• The Journal of the Korea Contents Association
• /
• v.13 no.12
• /
• pp.829-839
• /
• 2013

This study is a descriptive research study with the purpose of predicting and comparing factors of depression affecting residents in a metropolitan city by using logistic regression analysis and decision-making tree analysis. The subjects for the study were 462 residents ($20{\leq}aged{\angle}65$) in a metropolitan city. This study collected data between October 7, 2011 and October 21, 2011 and analyzed them with frequency analysis, percentage, the mean and standard deviation, ${\chi}^2$-test, t-test, logistic regression analysis, roc curve, and a decision-making tree by using SPSS 18.0 program. The common predicting variables of depression in community residents were social dysfunction, perceived physical symptom, and family support. The specialty and sensitivity of logistic regression explained 93.8% and 42.5%. The receiver operating characteristic (roc) curve was used to determine an optimal model. The AUC (area under the curve) was .84. Roc curve was found to be statistically significant (p=<.001). The specialty and sensitivity of decision-making tree analysis were 98.3% and 20.8% respectively. As for the whole classification accuracy, the logistic regression explained 82.0% and the decision making tree analysis explained 80.5%. From the results of this study, it is believed that the sensitivity, the classification accuracy, and the logistics regression analysis as shown in a higher degree may be useful materials to establish a depression prediction model for the community residents.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.4
• /
• pp.2567-2570
• /
• 2013

Background: This study analyzed Surveillance, Epidemiology and End Results (SEER) data to assess if socio-economic factors (SEFs) impact on endometrial cancer survival. Materials and Methods: Endometrial cancer patients treated from 2004-2007 were included in this study. SEER cause specific survival (CSS) data were used as end points. The areas under the receiver operating characteristic (ROC) curve were computed for predictors. Time to event data were analyzed with Kaplan-Meier method. Univariate and multivariate analyses were used to identify independent risk factors. Results: This study included 64,710 patients. The mean follow up time (S.D.) was 28.2 (20.8) months. SEER staging (ROC area of 0.81) was the best pretreatment predictor of CSS. Histology, grade, race/ethnicity and county level family income were also significant pretreatment predictors. African American race and low income neighborhoods decreased the CSS by 20% and 3% respectively at 5 years. Conclusions: This study has found significant endometrial survival disparities due to SEFs. Future studies should focus on eliminating socio-economic barriers to good outcomes.

• Journal of the Korean Physical Society
• /
• v.80
• /
• pp.247-256
• /
• 2022

Our objective was to investigate radiomics signatures and prediction models defined by four segmentation methods in using 2-[¹⁸F]fluoro-2-deoxy-d-glucose positron emission tomography (¹⁸F-FDG PET) imaging of lung metastases of soft-tissue sarcomas (STSs). For this purpose, three fixed threshold methods using the standardized uptake value (SUV) and gradient-based edge detection (ED) were used for tumor delineation on the PET images of STSs. The Dice coefficients (DCs) of the segmentation methods were compared. The least absolute shrinkage and selection operator (LASSO) regression and Spearman's rank, and Friedman's ANOVA test were used for selection and validation of radiomics features. The developed radiomics models were assessed using ROC (receiver operating characteristics) curve and confusion matrices. According to the results, the DC values showed the biggest difference between SUV40% and other segmentation methods (DC: 0.55 and 0.59). Grey-level run-length matrix_run-length nonuniformity (GLRLM_RLNU) was a common radiomics signature extracted by all segmentation methods. The multivariable logistic regression of ED showed the highest area under the ROC (receiver operating characteristic) curve (AUC), sensitivity, specificity, and accuracy (AUC: 0.88, sensitivity: 0.85, specificity: 0.74, accuracy: 0.81). In our research, the ED method was able to derive a significant model of radiomics. GLRLM_RLNU which was selected from all segmented methods as a meaningful feature was considered the obvious radiomics feature associated with the heterogeneity and the aggressiveness. Our results have apparently showed that radiomics signatures have the potential to uncover tumor characteristics.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.573-585
• /
• 2018

In this study, we fit the logistic regression model and naïve Bayesian classifier model using 11 risk factors to predict the incidence rate probability for type 2 diabetes mellitus. We then introduce how to construct a nomogram that can help people visually understand it. We use data from the 2013-2015 Korean National Health and Nutrition Examination Survey (KNHANES). We take 3 interactions in the logistic regression model to improve the quality of the analysis and facilitate the application of the left-aligned method to the Bayesian nomogram. Finally, we compare the two nomograms and examine their utility. Then we verify the nomogram using the ROC curve.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.15
• /
• pp.6255-6259
• /
• 2015

