Drug Toxicity Evaluation using Ordinal Logistic Regression with Multi-Features from ORd and To R – ORd In-silico Ventricular Cell Model

Nurul Qashri Mahardika T,Ali Ikhsanul Qauli,박혜림,Aroli Marcellinus,임기무 한국동물실험대체법학회 2024 동물실험대체법학회지 Vol.18 No.1

Several drugs have been withdrawn from the market since their potential to cause Torsade de Pointes (TdP), a potentially fatal form of ventricular arrhythmia. To mitigate this risk, the Comprehensive in Vitro Proarrhythmia Assay (CiPA) proposes assessing the arrhythmogenic potential of drugs via in-silico simulations based on pharmacological data obtained in vitro. Various studies have utilized in-silico models with machine learning algorithms to classify TdP risks and yield promising results. In this study, we applied an ordinal logistic regression approach to assess TdP risk using 364 feature pairs derived from 14 features of the modified ORd and ToR-ORd models. This method allowed us to analyze drug-induced features and classify TdP risk levels. Ordinal logistic regression enabled us to explore complex relationships between these features and TdP risk levels. Notably, combining under the ToR-ORd model with under the ORd model achieved excellent performance, with Area Under the Curve (AUC values of 0.98 for high-risk and 0.92 for low-risk categories. These findings suggest that our approach can significantly enhance the understanding and assessment of TdP risk, contributing to developing safer drugs for clinical use.

Goodness-of-fit tests for a proportional odds model

Hyun Yung Lee 한국데이터정보과학회 2013 한국데이터정보과학회지 Vol.24 No.6

The chi-square type test statistic is the most commonly used test in terms of mea-suring testing goodness-of-fit for multinomial logistic regression model, which has its grouped data (binomial data) and ungrouped (binary) data classified by a covariate pattern. Chi-square type statistic is not a satisfactory gauge, however, because the ungrouped Pearson chi- square statis-tic does not adhere well to the chi-square statistic and the ungrouped Pearson chi-square statistic is also not a satisfactory form of measurement in itself. Currently, goodness-of-fit in the ordinal setting is often assessed using the Pearson chi-square statistic and deviance tests. These tests involve creat-ing a contingency table in which rows consist of all possible cross-classifications of the model covariates, and columns consist of the levels of the ordinal response. I exam-ined goodness-of-fit tests for a proportional odds logistic regression model-the most commonly used regression model for an ordinal response variable. Using a simulation study, I investigated the distribution and power properties of this test and compared these with those of three other goodness-of-fit tests. The new test had lower power than the existing tests; however, it was able to detect a greater number of the different types of lack of fit considered in this study. I illustrated the ability of the tests to detect lack of fit using a study of aftercare decisions for psychiatrically hospitalized adolescents.

Goodness-of-fit tests for a proportional odds model

Lee, Hyun Yung The Korean Data and Information Science Society 2013 한국데이터정보과학회지 Vol.24 No.6

The chi-square type test statistic is the most commonly used test in terms of measuring testing goodness-of-fit for multinomial logistic regression model, which has its grouped data (binomial data) and ungrouped (binary) data classified by a covariate pattern. Chi-square type statistic is not a satisfactory gauge, however, because the ungrouped Pearson chi-square statistic does not adhere well to the chi-square statistic and the ungrouped Pearson chi-square statistic is also not a satisfactory form of measurement in itself. Currently, goodness-of-fit in the ordinal setting is often assessed using the Pearson chi-square statistic and deviance tests. These tests involve creating a contingency table in which rows consist of all possible cross-classifications of the model covariates, and columns consist of the levels of the ordinal response. I examined goodness-of-fit tests for a proportional odds logistic regression model-the most commonly used regression model for an ordinal response variable. Using a simulation study, I investigated the distribution and power properties of this test and compared these with those of three other goodness-of-fit tests. The new test had lower power than the existing tests; however, it was able to detect a greater number of the different types of lack of fit considered in this study. I illustrated the ability of the tests to detect lack of fit using a study of aftercare decisions for psychiatrically hospitalized adolescents.

농업인 손상조사 결합자료 생성 및 분석

이서현,김소라,박진우,이석훈 한국자료분석학회 2018 Journal of the Korean Data Analysis Society Vol.20 No.3

