• 수산해양기술연구
• /
• 제60권1호
• /
• pp.87-99
• /
• 2024

This study deals with the application of an artificial neural network (ANN) model to predict power consumption for utilizing seawater source heat pumps of recirculating aquaculture system. An integrated dynamic simulation model was constructed using the TRNSYS program to obtain input and output data for the ANN model to predict the power consumption of the recirculating aquaculture system with a heat pump system. Data obtained from the TRNSYS program were analyzed using linear regression, and converted into optimal data necessary for the ANN model through normalization. To optimize the ANN-based power consumption prediction model, the hyper parameters of ANN were determined using the Bayesian optimization. ANN simulation results showed that ANN models with optimized hyper parameters exhibited acceptably high predictive accuracy conforming to ASHRAE standards.

• Geomechanics and Engineering
• /
• 제36권4호
• /
• pp.381-390
• /
• 2024

The introduction of bio-based materials has been recommended in the geotechnical engineering field to reduce environmental pollutants such as heavy metals and greenhouse gases. However, bio-treated soil methods face limitations in field application due to short research periods and insufficient verification of engineering performance, especially when compared to conventional materials like cement. Therefore, this study aimed to develop a machine learning model for predicting the unconfined compressive strength, a representative soil property, of biopolymer-based soil treatment (BPST). Four machine learning algorithms were compared to determine a suitable model, including linear regression (LR), support vector regression (SVR), random forest (RF), and neural network (NN). Except for LR, the SVR, RF, and NN algorithms exhibited high predictive performance with an R² value of 0.98 or higher. The permutation feature importance technique was used to identify the main factors affecting the strength enhancement of BPST. The results indicated that the unconfined compressive strength of BPST is affected by mean particle size, followed by biopolymer content and water content. With a reliable prediction model, the proposed model can present guidelines prior to laboratory testing and field application, thereby saving a significant amount of time and money.

• Tuberculosis and Respiratory Diseases
• /
• 제72권3호
• /
• pp.310-317
• /
• 2012

Background: Airway obstruction and the extent of emphysema are reported to be responsible for reduced bone mineral density (BMD). Corresponding to different phenotypes of a pulmonary disease, different severity in extra pulmonary features may exist. We compared BMDs of subjects with or without airway obstruction and/or emphysema and investigated the relationships among BMD, the severity of airway obstruction, and the extent of emphysema. Methods: Using a university hospital database, we reviewed patients over 40 years old who performed spirometry, computed tomography of chest, and measurement of BMD of the lumbar (L) spine. According to the presence or absence of airway obstruction and/or emphysema, four groups were classified. Results: Among a total of 59 subjects, 33 (56%) had osteoporosis. The prevalence of osteoporosis in subjects with no airway obstruction and no emphysema, those with only emphysema, those with only airway obstruction, and those with both airway obstruction and emphysema were 42%, 57%, 64%, and 73%, respectively (p=0.047 by linear-by-linear association). The mean T-scores of BMD of L1 (p=0.032) and L1-4 spines were different among the four groups (p=0.034). Although the T-score of L1 BMD negatively correlated with the extent of emphysema (r=-0.275, p=0.035) and positively with each of body mass index (BMI) (r=0.520, p<0.001), forced expiratory volume in one second ($FEV_1$) (r=0.330, p=0.011), $FEV_1$/forced vital capacity (r=0.409, p=0.001), and forced expiratory flow at 25~75% of FVC ($FEF_{25-75%}$) (r=0.438, p=0.0001), respectively, multiple linear regression analysis indicated that BMI (p<0.001) and $FEF_{25-75%}$ were predictive of BMD (p=0.012). Conclusion: Low BMI and airway obstruction were strongly associated with reduced bone density rather than the extent of emphysema.