Purpose: To evaluate the prognostic value of alkaline phosphatase (ALP) and gamma-glutamyltransferase (GGT) in gallbladder cancer (GBC). Materials and Methods: Serum ALP and GGT levels and clinicopathological parameters were retrospectively evaluated in 199 GBC patients. Receiver operating characteristic (ROC) curve analysis was performed to determine the cut-off values of ALP and GGT. Then, associations with overall survival were assessed by multivariate analysis. Based on the significant factors, a prognostic score model was established. Results: By ROC curve analysis, $ALP{\geq}210U/L$ and $GGT{\geq}43U/L$ were considered elevated. Overall survival for patients with elevated ALP and GGT was significantly worse than for patients within the normal range. Multivariate analysis showed that the elevated ALP, GGT and tumor stage were independent prognostic factors. Giving each positive factor a score of 1, we established a preoperative prognostic score model. Varied outcomes would be significantly distinguished by the different score groups. By further ROC curve analysis, the simple score showed great superiority compared with the widely used TNM staging, each of the ALP or GGT alone, or traditional tumor markers such as CEA, AFP, CA125 and CA199. Conclusions: Elevated ALP and GGT levels were risk predictors in GBC patients. Our prognostic model provides infomration on varied outcomes of patients from different score groups.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.29 no.3
• /
• pp.129-136
• /
• 2018

Objectives: This study aimed to evaluate the efficacy of the attention-deficit/hyperactivity disorder (ADHD)-After School Checklist (ASK) by comparing the results of the Comprehensive Attention Test (CAT) and Clinical Global Impression-Severity (CGI-S) Scale and then by calculating the area under the receiver operating characteristic (ROC) curve. Methods: We performed correlation analyses on the ASK and CAT results and then the ASK and CGI-S results. We created a ROC curve and evaluated performance on the ASK as a diagnostic tool. We then analyzed the test results of 1348 subjects (male 56.8%), including 1201 subjects in the general population and 147 ADHD subjects, aged 6-15 years, from kindergarten to middle school in Seoul and Gyeonggi province, South Korea. Results: According to the correlation analyses, ASK scores and the Attention Quotient (AQ) of CAT scores showed a significant correlation of -0.20--0.29 (p<0.05). The t-test between ADHD scores and CGI-S also showed a significant correlation (t=-2.55, p<0.05). The area under the ROC curve was calculated as 0.81, indicating good efficacy of the ASK, and the cut-off score was calculated as 15.5. Conclusion: The ASK can be used as a valid tool not only to evaluate functional impairment of ADHD children and adolescents but also to screen ADHD.

• Korean Journal of Radiology
• /
• v.23 no.1
• /
• pp.139-149
• /
• 2022

Objective: To compare the effects of bone suppression imaging using deep learning (BSp-DL) based on a generative adversarial network (GAN) and bone subtraction imaging using a dual energy technique (BSt-DE) on radiologists' performance for pulmonary nodule detection on chest radiographs (CXRs). Materials and Methods: A total of 111 adults, including 49 patients with 83 pulmonary nodules, who underwent both CXR using the dual energy technique and chest CT, were enrolled. Using CT as a reference, two independent radiologists evaluated CXR images for the presence or absence of pulmonary nodules in three reading sessions (standard CXR, BSt-DE CXR, and BSp-DL CXR). Person-wise and nodule-wise performances were assessed using receiver-operating characteristic (ROC) and alternative free-response ROC (AFROC) curve analyses, respectively. Subgroup analyses based on nodule size, location, and the presence of overlapping bones were performed. Results: BSt-DE with an area under the AFROC curve (AUAFROC) of 0.996 and 0.976 for readers 1 and 2, respectively, and BSp-DL with AUAFROC of 0.981 and 0.958, respectively, showed better nodule-wise performance than standard CXR (AUAFROC of 0.907 and 0.808, respectively; p ≤ 0.005). In the person-wise analysis, BSp-DL with an area under the ROC curve (AUROC) of 0.984 and 0.931 for readers 1 and 2, respectively, showed better performance than standard CXR (AUROC of 0.915 and 0.798, respectively; p ≤ 0.011) and comparable performance to BSt-DE (AUROC of 0.988 and 0.974; p ≥ 0.064). BSt-DE and BSp-DL were superior to standard CXR for detecting nodules overlapping with bones (p < 0.017) or in the upper/middle lung zone (p < 0.017). BSt-DE was superior (p < 0.017) to BSp-DL in detecting peripheral and sub-centimeter nodules. Conclusion: BSp-DL (GAN-based bone suppression) showed comparable performance to BSt-DE and can improve radiologists' performance in detecting pulmonary nodules on CXRs. Nevertheless, for better delineation of small and peripheral nodules, further technical improvements are required.

