• Tuberculosis and Respiratory Diseases
• /
• v.86 no.3
• /
• pp.203-215
• /
• 2023

Background: Surgical resection is the standard treatment for early-stage lung cancer. Since postoperative lung function is related to mortality, predicted postoperative lung function is used to determine the treatment modality. The aim of this study was to evaluate the predictive performance of linear regression and machine learning models. Methods: We extracted data from the Clinical Data Warehouse and developed three sets: set I, the linear regression model; set II, machine learning models omitting the missing data: and set III, machine learning models imputing the missing data. Six machine learning models, the least absolute shrinkage and selection operator (LASSO), Ridge regression, ElasticNet, Random Forest, eXtreme gradient boosting (XGBoost), and the light gradient boosting machine (LightGBM) were implemented. The forced expiratory volume in 1 second measured 6 months after surgery was defined as the outcome. Five-fold cross-validation was performed for hyperparameter tuning of the machine learning models. The dataset was split into training and test datasets at a 70:30 ratio. Implementation was done after dataset splitting in set III. Predictive performance was evaluated by R² and mean squared error (MSE) in the three sets. Results: A total of 1,487 patients were included in sets I and III and 896 patients were included in set II. In set I, the R² value was 0.27 and in set II, LightGBM was the best model with the highest R² value of 0.5 and the lowest MSE of 154.95. In set III, LightGBM was the best model with the highest R² value of 0.56 and the lowest MSE of 174.07. Conclusion: The LightGBM model showed the best performance in predicting postoperative lung function.

• Radiation Oncology Journal
• /
• v.37 no.2
• /
• pp.117-126
• /
• 2019

Purpose: The purpose of this study was to characterize and evaluate the clinical significance of volume changes of soft tissue sarcomas during radiation therapy (RT), prior to definitive surgical resection. Materials and Methods: Patients with extremity or pelvis soft tissue sarcomas treated at our institution from 2013 to 2016 with RT prior to resection were identified retrospectively. Tumor volumes were measured using cone-beam computed tomography obtained daily during RT. Linear regression evaluated the linearity of volume changes. Kruskal-Wallis tests, Mann-Whitney U tests, and linear regression evaluated predictors of volume change. Logistic and Cox regression evaluated volume change as a predictor of resection margin status, histologic treatment response, and tumor recurrence. Results: Thirty-three patients were evaluated. Twenty-nine tumors were high grade. Prior to RT, median tumor volume was 189 mL (range, 7.2 to 4,885 mL). Sixteen tumors demonstrated significant linear volume changes during RT. Of these, 5 tumors increased and 11 decreased in volume. Myxoid liposarcoma (n = 5, 15%) predicted decreasing tumor volume (p = 0.0002). Sequential chemoradiation (n = 4, 12%) predicted increasing tumor volume (p = 0.008) and corresponded to longer times from diagnosis to RT (p = 0.01). Resection margins were positive in three cases. Five patients experienced local recurrence, and 7 experienced distant recurrence, at median 8.9 and 6.9 months post-resection, respectively. Volume changes did not predict resection margin status, local recurrence, or distant recurrence. Conclusion: Volume changes of pelvis and extremity soft tissue sarcomas followed linear trends during RT. Volume changes reflected histologic subtype and treatment characteristics but did not predict margin status or recurrence after resection.

• Journal of radiological science and technology
• /
• v.43 no.6
• /
• pp.503-509
• /
• 2020

This study aims to develop a patient-specific radiation exposure dose prediction model based on anthropometric data that can be easily measurable during CT examination, and to be used as basic data for DRL setting and radiation dose management system in the future. In addition, among the machine learning algorithms, the most suitable model for predicting exposure doses is presented. The data used in this study were chest CT scan data, and a data set was constructed based on the data including the patient's anthropometric data. In the pre-processing and sample selection of the data, out of the total number of samples of 250 samples, only chest CT scans were performed without using a contrast agent, and 110 samples including height and weight variables were extracted. Of the 110 samples extracted, 66% was used as a training set, and the remaining 44% were used as a test set for verification. The exposure dose was predicted through random forest, linear regression analysis, and SVM algorithm using Orange version 3.26.0, an open software as a machine learning algorithm. Results Algorithm model prediction accuracy was R^2 0.840 for random forest, R^2 0.969 for linear regression analysis, and R^2 0.189 for SVM. As a result of verifying the prediction rate of the algorithm model, the random forest is the highest with R^2 0.986 of the random forest, R^2 0.973 of the linear regression analysis, and R^2 of 0.204 of the SVM, indicating that the model has the best predictive power.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.5
• /
• pp.2414-2419
• /
• 2012

This paper presents an optimal operation control method for energy saving in the water reuse pumping system. A predictive horizon switching strategy is proposed to implement an optimal operation control and a linear programming (LP) algorithm is used to solve optimal problems in each time step. Energy costs are calculated for electricity on both TOU in the light, heavy, and maximum load time period and peak charges. The optimal operation in water reuse pumping systems is determined to reduce the TOU and peak costs. The simulation results show a power energy saving for water reuse pumping systems and power stability improvement.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.27 no.4
• /
• pp.491-501
• /
• 1990

Recently, four speech coding techniques, namely, SBC-APCM(sub-band coding adaptive PCM), RPE-LPC(regualr pulse excitation linear predictive codec), MPE-LTP(multi-pulse excited long-term prediction) and CELP (code-excited linear prediction) are proposed for digital mobile radio applications. However, a performance comparison of these coders in the Rayleigh fading environment has not been made yet. In this paper, the performances of the four spech coders in the random bit error and burst error environment are investigated. For the channel coding of SBC-APCM, RPE-LPC and MPE-LTP, the sensitivity of output bit stream is measured and a bit selective forward error correction is provided acording to the measured bit sensitivity. And for an attempt to improve the performance of CELP, an optimum quantizer is applied for transmitting scalar quantities in CELP. However, an improvement over the conventional approach is found to be negligible. For the channel coding of CELP, Reed-Solomon code, Golay code, convolutional code of rate 1/2 shows the best performance. Finally, from the simulation results, it is concluded that CELP is the best candidate for digital mobile radio and is followed by MPE-LTP, SBC-APCM and RPE-LPC.

