based on a sample median which is easy to use in practical situation and therefore properties for nonnormal distributions may be easily analyzed. Control limits and central lines are given for the more famous nonnormal distributions, such as Gamma, Beta, Lognormal, Weibull, Pareto, Truncated-normal distributions. Robustness of the proposed median control chart is compared with that of the $\bar{X}$-chart; the former tends to be superior to the latter as the probability distribution of the process becomes more skewed. The average run length to detect the assignable cause is also compared when the process has a Normal or a Gamma distribution for which the properties of X are easy to verify, the proposed chart is slightly worse than the $\bar{X}$-chart for the normally distributed product but much better for Gamma-distributed products. Average Run Lengths of the other distributions are also computed. To use the proposed control chart, the probability distribution of the process should be known or estimated. If it is not possible, the results of comparison of the robustness force us to use the proposed median control chart based oh a normal distribution. To estimate the distribution of the process, Sturge's formula is used to graph the histogram and the method of probability plotting, $\chi$$^2$-goodness of fit test and Kolmogorov-Smirnov test, are discussed with real case examples. A comparison of the proposed median chart and the $\bar{X}$ chart was also performed with these examples and the median chart turned out to be superior to the $\bar{X}$-chart.

  • PDF

한국인의 위험인지에 대한 경험적 분석 (An Empirical Review of Korean Perception for Technological Risks)

  • 정익재
    • 한국안전학회지
    • /
    • 제22권6호
    • /
    • pp.91-97
    • /
    • 2007
  • 본 연구는 기술위험에 대한 한국인의 인지수준을 경험적으로 분석하기 위하여 설문조사(표본 크기 1,870)를 실시하고, 응답자의 사회인구학적 변수를 배경으로 그 특성을 정리하였다. 설문에서 교통, 유해화학물질, 환경, 산업안전, 원자력 그리고 새로운 기술 등 6개 분야의 25개 위험에 대한 상대적인 위험수준을 평가하였다. 요인분석 결과, 응답자의 위험인지에서 독특한 행태적 특성을 발견하였다. 통계에 거한 객관적 위험평가와 주관적 위험인지는 뚜렷한 차이를 보이며, 응답자의 사회인구학적 변수는 이러한 차이를 의미있게 설명하고 있다. 예를 들면, 중 소도시에 거주하는 저소득 저학력의 30-40대 기혼 여성이 다른 사회집단 구성원보다 위험에 민감한 반응을 보였으며, 생소하거나 막연한 대상의 위험 수준을 높이 평가하는 경향이 있다. 이러한 연구결과는 위험인지에서 나타나는 개인 차원의 오류와 편견을 줄이고, 위험관리 정책과 안전규제를 효과적으로 집행하는데 요구되는 기반자료로서 활용할 수 있다. 특히, 위험인지의 사회집단별 차별성은 안전과 관련된 과학적인 지식과 정보를 누구에게 어떻게 전달할 것인지에 대한 정책적 함의를 제공한다. 현대사회의 위험관리는 기술공학적 접근과 더불어 사회 문화적 변수를 고려하여 추진되어야 한다는 점을 재확인한다.

"AMPQ-II 및 관리 매뉴얼"에 따른 학교 상담의 효과: 상담자 요인 및 회기 수를 중심으로 (Effectiveness of school counseling based on "the AMPQ-II and administrative manual": Focusing on the counselor and the number of session factors)

  • 설지원;김근영
    • 한국산학기술학회논문지
    • /
    • 제16권2호
    • /
    • pp.978-986
    • /
    • 2015
  • 정부는 학생의 정서행동적 문제를 예방하고, 위기에 즉각 개입하기 위해 국가적 차원에서 "학생정서행동특성 검사(AMPQ-II) 및 관리매뉴얼"을 통해 개입을 실시하고 있다. 하지만 국가의 개입이 실제로 효과가 얼마나 있는지, 어떠한 요인들이 개입효과에 영향을 미치는지에 대한 경험적 연구는 거의 없는 상황이다. 본 연구는 K지역 2개 중학교를 대상으로 48명의 관리대상 중학생이 개입 후 심리적 상태의 변화를 경험하였는지, 그리고 상담자의 자격증 종류와 상담 회기 수에 의해 그 효과가 다른지를 탐구하였다. 분석 결과 대다수의 학생들은 개입 이후 긍정적인 변화를 보고하는 것으로 드러났다. 국가공인 자격집단의 경우 민간 자격집단에 비해 개입효과가 떨어졌으며, 회기 수 수준별 개입효과의 차이는 발견되지 않았다. 본 연구의 결과 해석은 적은 표본수로 인해 조심스럽게 접근할 필요가 있다. 그럼에도 불구하고 학교상담의 핵심인력인 전문상담교사 및 국가공인 상담자들의 전문성 제고를 위한 노력이 필요함을 함의하였으며, 단순히 상담 회기 수를 늘리는 것은 의미가 없음을 시사한다는 점에서 추후 학교상담의 제도개선과 관련된 논의에 큰 영향을 미칠 것으로 기대된다.

