• The Korean Journal of Applied Statistics
• /
• v.32 no.4
• /
• pp.593-605
• /
• 2019

Significant literature exists on the area under the ROC curve (AUC) and the volume under the ROC surface (VUS) which are statistical measures of the discriminant power of classification models. Whereas the partial AUC is restricted on the false positive rate, the two-way partial AUC is restricted on both the false positive rate and true positive rate, which could be more efficient and accurate than partial AUC. The two-way partial AUC was suggested as more efficient and accurate than the partial AUC. Partial VUS as well as the three-way partial VUS were also developed for the ROC surface. A proposed AUC is expressed in this paper with probability and integration using two truncated distribution functions restricted on both the false positive rate and true positive rate. It is also found that this AUC has a relation with the two-way partial AUC. The three-way partial VUS for the ROC surface is also related to the VUS using truncated distribution functions. These AUC and VUS are represented and estimated in terms of Mann-Whitney statistics. Their parametric and non-parametric estimation methods are explored based on normal distributions and random samples.

• BMB Reports
• /
• v.56 no.6
• /
• pp.347-352
• /
• 2023

The protein family of poly (ADP-ribose) polymerases (PARPs) is comprised of multifunctional nuclear enzymes. Several PARP inhibitors have been developed as new anticancer drugs to combat resistance to chemotherapy. Herein, we characterized PARP4 mRNA expression profiles in cisplatin-sensitive and cisplatin-resistant ovarian cancer cell lines. PARP4 mRNA expression was significantly upregulated in cisplatin-resistant ovarian cancer cell lines, and this upregulation was associated with the hypomethylation of specific cytosine-phosphate-guanine (CpG) sites (cg18582260 and cg17117459) on its promoter. Reduced PARP4 expression was restored by treating cisplatin-sensitive cell lines with a demethylation agent, implicating the epigenetic regulation of PARP4 expression by promoter methylation. Depletion of PARP4 expression in cisplatin-resistant cell lines reduced cisplatin chemoresistance and promoted cisplatin-induced DNA fragmentation. The differential mRNA expression and DNA methylation status at specific PARP4 promoter CpG sites (cg18582260 and cg17117459) according to cisplatin responses, was further validated in primary ovarian tumor tissues. The results showed significantly increased PARP4 mRNA expressions and decreased DNA methylation levels at specific PARP4 promoter CpG sites (cg18582260 and cg17117459) in cisplatin-resistant patients. Additionally, the DNA methylation status at cg18582260 CpG sites in ovarian tumor tissues showed fairly clear discrimination between cisplatin-resistant patients and cisplatin-sensitive patients, with high accuracy (area under the curve = 0.86, P = 0.003845). Our findings suggest that the DNA methylation status of PARP4 at the specific promoter site (cg18582260) may be a useful diagnostic biomarker for predicting the response to cisplatin in ovarian cancer patients.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.31-40
• /
• 2024

In the realm of domestic online fashion platform industry the manual registration of product information by individual business owners leads to inconvenience and reliability issues, especially when dealing with simultaneous registrations of numerous product groups. Moreover, bias is significantly heightened due to the low quality of product images and an imbalance in data quantity. Therefore, this study proposes a ResNet50 model aimed at minimizing data bias through oversampling techniques and conducting multiple classifications for 13 fashion categories. Transfer learning is employed to optimize resource utilization and reduce prolonged learning times. The results indicate improved discrimination of up to 33.4% for data augmentation in classes with insufficient data compared to the basic convolution neural network (CNN) model. The reliability of all outcomes is underscored by precision and affirmed by the recall curve. This study is suggested to advance the development of the domestic online fashion platform industry to a higher echelon.