• Journal of the Korean Society of Radiology
• /
• v.81 no.5
• /
• pp.1164-1174
• /
• 2020

Purpose To evaluate a deep learning model to predict recurrence of thyroid tumor using preoperative ultrasonography (US). Materials and Methods We included representative images from 229 US-based patients (male:female = 42:187; mean age, 49.6 years) who had been diagnosed with thyroid cancer on preoperative US and subsequently underwent thyroid surgery. After selecting each representative transverse or longitudinal US image, we created a data set from the resulting database of 898 images after augmentation. The Python 2.7.6 and Keras 2.1.5 framework for neural networks were used for deep learning with a convolutional neural network. We compared the clinical and histological features between patients with and without recurrence. The predictive performance of the deep learning model between groups was evaluated using receiver operating characteristic (ROC) analysis, and the area under the ROC curve served as a summary of the prognostic performance of the deep learning model to predict recurrent thyroid cancer. Results Tumor recurrence was noted in 49 (21.4%) among the 229 patients. Tumor size and multifocality varied significantly between the groups with and without recurrence (p < 0.05). The overall mean area under the curve (AUC) value of the deep learning model for prediction of recurrent thyroid cancer was 0.9 ± 0.06. The mean AUC value was 0.87 ± 0.03 in macrocarcinoma and 0.79 ± 0.16 in microcarcinoma. Conclusion A deep learning model for analysis of US images of thyroid cancer showed the possibility of predicting recurrence of thyroid cancer.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.28 no.2
• /
• pp.415-433
• /
• 1998

Digora system is an intraoral indirect digital radiography system utilizing storage phosphor image plate. It has wide dynamic range which allows it to decrease the patient s exposure time and may increase diagnostic ability through image processing (such as edge enhancement, grey scale conversion, brightness change, and contrast enhancement). And also, it can transmit and storage image information. The purpose of this study was to evaluate the diagnostic ability of artificial proximal caries between Conventional radiograph and Digora images(unenhanced image, brightness & contrast controlled image, and edge enhanced image). ROC(Receiver Operating Characteristic) analysis, paired t-tests, and F-tests were done for the statistical evaluation of detectability. The following results were acquired: 1. In Grade I lesions, the mean ROC areas of Conventional radiograph, Digora unenhanced image, Digora controlled image, and Digora edge enhanced image were 0.953, 0.933, 0.965, 0.978 (p>0.05). 2. In Grade II lesions, the mean ROC areas of Conventional radiograph, Digora unenhanced image, Digora controlled image, and Digora edge enhanced image were 0.969, 0.964, 0.988, 0.994. Among theses areas, there was just statistical significance between Diagnostic abilities of Digora edge enhanced image and Conventional radiograph (p<0.05). 3. In the Interobserver variability, the ROC curve areas of Digora edge enhanced image was lowermost in these areas, regardless of the Carious lesion depths. In conclusion, intraoral indirect digital system, Digora system, has the potential possibility as an alternative of Conventional radiograph in the diagnosis of proximal caries.

• The Journal of Korean Obstetrics and Gynecology
• /
• v.25 no.3
• /
• pp.95-102
• /
• 2012

Purpose: The purpose of this study is to define the cut off values of cold hypersensitivity of hands by using digital infrared thermographic imaging(DITI). Methods: Thermographic images of 130 patients with cold hypersensitivity of hands(CHHG, n=65) and non-cold hypersensitivity of hands(NCHHG, n=65) were retrospectively reviewed. We used the temperature difference the palm(PC8) and the upper arm(LU4) for diagnosing cold hypersensitivity of hands. The temperature differences of between two groups were analysed using independent samples t-tests. The cut off values were calculated by ROC curve analysis. Analyses were undertaken using SPSS version 17.0. P value of < 0.05 was considered significant. Results: The temperature difference the palm(PC8) and the upper arm(LU4) were significantly different between groups(p < 0.001). Using receiver operating characteristic curve analysis, the sensitivity, specificity, and area under the curve were 70.8%, 73.8%, respectively both hands. The AUC was 0.822 on right hand and 0.818 on left hand. The optimum cut-off value was defined as $-0.05^{\circ}C$. Conclusions: These results suggest that DITI is a reliable instrument for estimating the cold hypersensitivity of hands.