• 한국산학기술학회논문지
• /
• 제18권6호
• /
• pp.302-311
• /
• 2017

본 연구는 병원 간호사의 감정 노동과 자원동원성이 이직 의도에 미치는 영향을 파악하기 위한 예측적 상관관계 조사 연구이다. 연구 대상자는 종합병원에서 근무한지 6개월 이상의 간호사 138명이고, 자료 수집은 2016년 4월 10일부터 17일까지 온라인상에서 자가보고식 설문지를 이용하여 진행하였다. 자료 분석은 SPSS statistics 21 프로그램을 이용하여 기술통계, t-test, One-way ANOVA, Scheffe's test와 multiple linear regression으로 분석하였다. 연구 결과, 감정노동은 연령, 결혼 유무, 직위, 임상 경력, 교대근무에 따라, 자원동원성은 성별과 임상 경력에 따라 유의한 차이가 나타났으며, 이직 의도는 연령에 따라 유의한 차이를 보였다. 회귀 분석 결과에서 이직 의도에 영향을 미치는 관련 변수로는 감정노동(${\beta}=.19$, p=.048), 자원동원성(${\beta}=-.18$, p=.038)이 유의한 영향을 미치는 것으로 나타났으며, 이들 요인이 이직 의도를 설명하는 정도는 3.7%였다. 연구결과를 바탕으로 병원 간호사의 이직 의도를 낮추기 위해 간호사의 감정노동과 자원동원성을 감소시키기 위한 구체적인 방안을 마련할 필요성을 제시한다. 이를 바탕으로 병원 조직 관리차원에서 이를 활용하여 전략적인 프로그램을 개발하고 적용한다면 병원간호사의 이직 의도를 감소시킬 수 있을 것으로 기대한다.

• 대한전자공학회논문지SP
• /
• 제42권2호
• /
• pp.131-142
• /
• 2005

현재 개발된 4 kbit/s이하의 저 전송율 음성부호화 시스템은 STC(Sinusoidal Transform Coding)나 MBE (Multi-band Excitation Coding)에 바탕을 두고 있다. 이러한 저 전송율 부호화기들은 대표적인 전이구간 신호인 유성음의 시작점과 끝점에서의 혼합신호(onset signal, offset signal), 비주기적인 신호(non-period signal) 등은 정확히 표현하지 못하기 때문에 자연스런 음질을 만들어 내지 못한다. 본 논문에서는 유성음에는 하모닉 모델, 무성음에서는 스토케스틱 모델, 전이구간에는 하모닉 기반의 비주기적인 펄스의 위치를 추적하는 방식을 사용하여 효과적으로 전이구간을 모델링 하는 방법과 2.4 kbit/s 다중모드 부호화방법을 제안한다. 제안한 방법은 원본신호에서 선형예측 부호화 방법으로 추출된 잔여신호를 신호의 성격에 따라 모델을 달리하는 방법이며, 자각의 신호의 성격에 따라 좋은 성능을 나타내는 모델을 사용하였다. 또한 효율적인 전이구간 모델링 방법의 도입으로 저 전송율에서 CELP(Code Excitation Linear Predictive) 부호화 방식에 의해 시간축에서 합성되는 여기신호와 선형위상을 이용한 하모닉 부호화 방식에 의해 주파수축에서 합성되는 여기신호를 효율적으로 결합이 가능하다는 것이 제안된 2.4 kbit/s 다중모드 부호화기의 장점이다. 제안된 방법의 2.4kbit/s 다중모드 부호화기는 미국 연방 표준부호화기인 2.4 kbit/s MELP(Mixed Excitation Linear Prediction) 부호화기보다 더 좋은 성능을 나타낸다.

