• Nuclear Engineering and Technology
• /
• 제54권4호
• /
• pp.1439-1448
• /
• 2022

Background: We investigated the feasibility of in vitro radiosensitivity prediction with gene expression using deep learning. Methods: A microarray gene expression of the National Cancer Institute-60 (NCI-60) panel was acquired from the Gene Expression Omnibus. The clonogenic surviving fractions at an absorbed dose of 2 Gy (SF2) from previous publications were used to measure in vitro radiosensitivity. The radiosensitivity prediction model was based on the convolutional neural network. The 6-fold cross-validation (CV) was applied to train and validate the model. Then, the leave-one-out cross-validation (LOOCV) was applied by using the large-errored samples as a validation set, to determine whether the error was from the high bias of the folded CV. The criteria for correct prediction were defined as an absolute error<0.01 or a relative error<10%. Results: Of the 174 triplicated samples of NCI-60, 171 samples were correctly predicted with the folded CV. Through an additional LOOCV, one more sample was correctly predicted, representing a prediction accuracy of 98.85% (172 out of 174 samples). The average relative error and absolute errors of 172 correctly predicted samples were 1.351±1.875% and 0.00596±0.00638, respectively. Conclusion: We demonstrated the feasibility of a deep learning-based in vitro radiosensitivity prediction using gene expression.

• 국제학술발표논문집
• /
• The 6th International Conference on Construction Engineering and Project Management
• /
• pp.597-598
• /
• 2015

In order to improve the reliability of cost estimation results using CBR, there has been a continuous issue on similarity measurement to accurately compute the distance among attributes and cases to retrieve the most similar singular or plural cases. However, these existing similarity measures have limitations in taking the covariance among attributes into consideration and reflecting the effects of covariance in computation of distances among attributes. To deal with this challenging issue, this research examines the weighted Mahalanobis distance based similarity measure applied to CBR cost estimation and carries out the comparative study on the existing distance measurement methods of CBR. To validate the suggest CBR cost model, leave-one-out cross validation (LOOCV) using two different sets of simulation data are carried out. Consequently, this research is expected to provide an analysis of covariance effects in similarity measurement and a basis for further research on the fundamentals of case retrieval.

• 응용통계연구
• /
• 제24권5호
• /
• pp.847-859
• /
• 2011

cDNA microarray-based comparative genomic hybridization(CGH) data includes low-intensity spots and thus a statistical strategy is needed to detect subtle differences between different cancer classes. In this study, genes displaying a high frequency of alteration in one of the different classes were selected among the pre-selected genes that show relatively large variations between genes compared to total variations. Utilizing copy-number changes of the selected genes, this study suggests a statistical approach to predict patients' classes with increased performance by pre-classifying patients with similar genetic alteration scores. Two-stage logistic regression model(TLRM) was suggested to pre-classify homogeneous patients and predict patients' classes for cancer prediction; a decision tree(DT) was combined with logistic regression on the set of informative genes. TLRM was constructed in cDNA microarray-based CGH data from the Cancer Metastasis Research Center(CMRC) at Yonsei University; it predicted the patients' clinical diagnoses with perfect matches (except for one patient among the high-risk and low-risk classified patients where the performance of predictions is critical due to the high sensitivity and specificity requirements for clinical treatments. Accuracy validated by leave-one-out cross-validation(LOOCV) was 83.3% while other classification methods of CART and DT performed as comparisons showed worse performances than TLRM.

• 항공우주산업기술동향
• /
• 제6권1호
• /
• pp.124-134
• /
• 2008

The rising demand for the high efficiency and high covertness in UAV motivates the miniature design of the high performing mission sensors, or payloads. One of the promising payload sensors, EO/IR sensor has evolved satisfying its demands and became the main stand-alone mission sensor for 200kg-range UAV. One aspect in development of EO/IR sensor concerns lack of specification criterions to represent its performance. Even though the high demand and competition among each manufacturer caused EO/IR features subject to rapid change collateral to new technology, the datasheets maintained the conventional outdated formats which leave some of the major components in ambiguity. Making comparisons or predicting actual performance with such datasheets is hardly worthwhile; yet, they could be important reference guide for the potential customers what to expect for the upcoming EO/IR. According to UAS Roadmap 2007-2032 published by DoD, one of the main potential customers as well as a main investor of EO/IR technology, EO/IR is expected to play key roll in solving urgent problems, such as see and avoid system. This paper will examine the recent representative EO/IR specialized in UAS missions through datasheets to find out current trend and eventually extrapolate the possible future trend.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.682-685
• /
• 2004