• Korean Journal of Radiology
• /
• v.25 no.1
• /
• pp.74-85
• /
• 2024

Objective: Idiopathic intracranial hypertension (IIH) is a condition of unknown etiology associated with venous sinus stenosis. This study aimed to develop a magnetic resonance venography (MRV)-based radiomics model for predicting a high trans-stenotic pressure gradient (TPG) in IIH patients diagnosed with venous sinus stenosis. Materials and Methods: This retrospective study included 105 IIH patients (median age [interquartile range], 35 years [27-42 years]; female:male, 82:23) who underwent MRV and catheter venography complemented by venous manometry. Contrast enhanced-MRV was conducted under 1.5 Tesla system, and the images were reconstructed using a standard algorithm. Shape features were derived from MRV images via the PyRadiomics package and selected by utilizing the least absolute shrinkage and selection operator (LASSO) method. A radiomics score for predicting high TPG (≥ 8 mmHg) in IIH patients was formulated using multivariable logistic regression; its discrimination performance was assessed using the area under the receiver operating characteristic curve (AUROC). A nomogram was constructed by incorporating the radiomics scores and clinical features. Results: Data from 105 patients were randomly divided into two distinct datasets for model training (n = 73; 50 and 23 with and without high TPG, respectively) and testing (n = 32; 22 and 10 with and without high TPG, respectively). Three informative shape features were identified in the training datasets: least axis length, sphericity, and maximum three-dimensional diameter. The radiomics score for predicting high TPG in IIH patients demonstrated an AUROC of 0.906 (95% confidence interval, 0.836-0.976) in the training dataset and 0.877 (95% confidence interval, 0.755-0.999) in the test dataset. The nomogram showed good calibration. Conclusion: Our study presents the feasibility of a novel model for predicting high TPG in IIH patients using radiomics analysis of noninvasive MRV-based shape features. This information may aid clinicians in identifying patients who may benefit from stenting.

• Korean Journal of Radiology
• /
• v.24 no.7
• /
• pp.668-680
• /
• 2023

Objective: To compare multiparametric magnetic resonance imaging (MRI) features of intraductal carcinoma of the prostate (IDC-P) with those of prostatic acinar adenocarcinoma (PAC) and develop prediction models to distinguish IDC-P from PAC and IDC-P with a high proportion (IDC ≥ 10%, hpIDC-P) from IDC-P with a low proportion (IDC < 10%, lpIDC-P) and PAC. Materials and Methods: One hundred and six patients with hpIDC-P, 105 with lpIDC-P and 168 with PAC, who underwent pretreatment multiparametric MRI between January 2015 and December 2020 were included in this study. Imaging parameters, including invasiveness and metastasis, were evaluated and compared between the PAC and IDC-P groups as well as between the hpIDC-P and lpIDC-P subgroups. Nomograms for distinguishing IDC-P from PAC, and hpIDC-P from lpIDC-P and PAC, were made using multivariable logistic regression analysis. The discrimination performance of the models was assessed using the receiver operating characteristic area under the curve (ROC-AUC) in the sample, where the models were derived from without an independent validation sample. Results: The tumor diameter was larger and invasive and metastatic features were more common in the IDC-P than in the PAC group (P < 0.001). The distribution of extraprostatic extension (EPE) and pelvic lymphadenopathy was even greater, and the apparent diffusion coefficient (ADC) ratio was lower in the hpIDC-P than in the lpIDC-P group (P < 0.05). The ROC-AUCs of the stepwise models based solely on imaging features for distinguishing IDC-P from PAC and hpIDC-P from lpIDC-P and PAC were 0.797 (95% confidence interval, 0.750-0.843) and 0.777 (0.727-0.827), respectively. Conclusion: IDC-P was more likely to be larger, more invasive, and more metastatic, with obviously restricted diffusion. EPE, pelvic lymphadenopathy, and a lower ADC ratio were more likely to occur in hpIDC-P, and were also the most useful variables in both nomograms for predicting IDC-P and hpIDC-P.

• Korean Journal of Radiology
• /
• v.2 no.1
• /
• pp.1-7
• /
• 2001

Objective: To determine the usefulness of perfusion MR imaging in assessing the histologic grade of cerebral gliomas. Materials and Methods: In order to determine relative cerebral blood volume (rCBV), 22 patients with pathologically proven gliomas (9 glioblastomas, 9 anaplastic gliomas and 4 low-grade gliomas) underwent dynamic contrast-enhanced T2^*-weighted and conventional T1- and T2-weighted imaging. rCBV maps were obtained by fitting a gamma-variate function to the contrast material concentration versus time curve. rCBV ratios between tumor and normal white matter (maximum rCBV of tumor / rCBV of contralateral white matter) were calculated and compared between glioblastomas, anaplastic gliomas and low-grade gliomas. Results: Mean rCBV ratios were 4.90°±1.01 for glioblastomas, 3.97°±0.56 for anaplastic gliomas and 1.75°±1.51 for low-grade gliomas, and were thus significantly different; p < .05 between glioblastomas and anaplastic gliomas, p < .05 between anaplastic gliomas and low-grade gliomas, p < .01 between glioblastomas and low-grade gliomas. The rCBV ratio cutoff value which permitted discrimination between high-grade (glioblastomas and anaplastic gliomas) and low-grade gliomas was 2.60, and the sensitivity and specificity of this value were 100% and 75%, respectively. Conclusion: Perfusion MR imaging is a useful and reliable technique for estimating the histologic grade of gliomas.

