• Journal of Biomedical Engineering Research
• /
• v.41 no.1
• /
• pp.42-47
• /
• 2020

A personal authentication system based on biosignals has received increasing attention due to its relatively high security as compared to traditional authentication systems based on a key and password. Electrocardiography (ECG) measured from the chest or wrist is one of the widely used biosignals to develop a personal authentication system. In this study, we investigated the feasibility of using similar ECG measured behind the ears to develop a personal authentication system. To this end, similar ECGs were measured from thirty subjects using a pair of three electrodes attached behind each of the ears during resting state during which the standard Lead-I ECG was also simultaneously measured from both wrists as baseline ECG. The three ECG components, Q, R, and S, were extracted for each subject as classification features, and authentication accuracy was estimated using support vector machine (SVM) based on a 5×5-fold cross-validation. The mean authentication accuracies of Lead I-ECG and similar ECG were 90.41 ± 8.26% and 81.15 ± 7.54%, respectively. Considering a chance level of 3.33% (=1/30), the mean authentication performance of similar ECG could demonstrate the feasibility of using similar ECG measured behind the ears on the development of a personal authentication system.

• Journal of Biomedical Engineering Research
• /
• v.32 no.2
• /
• pp.134-143
• /
• 2011

The selection of meaningful clinical tests and its reference values from a high-dimensional clinical data with imbalanced class distribution, one class is represented by a large number of examples while the other is represented by only a few, is an important issue for differential diagnosis between similar diseases, but difficult. For this purpose, this study introduces methods based on the concepts of both discernibility matrix and function in rough set theory (RST) with two discretization approaches, equal width and frequency discretization. Here these discretization approaches are used to define the reference values for clinical tests, and the discernibility matrix and function are used to extract a subset of significant clinical tests from the translated nominal attribute values. To show its applicability in the differential diagnosis problem, we have applied it to extract the significant clinical tests and its reference values between normal (N = 351) and abnormal group (N = 101) with either cholecystitis or cholelithiasis disease. In addition, we investigated not only the selected significant clinical tests and the variations of its reference values, but also the average predictive accuracies on four evaluation criteria, i.e., accuracy, sensitivity, specificity, and geometric mean, during l0-fold cross validation. From the experimental results, we confirmed that two discretization approaches based rough set approximation methods with relative frequency give better results than those with absolute frequency, in the evaluation criteria (i.e., average geometric mean). Thus it shows that the prediction model using relative frequency can be used effectively in classification and prediction problems of the clinical data with imbalanced class distribution.

• Journal of Biomedical Engineering Research
• /
• v.40 no.6
• /
• pp.242-249
• /
• 2019

Forearm electromyography (EMG) generated by wrist movements has been widely used to develop an electrical prosthetic hand, but EMG generated by finger movements has been rarely used even though 20% of amputees lose fingers. The goal of this study is to improve the classification performance of different finger movements using a deep learning algorithm, and thereby contributing to the development of a high-performance finger-based prosthetic hand. Ten participants took part in this study, and they performed seven different finger movements forty times each (thumb, index, middle, ring, little, fist and rest) during which EMG was measured from the back of the right hand using four bipolar electrodes. We extracted mean absolute value (MAV), root mean square (RMS), and mean (MEAN) from the measured EMGs for each trial as features, and a 5x5-fold cross-validation was performed to estimate the classification performance of seven different finger movements. A long short-term memory (LSTM) model was used as a classifier, and linear discriminant analysis (LDA) that is a widely used classifier in previous studies was also used for comparison. The best performance of the LSTM model (sensitivity: 91.46 ± 6.72%; specificity: 91.27 ± 4.18%; accuracy: 91.26 ± 4.09%) significantly outperformed that of LDA (sensitivity: 84.55 ± 9.61%; specificity: 84.02 ± 6.00%; accuracy: 84.00 ± 5.87%). Our result demonstrates the feasibility of a deep learning algorithm (LSTM) to improve the performance of classifying different finger movements using EMG.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10a
• /
• pp.53-58
• /
• 2006

