• 대한물리치료과학회지
• /
• 제27권2호
• /
• pp.25-35
• /
• 2020

Background: There are many situations where walking in an actual community needs to change direction along with walking on a straight path, and this situation needs to be reflected in assessing walking ability of the community. Therefore, in this study, we tried to determine whether the assessments can distinguish the level of walking in the community. Design: Retrospective cohort study. Methods: Fifty-two survivors with chronic stroke have participated in the study. According to the evaluation result of 10mWT, the subjects of 0.8m/s and above were classified as the group who could walk in the community (n=22), and the subjects of 0.4m/s~0.8m/s were classified into the group who could not walk in the community (n=30). Modified Rivermead Mobility Index, Postural Assessment Scale for Stroke, Fugl-Meyer Assessment, Berg Balance Scale, 10-meter Walk Test (10mWT) were used to evaluate the motor skills. Furthermore, Activities-specific Balance Confidence Scale was used to evaluate psychological factors, and Timed Up & Go Test (TUG), Figure-of-Eight Walk Test (F8WT), Four Square Step Test (FSST), Step Test (ST) were applied to evaluate dynamic balance and mobility. Results: As a result for distinguishing walking levels in the community, TUG was 14.25 seconds, F8WT was 13.34 seconds, FST was 19.43 seconds, and ST of affected side and non-affected side were 6.5 points and 7.5 points, respectively. TUG (AUC=0.923), F8WT (AUC=0.905), and FST (AUC=0.941) were highly accurate, but the ST of affected side and non-affected side (AUC=0.806, 0.705) showed the accuracy of the median degree, respectively. Conclusion: To distinguish walking levels in the community of survivors with chronic stroke, TUG and FSST have been found to be the best assessment tool, and in particular, FSST could be very valuable in clinical use as the most important assessment tool to distinguish walking levels in the community.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2011년도 학술발표회
• /
• pp.14-14
• /
• 2011

Groundwater in the Waikatoregion is a valuable resource for agriculture, water supply, forestry and industries. The 434,000 ha study area comprises the upper Waikato River catchment from the outflow of Lake Taupo (New Zealand's largest lake) through to Lake Karapiro (a man-made hydro lake with high recreational value) (Figure 1). Water quality in the area is naturally high. However, there are indications that this quality is deteriorating as a result of land use intensification and deforestation. Compounding this concern for decision makers is the lag time between land use changes and the realisation of effects on groundwater and surface water quality. It is expected that the effects of land use changes have not yet fully manifested, and additional intensification may take decadesto fully develop, further compounding the deterioration. Consequently, Environment Waikato (EW) have proposed a programme of work to develop a groundwater model to assist managing water quality and appropriate policy development within the catchment. One of the most important and critical decisions of any modelling exercise is the choice of the modelling platform to be used. It must not inhibit future decision making and scenario exploration and needs to allow as accurate representation of reality as feasible. With this in mind, EW requested that two modelling platforms, MODFLOW/MT3DMS and FEFLOW, be assessed for their ability to deliver the long-term modelling objectives for this project. The two platforms were compared alongside various selection criteria including complexity of model set-up and development, computational burden, ease and accuracy of representing surface water-groundwater interactions, precision in predictive scenarios and ease with which the model input and output files could be interrogated. This latter criteria is essential for the thorough assessment of predictive uncertainty with third-party software, such as PEST. This paper will focus on the attributes of each modelling platform and the comparison of the two approaches against the key criteria in the selection process. Primarily due to the ease of handling and developing input files and interrogating output files, MODFLOW/MT3DMS was selected as the preferred platform. Other advantages and disadvantages of the two modelling platforms were somewhat balanced. A preliminary regional groundwater numerical model of the study area was subsequently constructed. The model simulates steady state groundwater and surface water flows using MODFLOW and transient contaminant transport with MT3DMS, focussing on nitrate nitrogen (as a conservative solute). Geological information for this project was provided by GNS Science. Professional peer review was completed by Dr. Vince Bidwell (of Lincoln Environmental).