• 한국콘텐츠학회논문지
• /
• 제21권11호
• /
• pp.77-86
• /
• 2021

시간적인 차이를 두고 획득한 이질적인 과거 전쟁 결과 데이터를 하나의 모형으로 구축하는 방법으로 베이지안 추론에 의한 전쟁시뮬레이션 모형을 구축하는 방법을 제안하였다. 과거의 전쟁 결과를 분석하여 미래에 있을 수 있는 전쟁을 예측하는 방법으로 선형회귀모형을 적용하는 방법을 고려할 수 있다. 그러나 역사적으로 시대가 서로 달라 전장 환경의 변화가 반영된 이질적인 두 유형의 자료들이라면 모형의 가정사항 위반으로 하나의 선형회귀모형으로 적합하는 것은 적절하지 않다. 이러한 문제를 해결하기 위해 앞선 시대에 있는 자료를 비정보적 사전분포로 가정하여 사후분포를 구하고 이를 다음 시대에 얻은 자료를 분석하기 위한 사전분포로 활용하여 최종 사후분포를 추론하는 베이지안 추론 방법을 제안하였다. 베이지안 추론 방법의 또 다른 장점은 마코프 체인 몬테 카를로 방법으로 샘플링한 결과를 이용하여 불확실성이 반영된 사후분포나 사후예측분포를 추론할 수 있다는 점이다. 이렇게 했을 때 고전적인 선형회귀모형으로 분석하는 것보다 다양한 정보를 활용할 수 있을 뿐만 아니라 향후 추가적으로 획득되는 자료도 모형에 반영하여 모형을 계속 업데이트시킬 수 있다는 장점이 있다.

• Journal of Trauma and Injury
• /
• 제20권2호
• /
• pp.130-137
• /
• 2007

Purpose: Ultrasound is of proven accuracy in abdominal and thoracic trauma and may be useful for diagnosing extremity injury in situations where radiography is not available, such as disasters and military and space applications. However, the diagnosis of fractures is suggested by history and physical examination and is typically confirmed with radiography. As a alternative to radiography, we prospectively evaluated the utility of extremity ultrasound performed by trained residents of emergency medicine (EM) one patient with wrist and ankle extremity injuries. Methods: Initially, residents of EM performed physical examinations for fractures. The emergency ultrasound (EM US) was performed by trained residents, who used a portable ultrasound device with a 10- to 5-MHz linear transducer, on suspected patients before radiography examination. The results of emergency ultrasound and radiography and the final diagnosis were recorded, and correlation among them were determined by using Kappa s test Results: Thirty-nine patients were enrolled in our study. The average age was $36.6\;{\pm}\;19.3$ years. There were radius Fx. (n=21), radius-ulna Fx. (n=1), ulna Fx. (n=1), and contusion (n=2) injuries among the wrist injury and lat.-med. malleolar Fx. (n=13), lat. malleolar Fx. (n=6), and med. malleolar Fx. (n=3) injuries among the ankle injury. Comparing EM US with radiography, we found the sensitivity, specificity, positive predictive value, and negative predictive value of EM US for Fx. diagnosis to be 100%, 66.7%, 97.3%, 100% and those of radiography to be 97.2%, 100%, 100%, and 75%, respectively. Kappa s test for a correlation between the Fx. diagnosis of EM US and the final diagnosis of Fx was performed, and Kappa's value was 0.787 (P = 0.004).Conclusion: EM US for Fx. can be performed quickly and accurately by EM residents with excellent accuracy in remote locations such as disaster areas and in military and aerospace applications. EM US was as useful as radiography in our study and had a high correlation to the final diagnosis of Fx. Therefore, ultrasound should performed on patients with extremity injury to determine whether extremity evaluation should be added to the FAST (focused abdominal sonography trauma) examination.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권4호
• /
• pp.1861-1864
• /
• 2016

Although early diagnosis of thyroid nodule type is very important, the diagnostic accuracy of standard tests is a challenging issue. We here aimed to find an optimal combination of factors to improve diagnostic accuracy for distinguishing malignant from benign thyroid nodules before surgery. In a prospective study from 2008 to 2012, 345 patients referred for thyroidectomy were enrolled. The sample size was split into a training set and testing set as a ratio of 7:3. The former was used for estimation and variable selection and obtaining a linear combination of factors. We utilized smoothly clipped absolute deviation (SCAD) logistic regression to achieve the sparse optimal combination of factors. To evaluate the performance of the estimated model in the testing set, a receiver operating characteristic (ROC) curve was utilized. The mean age of the examined patients (66 male and 279 female) was $40.9{\pm}13.4years$ (range 15- 90 years). Some 54.8% of the patients (24.3% male and 75.7% female) had benign and 45.2% (14% male and 86% female) malignant thyroid nodules. In addition to maximum diameters of nodules and lobes, their volumes were considered as related factors for malignancy prediction (a total of 16 factors). However, the SCAD method estimated the coefficients of 8 factors to be zero and eliminated them from the model. Hence a sparse model which combined the effects of 8 factors to distinguish malignant from benign thyroid nodules was generated. An optimal cut off point of the ROC curve for our estimated model was obtained (p=0.44) and the area under the curve (AUC) was equal to 77% (95% CI: 68%-85%). Sensitivity, specificity, positive predictive value and negative predictive values for this model were 70%, 72%, 71% and 76%, respectively. An increase of 10 percent and a greater accuracy rate in early diagnosis of thyroid nodule type by statistical methods (SCAD and ANN methods) compared with the results of FNA testing revealed that the statistical modeling methods are helpful in disease diagnosis. In addition, the factor ranking offered by these methods is valuable in the clinical context.