• Annals of Hepato-Biliary-Pancreatic Surgery
• /
• v.28 no.2
• /
• pp.229-237
• /
• 2024

Backgrounds/Aims: While patients with borderline resectable pancreatic cancer (BRPC) are a target population for neoadjuvant chemotherapy (NAC), formal guidelines for neoadjuvant therapy are lacking. We assessed the perioperative and oncological outcomes in patients with BRPC undergoing NAC with FOLFIRINOX for patients undergoing upfront surgery (US). Methods: The AHPBA criteria for borderline resectability and/or a CA19-9 level > 100 µ/mL defined borderline resectable tumors retrieved from a prospectively populated institutional registry from 2007 to 2020. The primary outcome was overall survival (OS) at 1 and 3 years. A Cox Proportional Hazard model based on intention to treat was used. A receiver-operator characteristics (ROC) curve was constructed to assess the discriminatory capability of the use of CA19-9 > 100 µ/mL to predict resectability and mortality. Results: Forty BRPC patients underwent NAC, while 46 underwent US. The median OS with NAC was 19.8 months (interquartile range [IQR], 10.3-44.24) vs. 10.6 months (IQR, 6.37-17.6) with US. At 1 year, 70% of the NAC group and 41.3% of the US group survived (p = 0.008). At 3 years, 42.5 % of the NAC group and 10.9% of the US group survived (p = 0.001). NAC significantly reduced the hazard of death (adjusted hazard ratio, 0.20; 95% confidence interval, 0.07-0.54; p = 0.001). CA19-9 > 100 µ/mL showed poor discrimination in predicting mortality, but was a moderate predictor of resectability. Conclusions: We found a survival benefit of NAC with FOLFIRINOX for BRPC. Greater pre-treatment of CA19-9 and multivessel involvement on initial imaging were associated with progression of the disease following NAC.

• The Korean journal of internal medicine
• /
• v.39 no.2
• /
• pp.248-260
• /
• 2024

Background/Aims: The model for end-stage liver disease (MELD) serves as an indicator for short-term mortality among patients diagnosed with liver cirrhosis (LC) and is used to prioritize patients for liver transplantation. In 2021, the updated version of MELD, MELD-3.0, was introduced to improve the accuracy of the mortality prediction of MELD. Therefore, this study aimed to compare the efficacy of MELD 3.0 and MELD-Na in predicting mortality among Korean patients with LC. Methods: A retrospective review was conducted using the medical records of patients diagnosed with LC who were admitted to Konkuk University Hospital From 2011 to 2021. The study calculated the predictive values of MELD-Na and MELD-3.0 for 3- and 6-months mortality using the area under the receiver operating curve (AUROC) and compared the results using the DeLong test. Results: Of the 3,034 patients enrolled in the study, 339 (11.2%) died within 3 months and 421 (14.4%) died within 6 months. The AUROCs values for predicting 3 months mortality were 0.846 for MELD-Na and 0.851 for MELD-3.0. The corresponding AUROC values for predicting 6 months mortality were 0.843 for MELD-Na and 0.848 for MELD-3.0. MELD-3.0 exhibited better discrimination ability than MELD-Na for both 3 (p = 0.03) and 6 months mortality (p = 0.01). Conclusions: Our study found a significant difference between the performance of MELD-3.0 and MELD-Na in Korean patients with LC.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.16
• /
• pp.6923-6928
• /
• 2015

