• Physical Therapy Korea
• /
• v.27 no.2
• /
• pp.140-148
• /
• 2020

Background: Scapular winging (SW) could be caused by tightness or weakness of the periscapular muscles. Although data mining techniques are useful in classifying or predicting risk of musculoskeletal disorder, predictive models for risk of musculoskeletal disorder using the results of clinical test or quantitative data are scarce. Objects: This study aimed to (1) investigate the difference between young women with and without SW, (2) establish a predictive model for presence of SW, and (3) determine the cutoff value of each variable for predicting the risk of SW using the decision tree method. Methods: Fifty young female subjects participated in this study. To classify the presence of SW as the outcome variable, scapular protractor strength, elbow flexor strength, shoulder internal rotation, and whether the scapula is in the dominant or nondominant side were determined. Results: The classification tree selected scapular protractor strength, shoulder internal rotation range of motion, and whether the scapula is in the dominant or nondominant side as predictor variables. The classification tree model correctly classified 78.79% (p = 0.02) of the training data set. The accuracy obtained by the classification tree on the test data set was 82.35% (p = 0.04). Conclusion: The classification tree showed acceptable accuracy (82.35%) and high specificity (95.65%) but low sensitivity (54.55%). Based on the predictive model in this study, we suggested that 20% of body weight in scapular protractor strength is a meaningful cutoff value for presence of SW.

• Journal of Intelligence and Information Systems
• /
• v.9 no.1
• /
• pp.71-89
• /
• 2003

Although many studies demonstrate that one technique outperforms the others for a given data set, there is often no way to tell a priori which of these techniques will be most effective in the classification problems. Alternatively, it has been suggested that a better approach to classification problem might be to integrate several different forecasting techniques. This study proposes the linearly combining methodology of different classification techniques. The methodology is developed to find the optimal combining weight and compute the weighted-average of different techniques' outputs. The proposed methodology is represented as the form of mixed integer programming. The objective function of proposed combining methodology is to minimize total misclassification cost which is the weighted-sum of two types of misclassification. To simplify the problem solving process, cutoff value is fixed and threshold function is removed. The form of mixed integer programming is solved with the branch and bound methods. The result showed that proposed methodology classified more accurately than any of techniques individually did. It is confirmed that Proposed methodology Predicts significantly better than individual techniques and the other combining methods.

• Journal of Korean Neurosurgical Society
• /
• v.61 no.1
• /
• pp.66-74
• /
• 2018

Objective : The aim of this study was to identify the susceptibility genes responsible for lumbar spondylosis (LS) in Korean patients. Methods : Data from 1427 subjects were made available for radiographic grading and genome wide association studies (GWAS) analysis. Lateral lumbar spine radiographs were obtained and the various degrees of degenerative change were semi-quantitatively scored. A pilot GWAS was performed using the AffymetrixGenome-Wide Human single-nucleotide polymorphisms (SNPs), 500K array. A total of 352228 SNPs were analyzed and the association between the SNPs and case-control status was analyzed by stepwise logistic regression analyses. Results : The top 100 SNPs with a cutoff p-value of less than $3.7{\times}10^{-4}$ were selected for joint space narrowing, while a cutoff p-value of $6.0{\times}10^{-4}$ was applied to osteophytes and the Kellgren-Lawrence (K-L) osteoarthritis grade. The SNPs with the strongest effect on disc space narrowing, osteophytes, and K-L grade were serine incorporator 1 (rs155467, odds ratio [OR]=17.58, $p=1.6{\times}10^{-4}$), stromal interaction molecule 2 (STIM1, rs210781, OR=5.53, $p=5{\times}10^{-4}$), and transient receptor potential cation channel, subfamily C (rs11224760, OR=3.99, $p=4.8{\times}10^{-4}$), respectively. Leucine-rich repeat-containing G protein-coupled receptor 4 was significantly associated with both disc space narrowing and osteophytes (rs1979400, OR=2.01, $p=1.1{\times}10^{-4}$ for disc space narrowing, OR=1.79, $p=3{\times}10^{-4}$ for osteophytes), while zinc finger and BTB domain containing 7C was significantly and negatively associated with both osteophytes and a K-L grade >2 (rs12457004,OR=0.25, $p=5.8{\times}10^{-4}$ and OR=0.27, $p=5.3{\times}10^{-4}$, respectively). Conclusion : We identified SNPs that potentially contribute to the pathogenesis of LS. This is the first report of a GWAS in an Asian population.