The 'Korean agricultural workers’ occupational disease and injury survey', which the rural development administration has been carrying out since 2009, is used as basic data for establishing policies for preventing and managing injuries by identifying the cause and magnitude of injuries caused by farmers' agricultural work. The survey is conducted annually since 2012, but in-depth investigations of disease and injury are being conducted every two years. In this study, we conducted a matching of survey data on occupational injury of farmers collected across three years in 2013, 2015, and 2017, so that we could better utilize the results of the survey. Based on the matched data for three years, we obtained information about the experience of injury for three years, and analyzed the factors influencing it. The random forest model and the ordinal logistic model were used for the analysis. 농촌진흥청이 2009년부터 수행하고 있는 ‘농업인의 업무상 질병 및 손상조사’는 농업인들의 농작업으로 인한 재해 원인과 규모를 파악하여 재해 예방과 관리를 위한 정책 수립의 기초자료로 활용되고 있다. 이 조사는 2012년부터 매년 수행되고 있으나 질병과 손상에 관한 심층조사는 격년으로 수행되고 있다. 본 연구에서는 2013년과 2015년, 2017년에 걸쳐 횡단적으로 수집된 농업인의 업무상 질병 및 손상조사 자료를 분석하여 향후 손상조사가 질병 및 손상을 예방하기 위한 보다 효과적, 효율적인 정책 수립에 활용될 수 있는 가능성을 조사하였다. 결합된 3개년의 자료를 바탕으로 단년도 분석에서는 알 수 없는 3개년도 손상 경험 여부에 관한 정보를 얻고, 이에 영향을 미치는 요인을 분석하여 보았다. 분석에는 랜덤포레스트 모형과 순서 로지스틱 모형을 사용하였고, 분석 결과 손상의 예측에 중요도가 높은 상위 5개 변수들은 농업활동 제한여부와 층, 연령, 농업 종사기간, 가구소득인 것으로 나타났다.

성대 결절 및 폴립 병변 판별 예측모형에 대한 연구

박수정,심현섭,정성민,김한수,박애경,Park, Soo-Jung,Shim, Hyun-Sup,Chung, Sung-Min,Kim, Han-Soo,Park, Ae-Kyung 대한후두음성언어의학회 2004 대한후두음성언어의학회지 Vol.15 No.2

Background and Objectives : This study is purposed to investigate the statistically significant discrimination model for predicting vocal fold nodule and polyp's lesional grade, with patients' background data and objective voice evaluation parameters. Materials and Method : The retrospective research was carried out at the Ewha Womans University Hospital. 122 patients' voice examination data had been selected, and lesion screening (Grade I, II, and III) was conducted by 2 ENT specialists, with each patient's vocal fold pictures achieved during the laryngoscopy examination. Results : The Lesional Grade Discrimination Model with which the lesional grade of vocal fold nodules and polyps could be predicted was derived by the ordinal logistic regression analysis (using SPSS 10.0). With this model the lesional grades of 73 out of 122 patients(59.8%) were correctly predicted to their formerly screened ones. Conclusion : This model applied the multivariate approach, which statistically combined these currently used parameters, Jitter, Shimmer, MFR, MPT, and patient's background data such as gender and dysphonia period. It might explain the status of benign lesion of vocal folds, and furthermore expect the physiological function of vocal folds.

자율주행차의 사회적 수용성에 영향을 미치는 사회·경제적 요인

조선아,김규옥 한국교통연구원 2020 交通硏究 Vol.27 No.3

The purpose of this study is to derive socio-economic parameters that have major effects on public acceptance on autonomous vehicles. Currently, advanced countries vigorously supporting vehicle technology development to stimulate its’ commercialization which raising concerns on social acceptance of the technology. To examine the perception, authors utilized survey results conducted by recruiting 912 respondents consisted of three groups depending on their expertise in vehicles technology, transportation engineers and vocational drivers, and general public. Ordinal logistic regression model was employed for analyzing the data and drew statistically significant socio-economic factors. The results show that gender, age and prior experiences of car accidents were significantly affects responders perception on autonomous vehicles. Authors expect that results of this research could contribute to the field by providing socio-economic factors that have effects on public acceptance on autonomous vehicle. 본 논문의 목적은 자율주행차의 사회적 수용성에 영향을 미치는 사회·경제적 요인을 파악하는 것이다. 현재 각국은 자율주행차의 상용화를 촉진하기 위해 기술개발과 정책지원을 하고 있다. 새로운 자율주행 기술이 자동차에 접목되면서 사회적 수용성의 중요성이 커지고 있다. 본 논문에서는 자율주행차의 사회적 수용성에 영향을 미치는 요인을 도출하고자 912명을 대상으로 일반인, 전문가, 직업 운전자 등 3개 그룹으로 구분하여 설문조사를 실시하였다. 설문조사 결과를 기반으로 순서형 로지스틱 회귀분석 모델을 개발·적용하여 사회·경제적 요인 중에서 통계적으로 유의한 요인을 도출하였다. 분석 결과, 사회경제적 지표인 성별, 연령, 자동차 사고 경험이 자율주행차에 대한 수용성에 미치는 영향이 큰 것으로 나타났다. 본 연구는 자율주행차에 대한 일반 국민의 인식과 수용성에 영향을 미치는 요인을 제시하였다는 데 그 의의가 있다.