잠업단지의 경제효율에 관한 비교분석 (Comparative Analysis of Economic Efficiency by Major Sericultural Farming Areas in Korea)

  • 이질현;김문협;강석권
    • 한국잠사곤충학회지
    • /
    • 제14권2호
    • /
    • pp.95-103
    • /
    • 1972
  • The major purpose of this study is to collect the information related on the aspects of economic efficiency for solving the problems which are faced by farmers and areas, and providing scientific facts to farmers and related institutions for further development of sericultural sector in Korea. In order for obtaining the related information 12 sample areas among 23 major sericultural farming areas and 30 farm units in each area are selected and analyzed in this study. The fold suevey is made by member of this study team and graduate students in the Department of Sericultural Science with a prepared questionnaires. Cross-section and regression analysis methods are employed for processing the data in this study. The major findings obtained are as followings. 1. Sericultural earnings per Tanbo is, on the average, 22, 752 won in new cultivated areas and 29, 403 won in ordinary ones. There are big difference in the size of earnings by areas, especially, 46, 968 won in Kumo mountain area, compared with 16, 798 won in Yeoju and Yichun areas. General trend is finded that small scale farming units are made higher earnings and operating their farms efficiently. 2. Cocoon production expences per Tanbo is 16, 737 won in new cultivated areas and 19, 802 won in ordinary areas. There are also big difference in farming expences, especially, 27, 389 won in Sudang area, compared with 11, 689 won in Emjin area. 3. Sericultural income per Tanto is 10, 664 won in ordinary areas and 6, 898 won in new cultivated areas. Farmers in Kumo mountain area make the highest income of 21, 164 won and lowest income of 1, 296 won in Sudang area. It can be generized that about 30-50 a sized farmers make higher income. 4. Land, labor and capital productivities estimated by fitting Cobb-Douglas functions in ordinary areas are higher than in new cultivated areas, especially, labor productivity is higher in ordinary areas. 5. Changsung, Kwangna, Yunsun and Kumo mountain areas are technically and economically efficient. Sudang and Mujinchang areas are technically successful but economically inefficient and Emjin and Honam areas are technically inefficient but economically efficient. YeojuYichun, Chunwon and West Kyongnam are technically and economically inefficient. Technical and economic improvement program should be implemented for these areas. 6. Estimated Internal Rate of Return (IRR) on capital investment in Chongwon are is 23.5 percent. It is economically feasible, if we consider 20 percent of opportunity cost of capital in our economy.

  • PDF

Use of radiotherapy in patients with palliative double bypass for locally advanced pancreatic adenocarcinoma

  • Glinka, Juan;Diaz, Federico;Alva, Augusto;Mazza, Oscar;Claria, Rodrigo Sanchez;Ardiles, Victoria;Santibanes, Eduardo de;Pekolj, Juan;Santibanes, Martin de
    • Radiation Oncology Journal
    • /
    • 제36권3호
    • /
    • pp.210-217
    • /
    • 2018
  • Purpose: Pancreatic cancer (PC) has not changed overall survival in recent years despite therapeutic efforts. Surgery with curative intent has shown the best long-term oncological results. However, 80%-85% of patients with these tumors are unresectable at the time of diagnosis. In those patients, first therapeutic attempts are minimally invasive or surgical procedures to alleviate symptoms. The addition of radiotherapy (RT) to standard chemotherapy, ergo chemoradiation, in patients with locally advanced pancreatic cancer (LAPC) is still controversial. The study aims to compare outcomes in patients with a double bypass surgery due to LAPC treated or not with RT. Materials and Methods: A retrospective cohort study of patients with double bypass for LAPC were registered and divided into two groups: treated or not with postoperative RT. Baseline characteristics, postoperative complications, those related to RT and their relation to the main event (mortality) were compared. Results: Seventy-four patients were included. Surgical complications between the groups did not offer significant differences. Complications related to RT were mostly mild, and 86% of patients completed the treatment. Overall survival at 1 and 2 years for patients in the exposed group was 64% and 35% vs. 50% and 28% in the non-exposed group, respectively (p = 0.11; power 72%; hazard ratio = 0.53; 95% confidence interval, 0.24-1.18). Conclusion: We observed a tendency for survival improvement in patients with postoperative RT. However, we've not had enough power to demonstrate this difference, possibly due to the small sample size. It is indispensable to develop randomized and prospective trials to guide more specific treatment lines in this patients.