• Journal of Microbiology and Biotechnology
• /
• v.17 no.1
• /
• pp.146-153
• /
• 2007

A mathematical competition model between normal flora and an invading pathogen was devised to allow analysis of bacterial infections in a host. The normal flora includes the various microorganisms that live on or within the host and act as a primary human immune system. Despite the important role of the normal flora, no mathematical study has been undertaken on models of the interaction between it and invading pathogens against a background of antibiotic treatment. To quantify key elements of bacterial behavior in a host, pairs of nonlinear differential equations were used to describe three categories of human health conditions, namely, healthy, latent infection, and active infection. In addition, a cutoff value was proposed to represent the minimum population level required for survival. The recovery of normal flora after antibiotic treatment was also included in the simulation because of its relation to human health recovery. The significance of each simulation parameter for the bacterial growth model was investigated. The devised simulation showed that bacterial proliferation rate, carrying capacity, initial population levels, and competition intensity have a significant effect on bacterial behavior. Consequently, a model was established to describe competition between normal flora and an infiltrating pathogen. Unlike other population models, the recovery process described by the devised model can describe the human health recovery mechanism.

• Clinical and Experimental Pediatrics
• /
• v.59 no.12
• /
• pp.477-482
• /
• 2016

Purpose: We conducted a study to determine which factors may be useful as predictive markers in identifying Kawasaki disease (KD) patients with a high risk of resistance to intravenous immunoglobulin (IVIG) and developing coronary artery lesions (CAL). Methods: We enrolled 287 patients in acute phase of KD at a single center. The demographic, clinical and laboratory data were collected retrospectively. Results: There were 34 patients in the IVIG resistant group. The IVIG resistant group had significantly higher serum N-terminal-pro-brain natriuretic protein (NT-proBNP) levels (P<0.01) and polymorphonuclear neutrophil (PMN) percentage (P<0.01) in comparison to the IVIG responders. The results yielded sensitivity (78.8%, 60.6%), specificity (58.2%, 90%) and cutoff value (628.6 pg/mL, 80.3%) of NT-proBNP and PMN respectively, in predicting IVIG resistance. Despite IVIG administration, 13 of the 287 patients developed CAL. The patients in the CAL group had higher NT-proBNP levels (P<0.01) and higher PMN percentage (P<0.01). In these patients, the results yielded sensitivity (73.3%, 56.7%), specificity (67.9%, 88.9%) and cutoff value (853.4 pg/mL, 80.3%) of NT-proBNP and PMN respectively, for predicting CAL. The area under the curve (AUC) for predicting resistance to IVIG was NT-proBNP 0.712, PMN 0.802. The AUC for predicting CAL was NT-proBNP 0.739, and PMN 0.773. Conclusion: Serum NT-proBNP levels and PMN percentage were significantly elevated in patients with KD with IVIG resistance and CAL. Thus, they may be useful predicting markers for IVIG resistance and development of CAL in KD patients.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.18
• /
• pp.8625-8628
• /
• 2016

Background: Central lymph node metastasis(CLNM) is common in papillary thyroid microcarcinoma (PTMC). The aim of this study was to define the pathohistologic risk grading based on surgical outcomes. Materials and Methods: Statistical analysis was performed to figure out the optimal cut-off values of size in preoperative ultrasound images for defining the risk of CLNM in papillary thyroid microcarcinoma. Receiver operating characteristic curves (ROC) studies were carried out to determine the cutoff value(s) for the predictor(s). All the patients were divided into two groups according to the above size and the clinic-pathological and immunohistochemical parameters were compared to determine the significance of findings. Results: The optimal cut-off value of tumor size to predict the risk of CLNM in papillary thyroid microcarcinoma was 0.575 cm (area under the curve 0.721) according to the ROC curves. Significant differences were observed on the multifocality, extrathyroidal extension and central lymph node metastasis between two groups which were divided according to the tumor size by the cutoff values. Patients in two groups showed different positive rate and intensity of Ki67. Conclusions: The size of PTMC in ultrasound images are helpful to predict the aggressiveness of the tumors, it could be an easy predictor for PTMC prognosis and assist us to choose treatment.

• Journal of Korean Neurosurgical Society
• /
• v.59 no.5
• /
• pp.478-484
• /
• 2016

Objective : To assess the clinical and radiological factors as predictors for successful outcomes in lumbar disc herniation (LDH) treatment. Methods : Two groups of patients with single level LDH (L4-5) requiring treatment were retrospectively studied. The surgery group (SG) included 34 patients, and 30 patients who initially refused the surgery were included in the nerve blocks group (NG). A visual analogue scale (VAS) for leg and back pain and motor deficit were initially evaluated before procedures, and repeated at 1, 6, and 12 months. Radiological factors including the disc herniation length, disc herniation area, canal length-occupying ratio, and canal area-occupying ratio were measured and compared. Predicting factors of successful outcomes were determined with multivariate logistic regression analysis after the optimal cut off values were established with a receiver operating characteristic curve. Results : There was no significant demographic difference between two groups. A multivariate logistic regression analysis with radiological and clinical (12 months follow-up) data revealed that the high disc herniation length with cutoff value 6.31 mm [odds ratio (OR) 2.35; confidence interval (CI) 1.21-3.98] was a predictor of successful outcomes of leg pain relief in the SG. The low disc herniation length with cutoff value 6.23 mm (OR 0.05; CI 0.003-0.89) and high baseline VAS leg (OR 12.63; CI 1.64-97.45) were identified as predictors of successful outcomes of leg pain relief in the NG. Conclusion : The patients with the disc herniation length larger than 6.31 mm showed successful outcomes with surgery whereas the patients with the disc herniation length less than 6.23 mm showed successful outcomes with nerve block. These results could be considered as a radiological criteria in choosing optimal treatment options for LDH.