• 한국식품과학회지
• /
• 제53권5호
• /
• pp.657-668
• /
• 2021

본 연구는 일반우유와 무지방우유에서 장출혈성 대장균과 캠필로박터 제주니의 행동예측모델을 개발하고, 미생물학적 안전관리를 위한 기준의 적절성 평가를 위해 정량적 위해성평가를 수행하였다. 시중 마트에서 유통 판매되고 있는 일반우유(n=195)에서 장출혈성 대장균과 캠필로박터 제주니의 오염실태를 모니터링한 결과 모든 제품에서 장출혈성 대장균과 캠필로박터 제주니는 검출되지 않아 초기 오염도는 각각 -3.94 log CFU/mL로 동일하게 추정되었다. 장출혈성 대장균은 7℃ 이상의 온도에서 성장하였고, 캠필로박터 제주니는 4-25℃ 온도의 우유에서 사멸하였다. 우유에서 1차 모델에서 얻은 parameter를 사용하여 장 출혈성 대장균은 2차 성장모델을 캠필로박터 제주니는 2차 사멸예측모델을 개발하였다. 일반우유의 섭취패턴은 식품의약품안전처(2015) 연구에서 수행한 "50대 주요 축산식품의 섭취량 및 섭취패턴조사" 결과를 바탕으로 @RISK 프로그램을 활용하여 하루에 일반우유의 1회 섭취를 통하여 장출혈성 대장균과 캠필로박터 제주니에 의한 식중독 발생 확률을 추정하였다. 추정 결과 1일 1회 일반우유 섭취로 장출혈성 대장균과 캠필로박터 제주니로 인한 평균 식중독 발생 확률은 각각 5.70×10^-5, 9.86×10^-9 것으로 확인되었다. 본 연구에서 정량적 위해평가를 통해 일반우유에서 장출혈성 대장균과 캠필로박터 제주니의 위해수준을 산출한 결과 일반우유에서 장출혈성 대장균의 식중독 발생 가능성이 상대적으로 높으므로 우선관리 대상임을 알 수 있었고, 우유제조업체에서 교차오염 방지, 살균온도/시간 관리, 유통온도, 가정에서 온도 관리 등이 매우 중요할 것으로 사료된다.

• 대한응급의학회지
• /
• 제29권5호
• /
• pp.449-454
• /
• 2018

Objective: The SAFARI score was introduced to assess the risk of convulsive seizure during admission for aneurysmal subarachnoid hemorrhage in 2017. This study was conducted to determine whether the SAFARI score derived from the afore-mentioned study could be applied to patients with aneurysmal subarachnoid hemorrhage in Korea. Methods: We conducted a retrospective study of patients who were diagnosed with aneurysmal subarachnoid hemorrhage from March 2013 to October 2017. Patients' age, sex, blood pressure, pulse rate, body temperature, Glasgow-Coma Scale, Hunt-Hess scale, modified Fisher grade, size of ruptured aneurysm, surgery type, transfusion, and SAFARI score were compared between the seizure and non-seizure groups. The area under the receiver operator characteristic curves was calculated to evaluate the predictive ability for seizure during admission. Logistic regression analysis was used to analyze predictive factors for seizure during admission. Results: A total of 220 patients were included. Ninety-seven (44.1%) were male and 123 (55.9%) were female. The mean age of the patients was 65.8 years old (range, 56-75). The area under the curve of the SAFARI score for predicting seizure was 0.813. The SAFARI score was the only significant predictor of seizure during admission, while other factors were not statistically significant upon logistic regression analysis. Conclusion: The SAFARI score could be used for predicting seizure during admission in patients with aneurysmal subarachnoid hemorrhage.

• International journal of advanced smart convergence
• /
• 제12권3호
• /
• pp.89-103
• /
• 2023