MicroRNA (miRNA)는 약 22 nt의 작은 RNA 조각으로 이루어져 있으며 stem-loop 구조의 precursor 형태에서 최종적으로 만들어 진다. miRNA는 mRNA의 3‘UTR에 상보적으로 결합하여 유전자의 발현을 억제하거나 mRNA의 분해를 촉진한다. miRNA를 동정하기 위한 실험적인 방법은 조직 특이적인 발현, 적은 발현양 때문에 방법상 한계를 가지고 있다. 이러한 한계는 컴퓨터를 이용한 방법으로 어느 정도 해결될 수 있다. 하지만 miRNA의 서열상의 낮은 보존성은 homology를 기반으로 한 예측을 어렵게 한다. 또한 기계학습 방법인 support vector machine (SVM) 이나 naive bayes가 적용되었지만, 생물학적인 의미를 해석할 수 있는 generative model을 제시해 주지 못했다. 본 연구에서는 우수한 miRNA 예측을 보일 뿐만 아니라 학습된 모델로부터 생물학적인 지식을 얻을 수 있는 Bayesian network을 적용한다. 이를 위해서는 생물학적으로 의미 있는 특질들의 선택이 중요하다. 여기서는 position weighted matrix (PWM)과 Markov chain probability (MCP), Loop 크기, Bulge 수, spectrum, free energy profile 등을 특질로서 선택한 후 Information gain의 특질 선택법을 통해 예측에 기여도가 높은 특질 25개 와 27개를 최종적으로 선택하였다. 이로부터 Bayesian network을 학습한 후 miRNA의 예측 성능을 10 fold cross-validation으로 확인하였다. 그 결과 pre-/mature miRNA 각 각에 대한 예측 accuracy가 99.99% 100.00%를 보여, SVM이나 naive bayes 방법보다 높은 결과를 보였으며, 학습된 Bayesian network으로부터 이전 연구 결과와 일치하는 pre-miRNA 상의 의존관계를 분석할 수 있었다.

• Experimental and Molecular Medicine
• /
• v.50 no.11
• /
• pp.7.1-7.12
• /
• 2018

Recent findings from The Cancer Genome Atlas project have provided a comprehensive map of genomic alterations that occur in hepatocellular carcinoma (HCC), including unexpected mutations in apolipoprotein B (APOB). We aimed to determine the clinical significance of this non-oncogenetic mutation in HCC. An Apob gene signature was derived from genes that differed between control mice and mice treated with siRNA specific for Apob (1.5-fold difference; P < 0.005). Human gene expression data were collected from four independent HCC cohorts (n = 941). A prediction model was constructed using Bayesian compound covariate prediction, and the robustness of the APOB gene signature was validated in HCC cohorts. The correlation of the APOB signature with previously validated gene signatures was performed, and network analysis was conducted using ingenuity pathway analysis. APOB inactivation was associated with poor prognosis when the APOB gene signature was applied in all human HCC cohorts. Poor prognosis with APOB inactivation was consistently observed through cross-validation with previously reported gene signatures (NCIP A, HS, high-recurrence SNUR, and high RS subtypes). Knowledge-based gene network analysis using genes that differed between low-APOB and high-APOB groups in all four cohorts revealed that low-APOB activity was associated with upregulation of oncogenic and metastatic regulators, such as HGF, MTIF, ERBB2, FOXM1, and CD44, and inhibition of tumor suppressors, such as TP53 and PTEN. In conclusion, APOB inactivation is associated with poor outcome in patients with HCC, and APOB may play a role in regulating multiple genes involved in HCC development.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.35 no.11
• /
• pp.155-162
• /
• 2019

The construction industry is the highest safety accident causing industry as 28.55% portion of all industries' accidents in Korea. In particular, falling is the highest accidents type composed of 60.16% among the construction field accidents. Therefore, we analyzed the factors of major disaster affecting the fall accident and then derived feature importances by considering various variables. We used data collected from Korea Occupational Safety & Health Agency (KOSHA) for learning and predicting in the proposed model. We have an effort to predict the degree of occupational fall accidents by using the machine learning model, i.e., Adaboost, short for Adaptive Boosting. Adaboost is a machine learning meta-algorithm which can be used in conjunction with many other types of learning algorithms to improve performance. Decision trees were combined with AdaBoost in this model to predict and classify the degree of occupational fall accidents. HyOperpt was also used to optimize hyperparameters and to combine k-fold cross validation by hierarchy. We extracted and analyzed feature importances and affecting fall disaster by permutation technique. In this study, we verified the degree of fall accidents with predictive accuracy. The machine learning model was also confirmed to be applicable to the safety accident analysis in construction site. In the future, if the safety accident data is accumulated automatically in the network system using IoT(Internet of things) technology in real time in the construction site, it will be possible to analyze the factors and types of accidents according to the site conditions from the real time data.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.153-166
• /
• 2021