고지혈성 급성 췌장염에 대한 대시호탕의 효과 : 체계적 문헌고찰과 메타분석 (The Effect of Dachaihu Decoction for Hyperlipidemic Acute Pancreatitis: A Systematic Review and Meta-Analysis)

  • 김윤정;정유진;박동일
    • 대한한방내과학회지
    • /
    • 제41권3호
    • /
    • pp.306-325
    • /
    • 2020
  • Objectives: The aim of this study is to investigate the effect of a Dachaihu decoction for hyperlipidemic acute pancreatitis (HLAP) by systematic review and meta-analysis of Chinese clinical studies. Methods: China National Knowledge Infrastructure (CNKI) was utilized as the major search engine. The date of the literature search was March 7, 2020. Randomized controlled trials (RCTs) about using a Dachaihu decoction for HLAP were included in this study. Meta-analysis was performed by synthesizing outcome data, including total effective rate, abdomen pain relief time, first bowel movement time, blood amylase recovery time, and triglyceride (TG) levels (mmol/L). The selected literature was assessed using Cochrane's risk of bias (RoB). Results: Twelve of 44 RCTs met the inclusion criteria. Most studies were evaluated with RoB as having unclear risk. The total effective rate of herbal medicine treatment based on the Dachaihu decoction was significantly higher than that of symptomatic supportive treatment in 10 articles (risk ratio=1.15, 95% CI: 1.08 to 1.21, p<0.00001, I2=0%). Herbal medicine treatment based on a Dachaihu decoction was significantly more effective than symptomatic supportive treatment in terms of reducing abdomen pain relief time (in all articles; mean difference=-1.70, 95% CI: -1.91 to -1.41, p<0.00001, I2=45%), first bowel movement time (in 7 articles; mean difference=-1.46, 95% CI: -1.86 to -1.05, p<0.00001, I2=73%), blood amylase recovery time (in 8 articles; mean difference=-1.48, 95% CI: -2.04 to -0.92, p<0.00001, I2=90%), and TG levels (in 8 articles; mean difference=-1.59, 95% CI: -2.28to -0.91, p<0.00001, I2=90%). Only one article reported side effects of treatment among the intervention group and control group, citing pancreatic ulcer and pancreatic pseudocyst formation. Conclusions: This study suggests that herbal medicine treatment based on a Dachaihu decoction could yield higher efficacy for HLAP than symptomatic supportive treatment alone. However, the results might be somewhat biased because of the poor quality and small sample size of the included RCTs. Well-qualified clinical studies are needed to prove the effectiveness of Dachaihu decoction therapy for HLAP.

벵갈만 지역의 컨테이너항만 선택 기준에 관한 연구 (Empirical Analysis of Selection Criteria of Container Ports in the Bay of Bengal)

  • 뎅기르윈;김현덕
    • 한국항만경제학회지
    • /
    • 제34권4호
    • /
    • pp.69-84
    • /
    • 2018
  • 본 연구의 목적은 스리랑카의 콜롬보항만, 인도의 첸나이항만, 방글라데시의 치타공항만 그리고 미얀마의 양곤항만을 포함한 벵갈만 지역의 주요 4대 컨테이너항만의 지역 허브항만 선택기준에 대해 실증 분석하는데 있다. 연구 목적을 달성하기 위해 우선 항만선택기준에 관한 선행연구를 실시하였고, 선행 연구를 통해 도출된 항만선택기준을 전문가 자문을 통해 분류하였다. 이를 바탕으로 항만의 이용자인 해운회사. 프레이트 포워더, 물류서비스 제공자 그리고 항만물류전문가를 대상으로 설문지를 배포하였다. 연구방법론으로 AHP가 사용되었다. 주요 연구결과를 요약하면 다음과 같다. 첫째, 항만의 효율성이 가장 중요한 항만선택기준으로 평가되었다. 다음으로는 항만 비용, 지리적 위치 그리고 항만시설 순으로 나타났다. 둘째, 항만의 상대적 평가에서는 콜롬보항만이 항만 배후단지를 제외한 네 요인에서 중요한 것으로 분석되었다. 본 논문은 벵갈만 지역의 허브항만 선택 기준에 관한 기초 연구를 실시하였다는 점에서 연구의 의의가 있다. 그럼에도 불구하고 전문가를 대상으로 한 설문 표본이 작다는 점과 벵갈만 지역 항만에 대한 폭넓은 지식을 가진 전문가를 찾아내기가 쉽지 않다는 점이다. 향후, 이런 점을 보완할 수 있는 연구를 수행해야 할 것이다.