Obstructive sleep apnea (OSA) is one of the most prevalent sleep disorders that can lead to serious consequences, including hypertension and/or cardiovascular diseases, if not treated promptly. Continuous positive airway pressure (CPAP) is widely recognized as the most effective treatment for OSA, which needs the proper titration of airway pressure to achieve the most effective treatment results. However, the process of CPAP titration can be time-consuming and cumbersome. There is a growing importance in predicting personalized CPAP pressure before CPAP treatment. The primary objective of this study was to optimize the CPAP titration process for obstructive sleep apnea patients through EEG feature engineering with machine learning techniques. We aimed to identify and utilize the most critical EEG features to forecast key OSA predictive indicators, ultimately facilitating more precise and personalized CPAP treatment strategies. Here, we analyzed 126 OSA patients' PSG datasets before and after the CPAP treatment. We extracted 29 EEG features to predict the features that have high importance on the OSA prediction index which are AHI and SpO2 by applying the Shapley Additive exPlanation (SHAP) method. Through extracted EEG features, we confirmed the six EEG features that had high importance in predicting AHI and SpO2 using XGBoost, Support Vector Machine regression, and Random Forest Regression. By utilizing the predictive capabilities of EEG-derived features for AHI and SpO2, we can better understand and evaluate the condition of patients undergoing CPAP treatment. The ability to predict these key indicators accurately provides more immediate insight into the patient's sleep quality and potential disturbances. This not only ensures the efficiency of the diagnostic process but also provides more tailored and effective treatment approach. Consequently, the integration of EEG analysis into the sleep study protocol has the potential to revolutionize sleep diagnostics, offering a time-saving, and ultimately more effective evaluation for patients with sleep-related disorders.

• 터널과지하공간
• /
• 제33권6호
• /
• pp.594-609
• /
• 2023

TBM 공법은 굴착면 안정성 확보 및 주변환경에 비치는 영향을 최소화하기 때문에 도심지나 하·해저터널 등에서 적용 사례가 증가하는 추세이다. 디스크 커터의 수명을 예측하는 대표적인 모델 중 NTNU모델은 커터수명지수(Cutter Life Index, CLI)를 주요 매개 변수로 활용하지만 복잡한 시험절차와 시험장비의 희귀성으로 측정에 어려움이 있다. 본 연구에서는 다중선형회귀분석과 트리 기반의 머신러닝 기법으로 암석물성을 활용하여 CLI를 예측하였다. 문헌 조사를 통해 암석의 일축압축강도, 압열인장강도, 등 가석영함량과 세르샤 마모지수 등을 포함한 데이터베이스를 구축하였고 파생변수를 계산하여 추가하였다. 다중선형회귀분석은 통계적 유의성과 다중공선성을 고려하여 입력 변수를 선정하였고 머신러닝 예측 모델은 변수 중요도를 기반으로 입력 변수를 선정하였다. 학습용과 검증용 데이터를 8:2로 나누어 모델 간 예측 성능을 비교한 결과 XGBoost가 최적의 모델로 선정되었다. 본 연구에서 도출된 다중선형회귀모델과 XGBoost모델을 선행 연구와 예측 성능을 비교하여 타당성을 확인하였다.

• 응용통계연구
• /
• 제37권3호
• /
• pp.323-338
• /
• 2024

본 연구는 한국 프로배구 리그를 체계적으로 분석하고 대표적인 머신러닝 분류 방법을 활용하여 경기 결과를 예측하고자 한다. 이를 위해 2012/2013 시즌부터 2022/2023 시즌까지의 남자 프로배구와 여자 프로배구 리그 경기 데이터를 수집하였으며, 이 데이터는 경기 세부 내용을 상세하게 포함하고 있다. 데이터는 각 경기를 두 팀으로 분리한 경우와 홈팀을 기준으로 상대팀과의 성과 차이로 데이터를 가공한 경우로 두 가지 다른 데이터 구조를 모델에 적용했다. 이를 통해 남자 프로배구와 여자 프로배구 각각에 대해 총 4개의 예측 모형을 구축했다. 경기 종료 전에는 모형에서 사용하는 세부 변수 값들을 알 수 없기 때문에, 오늘 경기 직전까지의 3~4 경기의 결과를 전처리하여 이를 변수로 사용했다. 본 연구에서는 Decision Tree, Logistic Regression, Bagging, Random Forest, Xgboost, Adaboost, Light GBM 같은 다양한 머신러닝 기법을 분류에 활용하여, Random Forest를 사용한 모델이 가장 우수한 예측 성능을 보였다. 최종 선택한 모형에 대해 변수 중요도 그림과 부분 의존도 그림을 확인한 결과 성별과 데이터 구조에 따라 중요한 변수들이 다른 것으로 나타났지만, 공통적으로 세트 성공 수, 블로킹 득점, 범실 개수가 가장 중요한 변수임을 알 수 있었다. 본 승패 예측 모델은 사후적 예측이 아닌 경기 종료 전 사전 예측이 가능한 모형이라는 점에서 차별성을 가지며, 우리의 분석이 한국 프로배구 팀들에게 전략적 추론이 될 수 있을 것이라 기대한다.

