• Journal of Gastric Cancer
• /
• v.22 no.2
• /
• pp.120-134
• /
• 2022

Purpose: This study aimed to identify prognostic factors for patients with distant lymph node-involved gastric cancer (GC) using a machine learning algorithm, a method that offers considerable advantages and new prospects for high-dimensional biomedical data exploration. Materials and Methods: This study employed 79 features of clinical pathology, laboratory tests, and therapeutic details from 289 GC patients whose distant lymphadenopathy was presented as the first episode of recurrence or metastasis. Outcomes were measured as any-cause death events and survival months after distant lymph node metastasis. A prediction model was built based on possible outcome predictors using a random survival forest algorithm and confirmed by 5×5 nested cross-validation. The effects of single variables were interpreted using partial dependence plots. A contour plot was used to visually represent survival prediction based on 2 predictive features. Results: The median survival time of patients with GC with distant nodal metastasis was 9.2 months. The optimal model incorporated the prealbumin level and the prothrombin time (PT), and yielded a prediction error of 0.353. The inclusion of other variables resulted in poorer model performance. Patients with higher serum prealbumin levels or shorter PTs had a significantly better prognosis. The predicted one-year survival rate was stratified and illustrated as a contour plot based on the combined effect the prealbumin level and the PT. Conclusions: Machine learning is useful for identifying the important determinants of cancer survival using high-dimensional datasets. The prealbumin level and the PT on distant lymph node metastasis are the 2 most crucial factors in predicting the subsequent survival time of advanced GC.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.13
• /
• pp.5383-5387
• /
• 2014

Purpose: To investigate whether pretreatment serum carbohydrate antigen 19-9 (CA 19-9) levels are associated with pathological responses to preoperative chemoradiotherapy (CRT) in patients with rectal cancer. Materials and Methods: In total, 260 patients with locally advanced rectal cancer (cT3-4NanyM0) who underwent preoperative CRT and radical surgery were analyzed retrospectively. CRT consisted of 50.4 Gy pelvic radiotherapy and concurrent chemotherapy. Radical surgery was performed at a median of 7 weeks after CRT completion. Pathological CRT response criteria included downstaging (ypStage 0-I) and ypT0-1. A discrimination threshold of CA 19-9 level was determined using a receiver operating characteristics analysis. Results: The median CA 19-9 level was 8.0 (1.0-648.0) U/mL. Downstaging occurred in 94 (36.2%) patients and ypT0-1 in 50 (19.2%). The calculated optimal threshold CA 19-9 level was 10.2 U/mL for downstaging and 9.0 U/mL for ypT0-1. On multivariate analysis, CA 19-9 (${\leq}9.0U/mL$) was significantly associated with downstaging (odds ratio, 2.089; 95% confidence interval, 1.189-3.669; P=0.010) or ypT0-1 (OR, 2.207; 95%CI, 1.079-4.512; P=0.030), independent of clinical stage or carcinoembryonic antigen. Conclusions: This study firstly showed a significant association of pretreatment serum CA 19-9 levels with pathological CRT responses of rectal cancer. The CA 19-9 level is suggested to be valuable in predicting CRT responses of rectal cancer cases before treatment.

• International Journal of Highway Engineering
• /
• v.14 no.2
• /
• pp.145-155
• /
• 2012

At present, Seoul City is putting the bus exclusive lane system into practice according to mass transit revitalization policy. Starting with the installation of roadside bus exclusive lane in the past, at present, even the road sections for central- lane bus exclusive lane system are on the increase. The purpose of this research is to analyze the factors giving impacts on bus accident on central bus exclusive lane and roadside bus exclusive lane. In case of the central bus exclusive lane, the 6 variables, such as the number of bus routes, number of access & entrance to central lanes patterns, whether the stop line of central lanes retreats or not, separated distance between the stop line of central lanes and crosswalks, traffic volume, and number of bus routes stopping at bus stops on reversible lanes, were found to have a significant influence on bus accidents. In case of roadside bus exclusive lane sections, the four variables such as the number of right-turn bus routes, whether to be chronic illegal parking & stopping, time for the walk signal, and forms of land use, etc. were found to have a significant influence on bus accident.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.1
• /
• pp.11-21
• /
• 2008