친환경농산물 소비의 건강증진 기여 인식도 분석

정학균,김창길,문동현 한국유기농업학회 2012 韓國有機農業學會誌 Vol.20 No.2

The purposes of this study are to analyze the effect of consumption of environment- friendly agricultural products (EFAP) on improvement of family health, and to suggest directions for improvement of family health. A survey was conducted for qualitative analysis regarding relationship between EFAP consumption and family health. The method of his study was employed Cross-tabulation and an Ordinal Logistic Regression Model to derive more significant results in analyzing factors of improvement of family health. The result shows that improvement of health has a significant positive relationship with consumption of EFAP. In addition, those consumers with high reliability and quality contentment are more likely to experience improvement of health. As consumers constantly eat EFAP, they are more likely to experience improvement of health. In order to provide consumer reliability of EFAP, more strict certification management system with sound monitoring and an appropriate penalty for violation should be established

친환경농산물 소비의 건강증진 기여 인식도 분석

정학균,김창길,문동현,Jeong, Hak-Kyun,Kim, Chang-Gil,Moon, Dong-Hyun 한국유기농업학회 2012 한국유기농업학회지 Vol.20 No.2

The purposes of this study are to analyze the effect of consumption of environment-friendly agricultural products (EFAP) on improvement of family health, and to suggest directions for improvement of family health. A survey was conducted for qualitative analysis regarding relationship between EFAP consumption and family health. The method of his study was employed Cross-tabulation and an Ordinal Logistic Regression Model to derive more significant results in analyzing factors of improvement of family health. The result shows that improvement of health has a significant positive relationship with consumption of EFAP. In addition, those consumers with high reliability and quality contentment are more likely to experience improvement of health. As consumers constantly eat EFAP, they are more likely to experience improvement of health. In order to provide consumer reliability of EFAP, more strict certification management system with sound monitoring and an appropriate penalty for violation should be established.

실사고 데이터 기반 자율주행자동차 교통사고 심각도 영향요인 분석

박기옥,이요셉,이현미,장정아,윤일수 한국ITS학회 2024 한국ITS학회논문지 Vol.23 No.6

비자율주행자동차 교통사고와 달리 자율주행자동차가 관여된 교통사고는 예측하기 어려운 위험요소가 존재할 수 있다. 자율주행자동차 상용화 관련 규정 및 제도적 장치 마련을 위해서 는 자율주행자동차 교통사고에 영향을 미치는 요인들을 분석하고 정량화하는 것이 선행되어 야 한다. 따라서, 본 연구에서는 넓은 공간적 범위와 표본 수를 가진 미국 도로교통안전국의 자율주행자동차 교통사고 데이터를 활용하여 교통사고 발생 현황 및 특성 분석과 순서형 로지 스틱 회귀 모형 기반 심각도 영향요인 분석을 수행하였다. 교통사고 심각도 영향요인의 도출 결과, 주행 거리, 교통사고 발생 전 주행 행태, 충돌 대상물이 교통사고 심각도에 영향을 미치 는 것으로 나타났다. 본 연구는 실사고를 기반으로 자율주행자동차 교통사고의 심각도에 영향 을 미치는 요인들을 확인했다는 점에서 의의가 있으며, 도출된 영향 요인들은 향후 자율주행 자동차 교통사고 조사, 관련 규정 및 제도적 장치 마련에 활용될 수 있을 것으로 기대한다. Unlike non-automated vehicle accidents, traffic accidents involving automated vehicle(AV) may involve various unpredictable risk factors. Therefore, it is crucial to identify and quantify the factors affecting AV accidents to establish relevant regulations and institutional frameworks. This study utilizes AV data published by the U.S. National Highway Traffic Safety Administration, which provides a large sample size and wide spatial coverage, to analyze the occurrence and characteristics of AV accidents. Additionally, an ordinal logistic regression model is employed to assess the factors influencing accident severity. The results indicate that AV driving distance, driving behavior before the accident, and the type of collision object significantly affect accident severity. This research is meaningful as it identifies the key factors influencing the severity of AV accidents based on real-world data. The findings are expected to serve as a foundation for future investigations into AV accidents, as well as for the development of related regulations and institutional measures.