• Radiation Oncology Journal
• /
• v.31 no.3
• /
• pp.111-117
• /
• 2013

Purpose: To determine whether the maximum standardized uptake value (SUV) of [$^{18}F$] fluorodeoxyglucose uptake by positron emission tomography (FDG PET) ratio of lymph node to primary tumor (mSUVR) could be a prognostic factor for node positive non-small cell lung cancer (NSCLC) patients treated with definitive radiotherapy (RT). Materials and Methods: A total of 68 NSCLC T1-4, N1-3, M0 patients underwent FDG PET before RT. Optimal cutoff values of mSUVR were chosen based on overall survival (OS). Independent prognosticators were identified by Cox regression analysis. Results: The most significant cutoff value for mSUVR was 0.9 with respect to OS. Two-year OS was 17% for patients with mSUVR > 0.9 and 49% for those with mSUVR ${\leq}0.9$ (p = 0.01). In a multivariate analysis, including age, performance status, stage, use of chemotherapy, and mSUVR, only performance status (p = 0.05) and mSUVR > 0.9 (p = 0.05) were significant predictors of OS. Two-year OS for patients with both good performance (Eastern Cooperative Oncology Group [ECOG] ${\leq}1$) and mSUVR ${\leq}0.9$ was significantly better than that for patients with either poor performance (ECOG > 1) or mSUVR > 0.9, 23% (71% vs. 23%, p = 0.04). Conclusion: Our results suggested that the mSUVR was a strong prognostic factor among patients with lymph node positive NSCLC following RT. Addition of mSUVR to performance status identifies a subgroup at highest risk for death after RT.

• Investigative Magnetic Resonance Imaging
• /
• v.23 no.4
• /
• pp.351-360
• /
• 2019

Purpose: To investigate noninvasive biomarkers for predicting treatment response in patients with locally advanced HCC who underwent concurrent chemoradiotherapy (CCRTx). Materials and Methods: Thirty patients (55.5 ± 10.2 years old, M:F = 24:6) who underwent CCRTx due to advanced HCC were enrolled. Contrast-enhanced US (CEUS) and dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) were obtained before and immediately after CCRTx. The third CEUS was obtained at one month after CCRTx was completed. Response was assessed at three months after CCRTx based on RECIST 1.1. Quantitative imaging biomarkers measured with CEUS and MRI were compared between groups. A cutoff value was calculated with ROC analysis. Overall survival (OS) was compared by the Breslow method. Results: Twenty-five patients were categorized into the non-progression group and five patients were categorized into the progression group. Peak enhancement of the first CEUS before CCRTx (PE1) was significantly lower in the non-progression group (median, 18.6%; IQR, 20.9%) than that in the progression group (median, 59.1%; IQR, 13.5%; P = 0.002). There was no significant difference in other quantitative biomarkers between the two groups. On ROC analysis, with a cutoff value of 42.6% in PE1, the non-progression group was diagnosed with a sensitivity of 90.9% and a specificity of 100%. OS was also significantly longer in patients with PE1 < 42.6% (P = 0.014). Conclusion: Early treatment response and OS could be predicted by PE on CEUS before CCRTx in patients with HCC.

• Journal of Preventive Medicine and Public Health
• /
• v.40 no.4
• /
• pp.297-304
• /
• 2007

Objectives : We investigated the validity of the dipstick method (Mossman Associates Inc. USA) and the expired CO method to distinguish between smokers and non-smokers. We also elucidated the related factors of the two methods. Methods : This study included 244 smokers and 50 ex-smokers, recruited from smoking cessation clinics at 4 local public health centers, who had quit for over 4 weeks. We calculated the sensitivity, specificity and Kappa coefficient of each method for validity. We obtained ROC curve, predictive value and agreement to determine the cutoff of expired air CO method. Finally, we elucidated the related factors and compared their effect powers using the standardized regression coefficient. Results : The dipstick method showed a sensitivity of 92.6%, specificity of 96.0% and Kappa coefficient of 0.79. The best cutoff value to distinguish smokers was 5-6ppm. At 5 ppm, the expired CO method showed a sensitivity of 94.3%, specificity of 82.0% and Kappa coefficient of 0.73. And at 6 ppm, sensitivity, specificity and Kappa coefficient were 88.5%, 86.0% and 0.64, respectively. Therefore, the dipstick method had higher sensitivity and specificity than the expired CO method. The dipstick and expired CO methods were significantly increased with increasing smoking amount. With longer time since the last smoking, expired CO showed a rapid decrease after 4 hours, whereas the dipstick method showed relatively stable levels for more than 4 hours. Conclusions : The dipstick and expired CO methods were both good indicators for assessing smoking status. However, the former showed higher sensitivity and specificity and stable levels over longer hours after smoking, compared to the expired CO method.