In conventional video coding, the complexity of encoder is much higher than that of decoder. However, as more needs arises for extremely simple encoder in environments having constrained energy such as sensor network, much investigation has been carried out for eliminating motion prediction/compensation claiming most complexity and energy in encoder. The Wyner-Ziv coding, one of the representative schemes for the problem, reconstructs video at decoder by correcting noise on side information using channel coding technique such as turbo code. Since the encoder generates only parity bits without performing any type of processes extracting correlation information between frames, it has an extremely simple structure. However, turbo decoding errors occur in noisy side information. When there are high-motion or occlusion between frames, more turbo decoding errors appear in reconstructed frame and look like Salt & Pepper noise. This severely deteriorates subjective video quality even though such noise rarely occurs. In this paper, we propose a computationally extremely light encoder based on symbol-level Wyner-Ziv coding technique and a new corresponding decoder which, based on a decision whether a pixel has error or not, applies median filter selectively in order to minimize loss of texture detail from filtering. The proposed method claims extremely low encoder complexity and shows improvements both in subjective quality and PSNR. Our experiments have verified average PSNR gain of up to 0.8dB.

• Journal of Korean Society of Transportation
• /
• v.21 no.4
• /
• pp.133-144
• /
• 2003

The current procedure of estimating accident reduction benefit shows fixed accident rates for each level of roads without considering the various characteristics of roadway geometries, and traffics. In this study, in order to solve the problems mentioned in the above, models were developed considering the characteristics of roadway alignments and traffic characteristics. The developed models can be used to estimate the accident rates on new or improved roads, In this study, only urban highways were included as a beginning stage. First of all. factors influencing accident rates were selected. Those factors such as traffic volumes. number of signalized intersections, the number of connecting roads, number of pedestrian traffic signals, existence of median barrier, and the number of road lane are also selected based upon the obtainability at the planning stage of roads. The relationship between the selected factors and accident rates shows strong correlation statistically. In this study, roads were classified into 4 groups based on number of lanes, level of roads and the existence of median barriers. The regression analysis had been performed for each group with actual data associated with traffic, roads. and accidents. The developed regression models were verified with another data set. In this study, in order to develop the proposed models, only data on a limited area were used. In order to represent whole area of the country with the developed models. the models should be re-analyzed with vast data.

• Korean Journal of Agricultural Science
• /
• v.39 no.3
• /
• pp.413-420
• /
• 2012

Spectroscopic measurement method based on visible and near-infrared wavelengths was prominent technology for rapid and non-destructive evaluation of internal quality of fruits. Reflectance measurement was performed to evaluate firmness, soluble solid content, and acid content of truss tomatoes by hyperspectral reflectance imaging system. The Vis/NIR reflectance spectra was acquired from truss tomatoes sorted by 6 ripening stages. The multivariable analysis based on partial least square (PLS) was used to develop regression models with several preporcessing methods, such as smoothing, normalization, multiplicative scatter correction (MSC), and standard normal variate (SNV). The best model was selected in terms of coefficient of determination of calibration ($R_c^2$) and full cross validation ($R_{cv}^2$), and root mean standard error of calibration (RMSEC) and full cross validation (RMSECV). The results of selected models were 0.8976 ($R_p^2$), 6.0207 kgf (RMSEP) with gaussian filter of smoothing, 0.8379 ($R_p^2$), $0.2674^{\circ}Bx$ (RMSEP) with the mean of normalization, and 0.7779 ($R_p^2$), 0.1033% (RMSEP) with median filter of smoothing for firmness, soluble solid content (SSC), and acid content, respectively. Results show that Vis / NIR hyperspectral reflectance imaging technique has good potential for the measurement of internal quality of truss tomato.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.20 no.1
• /
• pp.119-128
• /
• 2017

G-induced Loss of Consciousness(G-LOC) can be predicted by measuring Electromyogram(EMG) signals. Existing studies have mainly focused on specific body parts and lacked of consideration with quantitative EMG indices. The purpose of this study is to analyze the indices of EMG signals by human body parts for monitoring G-LOC condition. The data of seven EMG features such as Root Mean Square(RMS), Integrated Absolute Value(IAV), and Mean Absolute Value(MAV) for reflecting muscle contraction and Slope Sign Changes(SSC), Waveform Length (WL), Zero Crossing(ZC), and Median Frequency(MF) for representing muscle contraction and fatigue was retrieved from high G-training on a human centrifuge simulator. A total of 19 trainees out of 47 trainees of the Korean Air Force fell into G-LOC condition during the training in attaching EMG sensor to three body parts(neck, abdomen, calf). IAV, MAV, WL, and ZC under condition after G-LOC were decreased by 17 %, 17 %, 18 %, and 4 % comparing to those under condition before G-LOC respectively. Also, RMS, IAV, MAV, and WL in neck part under condition after G-LOC were higher than those under condition before G-LOC; while, those in abdomen and calf part lower. This study suggest that measurement of IAV and WL by attaching EMG sensor to calf part may be optimal for predicting G-LOC.