중등교사들의 교육과정-수업-평가-기록 일체화 실행수준 점검 및 자기평가 분석

이재창(Lee Jae Chang),최정원(Choi Jeong Won) 학습자중심교과교육학회 2021 학습자중심교과교육연구 Vol.21 No.10

목적 본 연구는 중학교와 고등학교 교사를 대상으로 교육과정-수업-평가-기록 일체화 실행수준을 파악하고 일체화 실행수준에 대해 스스로 어떠한 평가를 하고 있는지 확인하였다. 또한, 교사들의 교육경력, 학교급, 일체화 관련 연수 이수 경험, 일체화 관련 교사학습공동체 활동 경험과 교육과정-수업-평가-기록 일체화 관련 실행 사이의 상관관계를 분석하였다. 더불어 이러한 배경변인들이 일체화 실행에 미치는 직접적인 효과를 살펴보았다.방법 설문조사는 관심기반실행모형에서 제시하는 실행수준 척도를 수정하여 사용함으로써 실행수준을 확인하였고, 선행연구에서 제시한 교육과정-수업-평가-기록 일체화의 절차를 토대로 본 연구자가 개발한 문항을 이용하여 각 요소별 실행수준에 대한 자기평가를 확인하였다. 설문진행은 2019년 8월부터 2020년 1월까지 이루어졌으며 대상 학교는 ○지역 중학교 5개교, 고등학교 5개교였다. 중학교 교사 110명, 고등학교 교사 132명으로 총 242명의 응답에 대해 분석하였다. 분석방법은 기술통계, 교차분석, 상관분석, 순서형 로지스틱 회귀분석을 사용하였다.결과 주요 연구결과로는 첫째, 학교급에 상관없이 교사들의 실행수준이 전반적으로 낮았다. 그리고 고급수준인 갱신단계(6수준)에 이르는 비율이 극히 적은 것으로 파악되었다. 둘째, 중학교 교사들은 학습자중심 수업 수행여부와 일체화에서의 모든 요소들을 연계하여 운영한다는 응답비율에서도 고등학교 교사보다 더 높은 비율을 보였다. 셋째, 일체화 관련 교사학습공동체 활동 경험이 있는 교사들은 일체화 요소들을 더 긴밀하게 연계하는 것으로 나타났다. 넷째, 교육경력, 일체화 관련 연수, 관련 교사학습공동체 활동은 모두 실행수준 0수준에 통계적으로 유의미한 부적 효과를 보이고 3, 4, 5수준에는 통계적으로 유의미한 정적효과를 보이는 것으로 밝혀졌다.결론 제언사항은 첫째, 중등교사들의 교육과정-수업-평가-기록 일체화 관련 실행역량을 강화하기 위하여 교사 연수나 교사학습공동체와 같은 전문성 신장을 위한 기회를 확대하여 제공할 것, 둘째, 연수와 교사학습공동체 활동에 있어서 목표, 내용, 주제, 수준을 다양화하고 각각을 전문화⋅차별화하여 구성할 것, 셋째, 교육과정-수업-평가-기록 일체화에서 수업이 나머지 요소들에 미치는 영향 및 수업방식과 일체화 요소 간 연계성에 대한 심도 있는 후속 연구를 실시할 것을 제시하였다. Objectives This study identified the level of use on the alignment with curriculum - instruction - evaluation - record and self-evaluation with teachers in middle and high schools. In addition, the study analyzed correlation between teacher s educational career, level of school, experience in alignment-related training and experience in alignment-related teacher learning community activities and alignment with curriculum - instruction - evaluation - record. It also reviewed the direct effect of these background variables on adoption with alignment. Methods The survey identified overall adoption level by modifying and using scale of adoption level presented by concerns-based adoption model, and teacher’s self-evaluation was identified by using questions developed this researcher based on the procedures of alignment with curriculum - instruction - evaluation - record presented by pilot study. The survey was conducted from August 2019 to January 2020 and the subject schools included 5 middle schools and 5 high schools in the area of ○. A total of 242 responses were analyzed including 110 middle school teachers and 132 high school teachers. Descriptive statistics, crossover analysis, correlation analysis and ordinal logistic regression were used for analysis. Results The results of the study are as follows. First, regardless of level of school, the teacher’s level of adoption of the alignment with curriculum-instruction-evaluation-record are overall low. And there are very few teacher who reach the renewal level(level six). Second, middle school teather’s response rate are higher than high school teather’s not only student centered instruction factor but also connecting all the element of the alignment with curriculum-instruction-evaluation-record. Third, the teacher who has participated in alignment-related teacher learning community activities connects all the element of the alignment with curriculum-instruction-evaluation- record more. Fourth, teacher s educational career, experience in alignment-related training and experience in alignment-related teacher learning community activities have negative effectiveness in level 0 and positive effectiveness in level 3, 4, 5. Conclusions The suggestions were given as follows. First, opportunities to develop professionalism such as teacher s training and teacher s learning community must be expanded and provided to strengthen adoption capacity related to alignment with curriculum - instruction - evaluation - record of secondary school teachers. Second, goals, contents, topics and levels must be diversified in training and activities of teacher s learning community and each of them must be specialized and differentiated. Third, in-depth follow-up study on the effect of instruction on remaining factors in the alignment with curriculum - instruction - evaluation - record and on connectivity between factors of alignment must be conducted.