• 대한세포병리학회지
• /
• 제18권2호
• /
• pp.100-111
• /
• 2007

Fine needle aspiration (FNA) cytology of breast disease is recognized to be highly accurate and cost effective, especially when this is used in combination with clinical examination and imaging as part of a triple approach. A probabilistic/categorical approach is used for the classification of breast FNA specimens. Criteria are defined from the perspective of the likelihood of making a definitive diagnosis of cancer on excision. This approach is an accurate way of classifying breast FNA specimens, and this can be reliably applied regardless of the level of experience of the pathologist for interpreting the case. When a definitive diagnosis of malignancy is made, the next step is to determining the specific histologic types of the malignancy according to their cytological features. In order to make an accurate diagnosis of carcinoma and for correct typing a tumor, an adequate, correctly sampled aspirate without any other artifacts is required.

• 한국지능시스템학회논문지
• /
• 제23권5호
• /
• pp.412-417
• /
• 2013

태양광 발전 시스템의 해석적 모델링은 시스템의 동특성을 예측하거나 고장검출 및 진단 등과 같은 고급 공학 기술에 중요하게 적용할 수 있어 최근 많은 각광을 받고 있다. 본 논문은 대용량 학습 데이터를 갖는 태양광 발전 시스템에 대한 확률론적 모델링을 제시한다. 우선 태양광 일사량과 온도 입력 변수에 대한 태양광 시스템의 출력 전력과의 입출력 함수관계를 정의한다. 이 함수관계를 바탕으로 세 확률변수(일사량, 온도, 전력)에 대하여 조건부 확률 식으로 표현한다. 조건부 확률 분포 추정은 대용량 데이터 시스템에 적합한, 전체 표본 데이터 수 대비 관련 변수의 경우의 수에 대한 비율로 나타내었다. 추정한 확률분포를 통해 평균값 이론을 적용하여 시스템의 출력을 추정하게 된다. 본 논문에서 제안한 모델링 기법은 두 태양광 발전 단지의 사례 연구를 통해 성능을 검증하였다.

• 대한예방의학회:학술대회논문집
• /
• 대한예방의학회 1994년도 교수 연수회(역학)
• /
• pp.150-153
• /
• 1994

Receiver operating characteristic (ROC) curves have been frequently used to compare probability models applied to medical problems. Though the curves are a measure of the discriminatory power of a model. they do not reflect the model's accuracy. A supplementary accuracy curve is derived which will be coincident with the ROC curve if the model is reliable. will be above the ROC curve if the model's probabilities are too high or below if they are too low. A clinical example of this new graphical presentation is given.

• Journal of Electrical Engineering and Technology
• /
• 제6권6호
• /
• pp.742-749
• /
• 2011

Power transmission lines are one of the most important components of electric power system. Failures in the operation of power transmission lines can result in serious power system problems. Hence, fault diagnosis (transient or permanent) in power transmission lines is very important to ensure the reliable operation of the power system. A hidden Markov model (HMM), a powerful pattern recognizer, classifies events in a probabilistic manner based on fault signal waveform and characteristics. This paper presents application of HMM to classify faults in overhead power transmission lines. The algorithm uses voltage samples of one-fourth cycle from the inception of the fault. The simulation performed in EMTPWorks and MATLAB environments validates the fast response of the classifier, which provides fast and accurate protection scheme for power transmission lines.

• 한국철도학회논문집
• /
• 제20권2호
• /
• pp.223-233
• /
• 2017

철도에 사용되고 있는 분기기는 중요한 시스템 중 하나이다. 분기기 시스템의 건전성은 철도운용 안정성에 매우 중요하다. 분기기 시스템을 진단하기 위하여, LVDT와 accelerometer를 분기기에 설치하였다. LVDT는 분기기에서 변위가 발생하는 부분에 설치하여, 분기기의 이동과 차량의 주행에서 발생되는 변위를 측정하였다. Accelerometers는 충격과 진동이 발생하는 부분에 설치하여 충격량과 진동을 측정하였다. 측정된 데이터를 이용하여 변수화를 위한 데이터를 추출하였으며, 이 변수들은 진단에 사용하였다. 진단 알고리즘은 확률분포와 인공신경망을 사용하였다. 변수화된 값이 확률분포를 이용하여 판단할 수 있으면 확률분포를 사용하였으며, 형태를 보고 판단할 필요가 있으면 인공신경망을 활용하였다. 본 논문에서는 정상적인 상태에서 데이터를 측정하여, 정상상태의 조건을 위한 학습을 수행하였다.