This paper investigated a method for classifying emotional states by using pulse wave signal. It focused on finding effective features for emotional state classification. The emptional states considered here consisted of interest and neutral. Classification experiments utilized 65 and 60 samples of interest and neutral states respectively. We have investigated 19 features derived from pulse wave signals by using both time domain and frequency domain analysis methods with 2 classifiers of minimum distance (normalized Euclidean distanece) and ${\kappa}$-Nearest Neighbour. The Leave-one-out cross validation was used as an evaluation mehtod. Based on experimental results, the most efficient features were a combination of 4 features consisting of (i) the mean of the first differences of the smoothed pulse rate time series signal, (ii) the mean of absolute values of the second differences of thel normalized interbeat intervals, (iii) the root mean square successive difference, and (iv) the power in high frequency range in normalized unit, which provided 80.8% average accuracy with ${\kappa}$-Nearest Neighbour classifier.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2003년도 봄 학술발표논문집 Vol.30 No.1 (B)
• /
• pp.322-324
• /
• 2003

생체인식 기술은 현재까지 많은 발전을 거듭하고 있으며 국내에서도 연구는 물론 표준화작업 및 데이터 베이스 구축이 활발히 진행되고 있다. 생체인식은 신체의 여러 부분을 이용하는 방법과 습관에서 비롯된 특징을 이용하는 방법이 있는데, 본 연구에서는 이 중에서 개인의 필기습관 정보를 이용하여 인식하였다. 본 연구에서는 필기습관에 주목하여 서명하는 사람의 습관이 잘 드러나는 펜의 기울임과 눌림, 펜의 방위각도 둥의 성분이 표현되어지는 동적인 생채정보를 감지하고 특성을 추출할 수 있는 타블렛과 펜을 사용하여 서명정보를 추출한다. 이렇게 생성된 서명정보의 특징을 추출하기 위하여 패턴인식분야에 널리 활용하고 있는 주성분요소분석(PCA, Principal Component Analysis), 독립성분요소분석(ICA, Independent Component Analysis)기법에 적용하였다. 생성된 두 특징벡터 사이의 거리를 Euclidean Distance를 이용하여 구하고 Nearest Neighbor를 비교하여 인식률을 알아보고 교차인식(Cross Validation) 기법 중 하나인 Leave-One-Out 방법을 이용한 분류성능 측정을 통하여 데이터의 신뢰수준을 알아보았다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2006년도 Proceedings of ISRS 2006 PORSEC Volume I
• /
• pp.406-408
• /
• 2006

Various geological thematic maps such as grain size or ground water level maps have been generated by interpolating sparsely sampled ground survey data. When there are sampled data at a limited number of locations, to use secondary information which is correlated to primary variable can help us to estimate the attribute values of the primary variable at unsampled locations. This paper applies two multivariate geostatistical algorithms to integrate remote sensing imagery with sparsely sampled ground survey data for spatial estimation of grain size: simple kriging with local means and kriging with an external drift. High-resolution IKONOS imagery which is well correlated with the grain size is used as secondary information. The algorithms are evaluated from a case study with grain size observations measured at 53 locations in the Baramarae beach of Anmyeondo, Korea. Cross validation based on a one-leave-out approach is used to compare the estimation performance of the two multivariate geostatistical algorithms with that of traditional ordinary kriging.

• Communications for Statistical Applications and Methods
• /
• 제24권1호
• /
• pp.67-80
• /
• 2017

Over the last few decades, ensemble forecasts based on global climate models have become an important part of climate forecast due to the ability to reduce uncertainty in prediction. Moreover in ensemble forecast, assessing the prediction uncertainty is as important as estimating the optimal weights, and this is achieved through a probabilistic forecast which is based on the predictive distribution of future climate. The Bayesian model averaging has received much attention as a tool of probabilistic forecasting due to its simplicity and superior prediction. In this paper, we propose a new Bayesian model averaging method for probabilistic ensemble forecasting. The proposed method combines a deterministic ensemble forecast based on a multivariate regression approach with Bayesian model averaging. We demonstrate that the proposed method is better in prediction than the standard Bayesian model averaging approach by analyzing monthly average precipitations and temperatures for ten cities in Korea.

• 한국멀티미디어학회논문지
• /
• 제9권6호
• /
• pp.762-772
• /
• 2006

Instance based learning algorithm is the best known lazy learner and has been successfully used in many areas such as pattern analysis, medical analysis, bioinformatics and internet applications. However, its feature weighting scheme is too naive that many other extensions are proposed. Our version of IB3 named as eXtended IBL (XIBL) improves feature weighting scheme by backward stepwise regression and its distance function by VDM family that avoids overestimating discrete valued attributes. Also, XIBL adopts leave-one-out as its noise filtering scheme. Experiments with common artificial domains show that XIBL is better than the original IBL in terms of accuracy and noise tolerance. XIBL is applied to two important applications - intrusion detection and spam mail filtering and the results are promising.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.518-518
• /
• 2000

We have researched quantitative structure activity relationships between molecular structure of medicines and physiological activity. Since they are non-linear, neural networks are useful tool to research them. There are many ranks for the non-linearity; therefore, the neuron function must be selected carefully. As the results of some trial calculations, Ire find the sigmoid-linear functions pair. We call the neural network constructed of the pair as ANN. The inter- or extrapolation abilities of the ANN are excellent; therefere, ANN is a superior predictor for the relationships. We evaluated the anticarcinogenic medicines, Carboquinone derivatives, by the developed ANN and leave-one-out method.