• Bulletin of the Korean Chemical Society
• /
• v.29 no.4
• /
• pp.833-841
• /
• 2008

Principal component-genetic algorithm-multiparameter linear regression (PC-GA-MLR) and principal component-genetic algorithm-artificial neural network (PC-GA-ANN) models were applied for prediction of melting point for 323 drug-like compounds. A large number of theoretical descriptors were calculated for each compound. The first 234 principal components (PC’s) were found to explain more than 99.9% of variances in the original data matrix. From the pool of these PC’s, the genetic algorithm was employed for selection of the best set of extracted PC’s for PC-MLR and PC-ANN models. The models were generated using fifteen PC’s as variables. For evaluation of the predictive power of the models, melting points of 64 compounds in the prediction set were calculated. Root-mean square errors (RMSE) for PC-GA-MLR and PC-GA-ANN models are 48.18 and $12.77{^{\circ}C}$, respectively. Comparison of the results obtained by the models reveals superiority of the PC-GA-ANN relative to the PC-GA-MLR and the recently proposed models (RMSE = $40.7{^{\circ}C}$). The improvements are due to the fact that the melting point of the compounds demonstrates non-linear correlations with the principal components.

• Bulletin of the Korean Chemical Society
• /
• v.28 no.9
• /
• pp.1472-1476
• /
• 2007

Artificial neural networks (ANNs) are successfully developed for the modeling and prediction of normalized polarity parameter (ETN) of 216 various solvents with diverse chemical structures using a quantitative-structure property relationship. ANN with architecture 5-9-1 is generated using five molecular descriptors appearing in the multi-parameter linear regression (MLR) model. The most positive charge of a hydrogen atom (q+), total charge in molecule (qt), molecular volume of solvent (Vm), dipole moment (μ) and polarizability term (πI) are input descriptors and its output is ETN. It is found that properly selected and trained neural network with 192 solvents could fairly represent the dependence of normalized polarity parameter on molecular descriptors. For evaluation of the predictive power of the generated ANN, an optimized network is applied for prediction of the ETN values of 24 solvents in the prediction set, which are not used in the optimization procedure. Correlation coefficient (R) and root mean square error (RMSE) of 0.903 and 0.0887 for prediction set by MLR model should be compared with the values of 0.985 and 0.0375 by ANN model. These improvements are due to the fact that the ETN of solvents shows non-linear correlations with the molecular descriptors.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.3
• /
• pp.160-166
• /
• 1989

We propose adaptive algorithms computing the coefficients of spatial domain predictors. The method uses the LMS approach to compute the coefficients of the predictors realized by using the TDL(tapped-delay-line) and the ESC (escalator) structures. The predictors to be presented differ from the conventional ones in the sense that the relevant weights are updated such that the sum of the mean squared values of the forward and the backward prediction errors is minimized. Using the coefficients of such spatial domain predictors yields improved linear predictive spatial spectrums. The algorithms are applied to the problems of estimating incident angles of multiple narrow-band signals received by a linear array of sensors. Simulation results demonstrating the performances of the proposed methods are presented.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.585-588
• /
• 2019

Predictions of Energy Consumption for Industries gain an important place in energy management and control system, as there are dynamic and seasonal changes in the demand and supply of energy. This paper presents and discusses the predictive models for energy consumption of the steel industry. Data used includes lagging and leading current reactive power, lagging and leading current power factor, carbon dioxide (tCO2) emission and load type. In the test set, four statistical models are trained and evaluated: (a) Linear regression (LR), (b) Support Vector Machine with radial kernel (SVM RBF), (c) Gradient Boosting Machine (GBM), (d) random forest (RF). Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) are used to measure the prediction efficiency of regression designs. When using all the predictors, the best model RF can provide RMSE value 7.33 in the test set.

• Nuclear Engineering and Technology
• /
• v.53 no.12
• /
• pp.4042-4051
• /
• 2021

In this work, we applied a multilevel modeling technique to estimate the diametral creep in the pressure tubes of Korean Canada Deuterium Uranium (CANDU) units. Data accumulated from in-service inspections were used to develop the model. To confirm the strength of the multilevel models, a 2-level multilevel model considering the relationship between channels for a CANDU unit was compared with existing linear models. The multilevel model exhibited a very robust prediction accuracy compared to the linear models with different data pooling methods. A 3-level multilevel model, which considered individual bundles, channels, and units, was also implemented. The influence of the channel installation direction was incorporated into the three-stage multilevel model. For channels that were previously measured, the developed 3-level multilevel model exhibited a very good predictive power, and the prediction interval was very narrow. However, for channels that had never been measured before, the prediction interval widened considerably. This model can be sufficiently improved by the accumulation of more data and can be applied to other CANDU units.