• 한국산업보건학회지
• /
• 제18권3호
• /
• pp.239-247
• /
• 2008

Low back disorders (LBDs) have been the most common musculoskeletal problem in Korean workplaces. It affects many workers, and is associated with high costs to many companies as well as the individual, which can negatively influence even the quality of life of workers. The _evaluation of low back disorder risk associated with manual materials handling tasks can be performed using variety of ergonomic assessment tools such as National Institute for Occupational Safety and Health (NIOSH) Revised Lifting Equation (NLE), the Washington Administrative Code 296-62-0517 (WAC), the Snook Tables etc. But most of these tools provide limited information for choosing the most appropriate assessment method for a particular job and in finding out advantage and disadvantage of the methods, and few have been assessed for their predictive ability. The focus of this study was to _evaluate spinal loads in real time with lifting and pulling heavy cow leathers in variety of postures. Data for estimating mean trunk motions were collected as employees did their work at the job site, using the Lumbar Motion Monitor. Eight employees (2 males, 6 females) were selected in this study, in which the load weight and the vertical start and destination heights of the activity remained constant throughout the task. Variance components (three dimensional spaces) of mean trunk kinematic measures were estimated in a hierarchical design. They were used to compute velocity and acceleration of multiple employees performing the same task and to repetitive movements within a task. Therefore, a results of this study could be used as a quantitative, objective measure to design the workplace so that the risk of occupationally related low back disorder should be minimized.

• 대한화학회지
• /
• 제55권5호
• /
• pp.733-740
• /
• 2011

Multiple regression analysis was used for the calculation of pKa values of 15 substituted benzaldoximes by using various types of descriptors as parameters. These descriptors are based on quantum mechanical treatments. They were derived by employing semi-empirical calculation represented by the PM3 model and an Abinitio method expressed by Hartree-Fock(HF) model performed at the 6-311 G(d, p) level of theory. The parameters tested for their ability to represent the variations observed in the experimental pKa(s) are atomic and structural properties including Muliken charges on the atoms of hydroxyl group and C=N bond, the angle $C_6-C_1-C_7$, and length of O-H bond. Molecular properties are also used like energies of HOMO and LUMO, hardness(${\eta}$), chemical potential(${\mu}$), total energy(TE), dipole of molecule(DM), and electrophilicity index(W). The relation between pKa values and each of these parameters of the studied compounds is investigated. Depending on these relations, two sets of parameters were constructed for comparison between the PM3 and HF methods. The results obtained favor the Abinitio method for such applications although both models proved to have high predictive power and have sufficient reliability to describe the effect of substituents on pKa values of benzaldoxime compounds under consideration which is clear from the values of correlation coefficient $R^2$ obtained and the consistency between the experimental and the calculated values.

• 통합자연과학논문집
• /
• 제10권2호
• /
• pp.85-94
• /
• 2017

Fibroblast growth factor receptor (FGFR) belongs to the family of receptor tyrosine kinase. They play important roles in cell proliferation, differentiation, development, migration, survival, wound healing, haematopoiesis and tumorigenesis. FGFRs are reported to cause several types of cancers in humans which make it an important drug target. In the current study, HQSAR analysis was performed on a series of recently reported 1H-Pyrazolo [3,4-b]pyridine derivatives as FGFR antagonists. The model was developed with Atom (A) and bond (B) connection (C), chirality (Ch), hydrogen (H) and donor/acceptor (DA) parameters and with different set of atom counts to improve the model. A reasonable HQSAR model ($q^2=0.701$, SDEP=0.654, NOC=5, $r^2=0.926$, SEE=0.325, BHL=71) was generated which showed good predictive ability. The contribution map depicted the atom contribution in inhibitory effect. A contribution map for the most active compound (compound 24) indicated that hydrogen and nitrogen atoms in the side chains of ring B as well as hydrogen atoms in the side chain of ring C and the nitrogen atom in the ring D contributed positively to the activity in inhibitory effect whereas, the lowest active compound (compound 04) showed negative contribution to inhibitory effect. Thus results of our study can provide insights in the designing potent and selective FGFR kinase inhibitors.