A main goal of pharmacogenomics studies is to predict individual's drug responsiveness based on high dimensional genetic variables. Due to a large number of variables, feature selection is required in order to reduce the number of variables. The selected features are used to construct a predictive model using machine learning algorithms. In the present study, we applied several hybrid feature selection methods such as combinations of logistic regression, ReliefF, TurF, random forest, and LASSO to a next generation sequencing data set of 400 epilepsy patients. We then applied the selected features to machine learning methods including random forest, gradient boosting, and support vector machine as well as a stacking ensemble method. Our results showed that the stacking model with a hybrid feature selection of random forest and ReliefF performs better than with other combinations of approaches. Based on a 5-fold cross validation partition, the mean test accuracy value of the best model was 0.727 and the mean test AUC value of the best model was 0.761. It also appeared that the stacking models outperform than single machine learning predictive models when using the same selected features.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.4
• /
• pp.12-22
• /
• 2021

This article suggests the machine learning model, i.e., classifier, for predicting the production quality of free-machining 303-series stainless steel(STS303) small rolling wire rods according to the operating condition of the manufacturing process. For the development of the classifier, manufacturing data for 37 operating variables were collected from the manufacturing execution system(MES) of Company S, and the 12 types of derived variables were generated based on literature review and interviews with field experts. This research was performed with data preprocessing, exploratory data analysis, feature selection, machine learning modeling, and the evaluation of alternative models. In the preprocessing stage, missing values and outliers are removed, and oversampling using SMOTE(Synthetic oversampling technique) to resolve data imbalance. Features are selected by variable importance of LASSO(Least absolute shrinkage and selection operator) regression, extreme gradient boosting(XGBoost), and random forest models. Finally, logistic regression, support vector machine(SVM), random forest, and XGBoost are developed as a classifier to predict the adequate or defective products with new operating conditions. The optimal hyper-parameters for each model are investigated by the grid search and random search methods based on k-fold cross-validation. As a result of the experiment, XGBoost showed relatively high predictive performance compared to other models with an accuracy of 0.9929, specificity of 0.9372, F₁-score of 0.9963, and logarithmic loss of 0.0209. The classifier developed in this study is expected to improve productivity by enabling effective management of the manufacturing process for the STS303 small rolling wire rods.

• Structural Engineering and Mechanics
• /
• v.83 no.3
• /
• pp.293-304
• /
• 2022

In this work, a novel reliability approach for combined high and low cycle fatigue (CCF) estimation is developed by combining active learning strategy with least squares support vector machines (LS-SVM) (named as ALS-SVM) surrogate model to address the multi-resources uncertainties, including working loads, material properties and model itself. Initially, a new active learner function combining LS-SVM approach with Monte Carlo simulation (MCS) is presented to improve computational efficiency with fewer calls to the performance function. To consider the uncertainty of surrogate model at candidate sample points, the learning function employs k-fold cross validation method and introduces the predicted variance to sequentially select sampling. Following that, low cycle fatigue (LCF) loads and high cycle fatigue (HCF) loads are firstly estimated based on the training samples extracted from finite element (FE) simulations, and their simulated responses together with the sample points of model parameters in Coffin-Manson formula are selected as the MC samples to establish ALS-SVM model. In this analysis, the MC samples are substituted to predict the CCF reliability of turbine blades by using the built ALS-SVM model. Through the comparison of the two approaches, it is indicated that the reliability model by linear cumulative damage rule provides a non-conservative result compared with that by the proposed one. In addition, the results demonstrate that ALS-SVM is an effective analysis method holding high computational efficiency with small training samples to gain accurate fatigue reliability.

• Geomechanics and Engineering
• /
• v.30 no.3
• /
• pp.259-272
• /
• 2022

Rockburst is a dynamic, multivariate, and non-linear phenomenon that occurs in underground mining and civil engineering structures. Predicting rockburst is challenging since conventional models are not standardized. Hence, machine learning techniques would improve the prediction accuracies. This study describes decision based uncertainty models to predict rockburst in underground engineering structures using gradient boosting algorithms (GBM). The model input variables were uniaxial compressive strength (UCS), uniaxial tensile strength (UTS), maximum tangential stress (MTS), excavation depth (D), stress ratio (SR), and brittleness coefficient (BC). Several models were trained using different combinations of the input variables and a 3-fold cross-validation resampling procedure. The hyperparameters comprising learning rate, number of boosting iterations, tree depth, and number of minimum observations were tuned to attain the optimum models. The performance of the models was tested using classification accuracy, Cohen's kappa coefficient (k), sensitivity and specificity. The best-performing model showed a classification accuracy, k, sensitivity and specificity values of 98%, 93%, 1.00 and 0.957 respectively by optimizing model ROC metrics. The most and least influential input variables were MTS and BC, respectively. The partial dependence plots revealed the relationship between the changes in the input variables and model predictions. The findings reveal that GBM can be used to anticipate rockburst and guide decisions about support requirements before mining development.