• 전기학회논문지
• /
• 제62권6호
• /
• pp.870-875
• /
• 2013

After introducing diagnosis equipment power failure prevention activities for distribution system have become more active. To do facility diagnosis and maintenance work more efficiently we need to evaluate reliability for the system and should determine the priority line with appropriate criteria. Thus, to calculate risk factor for the power distribution line that are composed of many component facilities its historical failure events for the last 5 years are collected and analysed. The failure statics show that more than 60% of various failures are related to environment factors randomly and about 20% of the failures are refer to the aging. As a strategic evaluation system reflecting these environmental influence is needed, a system on the basis of the probabilistic approach related statical variables in terms of failure rate and failure probability of electrical components is proposed. The figures for the evaluation are derived from the field failure DB. With adopting Bayesian rule we can calculate easily about conditional probability query. The proposed evaluation system is demonstrated with model system and the calculated indices representing the properties of the model line are discussed.

• Industrial Engineering and Management Systems
• /
• 제9권1호
• /
• pp.41-53
• /
• 2010

An existing risk diagnosing methodology (RDM) diagnoses corporate risk for product-innovation projects. However, it cannot evaluate and compare the risk levels of multiple alternatives in the product development stage. This paper proposes a modified risk diagnosis method to fill the gap of risk evaluation in selections of innovative product alternatives and the application of the method will be also illustrated by a case problem on alternative selections in electrical dimmer designs. With RDM as the foundation, a modified RDM (MRDM) is proposed to deal with the problem of selecting innovative project alternatives during the early stages of product development. The Bayesian network; a probabilistic graphical model, is adopted to support the risk pre-assessment stage in the MRDM. The MRDM is proposed by incorporating the risk pre-assessment stage into the foundation. By evaluating the engineering design risks in two electrical dimmer switches, an application of the MRDM in product innovation development is successfully exemplified. This paper strengthens the existing methodology for RDM in innovative product development projects to accommodate innovative alternatives. It is advantageous for companies to identify and measure the risks associated in product development so as to plan for appropriate risk mitigation strategies.

• International journal of advanced smart convergence
• /
• 제13권2호
• /
• pp.61-68
• /
• 2024

In the quest for advancing diabetes diagnosis, this study introduces a novel two-step machine learning approach that synergizes the probabilistic predictions of Logistic Regression with the classification prowess of Random Forest. Diabetes, a pervasive chronic disease impacting millions globally, necessitates precise and early detection to mitigate long-term complications. Traditional diagnostic methods, while effective, often entail invasive testing and may not fully leverage the patterns hidden in patient data. Addressing this gap, our research harnesses the predictive capability of Logistic Regression to estimate the likelihood of diabetes presence, followed by employing Random Forest to classify individuals into diabetic, pre-diabetic or nondiabetic categories based on the computed probabilities. This methodology not only capitalizes on the strengths of both algorithms-Logistic Regression's proficiency in estimating nuanced probabilities and Random Forest's robustness in classification-but also introduces a refined mechanism to enhance diagnostic accuracy. Through the application of this model to a comprehensive diabetes dataset, we demonstrate a marked improvement in diagnostic precision, as evidenced by superior performance metrics when compared to other machine learning approaches. Our findings underscore the potential of integrating diverse machine learning models to improve clinical decision-making processes, offering a promising avenue for the early and accurate diagnosis of diabetes and potentially other complex diseases.

• Nuclear Engineering and Technology
• /
• 제55권5호
• /
• pp.1901-1910
• /
• 2023

The safety-class (1E) digital control system (DCS) of nuclear power plant characterized structural multiple redundancies, therefore, it is important to quantitatively evaluate the reliability of DCS in different degree of backup loss. In this paper, a reliability evaluation model based on T-S fuzzy fault tree (FT) is proposed for 1E DCS of nuclear power plant, in which the connection relationship between components is described by T-S fuzzy gates. Specifically, an output rejection control system is chosen as an example, based on the T-S fuzzy FT model, the key indicators such as probabilistic importance are calculated, and for a further discussion, the T-S fuzzy FT model is transformed into Bayesian Network(BN) equivalently, and the fault diagnosis based on probabilistic analysis is accomplished. Combined with the analysis of actual objects, the effectiveness of proposed method is proved.

• Smart Structures and Systems
• /
• 제32권1호
• /
• pp.61-81
• /
• 2023

This study presents an ensemble learning based Bayesian model updating approach for structural damage diagnosis. In the developed framework, the structure is initially decomposed into a set of substructures. The autoregressive moving average (ARMAX) model is established first for structural damage localization based structural motion equation. The wavelet packet decomposition is utilized to extract the damage-sensitive node energy in different frequency bands for constructing structural surrogate models. Four methods, including Kriging predictor (KRG), radial basis function neural network (RBFNN), support vector regression (SVR), and multivariate adaptive regression splines (MARS), are selected as candidate structural surrogate models. These models are then resampled by bootstrapping and combined to obtain an ensemble model by probabilistic ensemble. Meanwhile, the maximum entropy principal is adopted to search for new design points for sample space updating, yielding a more robust ensemble model. Through the iterations, a framework of surrogate ensemble learning based model updating with high model construction efficiency and accuracy is proposed. The specificities of the method are discussed and investigated in a case study.