Background: Early detection of various kinds of cancers nowadays is needed including colorectal cancer due to the highly significant effects in improving cancer treatment. The aim of this study was to evaluate the potential value of adiponectin, visfatin and resistin as early biomarkers for colorectal cancer patients. Materials and Methods: Serum levels of adiponectin, visfatin and resistin were measured by a sandwich-enzyme-linked (ELISA) assay technique in 114 serum samples comprising 34 patients with colorectal cancer (CRC), 27 with colonic polyps (CP), 24 with inflammatory bowel disease (IBD) and 29 healthy controls. The diagnostic accuracy of each serum marker was evaluated using receiver-operating characteristic (ROC) curve analysis. Results: The mean concentration of adiponectin was significantly higher in CRC and CP groups than IBD and control groups (P-value <0.05). Also the mean concentration of serum resistin was significantly elevated in the IBD and control groups compared to CRC and CP groups (P-value = 0.014). However, no significant difference was noted in patients of the CRC and CP groups. On the other hand, the mean concentration of visfatin was significantly elevated in CRC and control groups compared to CP and IBD groups (P-value = 0.03). ROC analysis curves for the studied markers revealed that between CRC and IBD groups serum level of adiponectin had a sensitivity of 76.7% and a specificity of 76% at a cut off value of 3940, +LR being 3.2 and -LR 0.31 with AUC 0.852, while serum level of adiponectin between CP and IBD had a sensitivity of 77.8% and a specificity of 75% at a cut off value of 3300, with +LR=3.11 and -LR = 0.3 with AUC 0.852. On the other hand the serum level of visfatin between CRC and CP groups had a sensitivity of 65.5% and a specificity of 66.7 at a cut off value of 2.4, +LR being 1.67 and -LR 0.52 with AUC 0.698. Also the serum level of resistin had a sensitivity of 62.5% and a specificity of 70.3% at a cut off value of 24500, with +LR=2.1 and -LR = 0.53 with AUC 0.685 between control and other groups. On the other hand by comparing control vs CP groups resistin had a sensitivity of 81.8% and a specificity of 70.8% at a cut off value of 17700, with +LR=2.8 and -LR = 0.26 with AUC 0.763 while visfatin had a sensitivity of 68.2% and a specificity of 70.8% at a cut off value of 2.7, with +LR=2.34 and -LR = 0.0.45 with AUC 0.812. Conclusions: These findings support potential roles of adiponectin, visfatin and resistin in early detection of CRC and discrimination of different groups of CRC, CP or IBD patients from normal healthy individuals.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.1
• /
• pp.363-368
• /
• 2014

Determination of the cause of malignant pleural effusions is important for treatment and management, especially in cases of unknown primaries. There are limited biomarkers available for prediction of the cause of malignant pleural effusion in clinical practice. Hence, we evaluated pleural levels of five tumor biomarkers (CEA, AFP, CA125, CA153 and CA199) in predicting the cause of malignant pleural effusion in a retrospective study. Kruskal-Wallis or Mann-Whitney U tests were carried out to compare levels of tumor markers in pleural effusion among different forms of neoplasia - lung squamous cell carcinoma, adenocarcinoma, or small cell carcinoma, mesothelioma, breast cancer, lymphoma/leukemia and miscellaneous. Receiver operator characteristic analysis was performed to evaluate sensitivity and specificity of biomarkers. The Kruskal-Wallis test showed significant differences in levels of pleural effusion CEA (P<0.01), AFP (P<0.01), CA153 (P<0.01) and CA199 (P<0.01), but not CA125 (P>0.05), among the seven groups. Receiver operator characteristic analysis showed that, compared with other four tumor markers, CA153 was the best biomarker in diagnosing malignant pleural effusions of lung adenocarcinoma (area under curve (AUC): 0.838 (95%confidence interval: 0.787, 0.888); cut-off value: 10.2U/ml; sensitivity: 73.2% (64.4-80.8)%, specificity: 85.2% (77.8-90.8)%), lung squamous cell carcinoma (AUC: 0.716 (0.652, 0.780); cut-off value: 14.2U/ml; sensitivity: 57.6% (50.7-64.3)%, specificity: 91.2% (76.3-98.0)%), and small-cell lung cancer (AUC: 0.812 (0.740, 0.884); cut-off value: 9.7U/ml; sensitivity: 61.5% (55.0-67.8)%, specificity: 94.1% (71.2-99.0)%); CEA was the best biomarker in diagnosing MPEs of mesothelioma (AUC: 0.726 (0.593, 0.858); cut-off value: 1.43ng/ml; sensitivity: 83.7% (78.3-88.2)%, specificity: 61.1% (35.8-82.6)%) and lymphoma/leukemia (AUC: 0.923 (0.872, 0.974); cut-off value: 1.71ng/ml; sensitivity: 82.8% (77.4-87.3)%, specificity: 92.3% (63.9-98.7)%). Thus CA153 and CEA appear to be good biomarkers in diagnosing different causes of malignant pleural effusion. Our findings implied that the two tumor markers may improve the diagnosis and treatment for effusions of unknown primaries.