• Journal of Korea Water Resources Association
• /
• v.30 no.6
• /
• pp.597-610
• /
• 1997

One-dimensional numerical models using finite difference methods for unsteady sediment transport on alluvial river channel are developed. The Preissmann implicit scheme and the Lax-Wendroff two-step explicit scheme with the Method of Characteristics for water motion and a forward time centered space explicit scheme for sediment motion are developed to simulate the sediment transport rate and the variation of channel bed level. The program correctness of each model is successfully verified using volume conservation tests. The sensitivity studies show that higher peak stage level, steeper channel slope and longer flooding duration produce more channel bed erosion. and median grain size, $D_{50}=0.4mm$ give maximum volume loss in this study. Finally, the numerical models are found to produce reasonable results from the various sensitivity tests which reveal that the numerical models have properly responded to the changes of each model parameter.

• Journal of Biomedical Engineering Research
• /
• v.28 no.4
• /
• pp.556-568
• /
• 2007

Unsatisfied results of return of spontaneous circulation (ROSC) estimates were caused by the fact that the predictability of the predictors was insufficient. This unmet estimate of the predictors may be affected by transitional events due to behaviors which occur during cardiopulmonary resuscitation (CPR). We thus hypothesized that the discrepancy of ROSC estimates found in statistical characteristics due to transitional CPR events, may affect the performance of the predictors, and that the performance of the classifier dichotomizing between ROSC and No-ROSC might be different during CPR. In a canine model (n=18) of prolonged ventricular fibrillation (VF), standard CPR was provided with administration of two doses of epinephrine 0 min or 3 min later of the onset of CPR. For the analysis of the likelihood of a successful defibrillation during CPR, Support Vector Classification was adopted to evaluate statistical peculiarity combining time and frequency based predictors: median frequency, frequency band-limited power spectrum, mean segment amplitude, and zero crossing rates. The worst predictable period showed below about 1 min after the onset of CPR, and the best predictable period could be observed from about 1.5 min later of the administering epinephrine through 2.0-2.2 min. As hypothesized, the discrepancy of statistical characteristics of the predictors was reflected in the differences of the classification performance during CPR. These results represent a major improvement in defibrillation prediction can be achieved by a specific timing of the analysis, as a change in CPR transition.

• Journal of Korean Neurosurgical Society
• /
• v.58 no.5
• /
• pp.448-453
• /
• 2015

Objective : Recently, the survival of patients with hepatocellular carcinoma (HCC) has been prolonged with improvements in various diagnostic tools and medical treatment modalities. Consequently, spine metastases from HCC are being diagnosed more frequently. The accurate prediction of prognosis plays a critical role in determining a patient's treatment plan, including surgery for patients with spinal metastases of HCC. We investigated the clinical features, surgical outcomes, and prognostic factors of HCC presenting with spine metastases, in patients who underwent surgery. Methods : A retrospective review was conducted on 33 HCC patients who underwent 36 operations (three patients underwent surgical treatment twice) from February 2006 to December 2013. The median age of the patients was 56 years old (range, 28 to 71; male : female=30 : 3). Results : Overall survival was not correlated with age, sex, level of metastases, preoperative Child-Pugh classification, preoperative ambulatory function, preoperative radiotherapy, type of operation, administration of Sorafenib, or the Tokuhashi scoring system. Only the Tomita scoring system was shown to be an independent prognostic factor for overall survival. Comparing the Child-Pugh classification and ambulatory ability, there were no statistically differences between patients pre- and post-operatively. Conclusion : The Tomita scoring system represents a practicable and highly predictive prognostic tool. Even though surgical intervention may not restore ambulatory function, it should be considered to prevent deterioration of the patient's overall condition. Additionally, aggressive management may be needed if there is any ambulatory ability remaining.