• 한국재난정보학회 논문집
• /
• 제16권1호
• /
• pp.133-143
• /
• 2020

연구목적: 본 연구는 국내에서 발생한 지진관측기록의 P파 성분으로부터 PGV를 예측하기 위한 예측식을 도출하고 지진현장경보(Onsite EEW)신뢰성을 검증함을 목적으로 한다. 연구방법: 국내에서 발생한 627개 지진 이벤트에 대한 관측기록으로부터 P파 외의 잡음을 제거하여 기반암에서의 PGV와의 예측식을 도출하고, 이를 이용한 지진현장경보 시뮬레이션을 통해 PGV의 예측치와 관측치 비교로부터 신뢰성을 검증한다. 연구결과: P파 잡음을 제거한 지진 관측기록으로부터 Filter Picker를 사용하여 P파를 추출하고, PGV와의 회귀분석을 통해 지진현장경보를 위한 예측식을 도출했다. 현장경보 시뮬레이션 결과 경보대상 구간인 MMI IV 이상 구간에서 MMI±1 오차범위 내 80%의 성공률을 얻었다. 결론: 본 연구를 통해 국내 지진기록을 이용한 지진현장경보의 설계 가능성과 성능을 확인하였다. 유효성을 높이기 위해, 해외 지진다발지역의 중규모지진의 관측기록을 분석기록으로 추가하고, 오탐지 제어 및 지표에서의 지진파 증폭에 효과 구현이 필요하다.

• 대한예방한의학회지
• /
• 제16권3호
• /
• pp.37-51
• /
• 2012

Objectives : The objective of this study is to determine the present state of patients with breast cancer use of Korean medicine(KM) and predictive factors for the use. Through this, the present study is intended to present reasonable treatment approaches for patients with breast cancer as well as communicating correct information on KM to healthcare providers and presenting objective alternatives for patients with breast cancer management based on the subjects' experience in health benefits obtained from their use of KM. Methods : To collect data for the present study, questionnaire surveys were conducted on outpatients who visited four hospitals located in Seoul, Korea during around three weeks from May 31, 2012. Although the total number of the questionnaire sheet distributed in the form of directly asking questionnaire questions was 300, 12 incomplete questionnaire sheets were excluded. Therefore, the number of questionnaire sheets actually used in analyses was 288 and thus the collect rate was 96%. Results : Major results of this study are as follows. First, the number of subjects who responded to the questionnaire was 288 in total. Forty-six percent of the patients reported KM usage and the most commonly used ginseng and qigong/exercise. KM use was found to be associated with age, experiencing side effects of cancer treamnent. Factors that affect the use of KM were analyzed by Linear Logistic Regression and the results showed that age, experiencing side effects of cancer treatment, effectiveness of cancer treatment, and satisfaction of the treatment were factors that were related with relatively more frequent use of KM. Conclusions : Comparing the previous studies, it could be seen that patients with breast cancer were highly interested in and used KM in which conventional medicine and KM are used simultaneously. Knowledge on the integrative use of KM and conventional therapies is necessary for cancer physicians and traditional Korean medical doctors to help patients make informed choices. KM use may play a role in the positive benefits associated with process of breast center treatment. Healthcare providers should communicate correct information on the KM use that has been scientifically verified and talk with each other openly. The fact that the significant correlation between predictive factors for the use of KM was identified trough the present study is quite meaningful.