• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.12
• /
• pp.445-452
• /
• 2022

Class distribution of unbalanced data is an important part of the digital world and is a significant part of cybersecurity. Abnormal activity of unbalanced data should be found and problems solved. Although a system capable of tracking patterns in all transactions is needed, machine learning with disproportionate data, which typically has abnormal patterns, can ignore and degrade performance for minority layers, and predictive models can be inaccurately biased. In this paper, we predict target variables and improve accuracy by combining estimates using Synthetic Minority Oversampling Technique (SMOTE) and Light GBM algorithms as an approach to address unbalanced datasets. Experimental results were compared with logistic regression, decision tree, KNN, Random Forest, and XGBoost algorithms. The performance was similar in accuracy and reproduction rate, but in precision, two algorithms performed at Random Forest 80.76% and Light GBM 97.16%, and in F1-score, Random Forest 84.67% and Light GBM 91.96%. As a result of this experiment, it was confirmed that Light GBM's performance was similar without deviation or improved by up to 16% compared to five algorithms.

• Molecules and Cells
• /
• v.39 no.8
• /
• pp.619-624
• /
• 2016

Glioblastoma stem cells (GBM-SCs) are believed to be a subpopulation within all glioblastoma (GBM) cells that are in large part responsible for tumor growth and the high grade of therapeutic resistance that is so characteristic of GBM. MicroRNAs (miR) have been implicated in regulating the expression of oncogenes and tumor suppressor genes in cancer stem cells, including GBM-SCs, and they are a potential target for cancer therapy. In the current study, miR-203 expression was reduced in $CD133^+$ GBM-SCs derived from six human GBM biopsies. MicroRNA-203 transfected GBM-SCs had reduced capacity for self-renewal in the cell sphere assay and increased expression of glial and neuronal differentiation markers. In addition, a reduced proliferation rate and an increased rate of apoptosis were observed. Therefore, miR-203 has the potential to reduce features of stemness, specifically in GBM-SCs, and is a logical target for GBM gene therapy.

• Journal of Korean Neurosurgical Society
• /
• v.52 no.5
• /
• pp.484-487
• /
• 2012

Glioblastoma multiforme (GBM) is the most common primary brain tumor and the most malignant astrocytoma in adults, with rare extra-cranial metastases, especially for subcutaneous metastases. It could be easily misdiagnosed as primary subcutaneous tumor. In this report, we describe a patient with pontine GBM who developed a subcutaneous swelling at the ipsilateral posterior cervical region 8 months after operation, and the pathological and immunocytochemical examination carry the same characteristics as the primary intracranial GBM cells, which defined it as subcutaneous metastasis. GBM with subcutaneous metastasis is extremely rare, and knowledge of a prior intracranial GBM, pathological examinations and immunocytochemical tests with markers typically expressed by GBM are of vital importance for the diagnosis of GBM metastasis. Surgical resection of subcutaneous swelling, followed by chemotherapy and radiotherapy, could be the best strategy of treatment for the patients with GBM subcutaneous metastasis.

• Journal of the Korea Convergence Society
• /
• v.8 no.10
• /
• pp.215-223
• /
• 2017

This article has tried to analyse the effect of the corporate earning return tax empirically through analysis on the impact of previous internal reservation on the dividends rate of the current year. In addition to this, this article has tried to the effectiveness of government policies with leverage ratio as a moderating variable. Moreover, DRF and GBM model were used to see the effect again. As a result of the actual proof analysis, OCF, ROE, FOR have a significance level of 99% in model1, model2, model3. However, ADV and MSE has appeared not to be meaningful in all models. In the result of DRF and GBM model for convergence was higher than GBM in depth and leaves. However, when it comes to a model explaining capability, GBM high than DRF. The further study will be required to examine the effect of government policy by time series analysis in the period of enforcement of the reflux tax, from 2015 to 2017.

• Journal of the Korea Convergence Society
• /
• v.13 no.2
• /
• pp.249-255
• /
• 2022

This study proposed an analysis framework for real-time prediction of CNC processing defects using machine learning-based models that are recently attracting attention as processing defect prediction methods, and applied it to CNC machines. Analysis shows that the XGBoost, CatBoost, and LightGBM models have the same best accuracy, precision, recall, F1 score, and AUC, of which the LightGBM model took the shortest execution time. This short run time has practical advantages such as reducing actual system deployment costs, reducing the probability of CNC machine damage due to rapid prediction of defects, and increasing overall CNC machine utilization, confirming that the LightGBM model is the most effective machine learning model for CNC machines with only basic sensors installed. In addition, it was confirmed that classification performance was maximized when an ensemble model consisting of LightGBM, ExtraTrees, k-Nearest Neighbors, and logistic regression models was applied in situations where there are no restrictions on execution time and computing power.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.8
• /
• pp.3629-3633
• /
• 2014

Background: Glioblastoma (GBM) is an immunosuppressive tumor whose median survival time is only 12-15 months, and patients with GBM have a uniformly poor prognosis. It is known that heredity contributes to formation of glioma, but there are few genetic studies concerning GBM. Materials and Methods: We genotyped six tagging SNPs (tSNP) in Han Chinese GBM and control patients. We used Microsoft Excel and SPSS 16.0 statistical package for statistical analysis and SNP Stats to test for associations between certain tSNPs and risk of GBM in five different models. ORs and 95%CIs were calculated for unconditional logistic-regression analysis with adjustment for age and gender. The SHEsis software platform was applied for analysis of linkage disequilibrium, haplotype construction, and genetic associations at polymorphism loci. Results: We found rs891835 in CCDC26 to be associated with GBM susceptibility at a level of p=0.009. The following genotypes of rs891835 were found to be associated with GBM risk in four different models of gene action: i) genotype GT (OR=2.26; 95%CI, 1.29-3.97; p=0.019) or GG (OR=1.33; 95%CI, 0.23-7.81; p=0.019) in the codominant model; ii) genotypes GT and GG (OR=2.18; 95%CI, 1.26-3.78; p=0.0061) in the dominant model; iii) GT (OR=2.24; 95%CI, 1.28-3.92; p=0.0053) in the overdominant model; iv) the allele G of rs891835 (OR=1.85; 95%CI, 1.14-3.00; p=0.015) in the additive model. In addition, "CG" and "CGGAG" were found by haplotype analysis to be associated with increased GBM risk. In contrast, genotype GG of CCDC26 rs6470745 was associated with decreased GBM risk (OR=0.34; 95%CI, 0.12-1.01; p=0.029) in the recessive model. Conclusions: Our results, combined with those from previous studies, suggest a potential genetic contribution of CCDC26 to GBM progression among Han Chinese.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.2
• /
• pp.463-466
• /
• 2016

Glioblastoma (GBM) is the most common and aggressive type of primary brain neoplasm. The current standard therapy for GBM consists of maximal surgical resection within safe limits, followed by radiation therapy (RT) and chemotherapy with temozolomide. Despite advances in treatment, the prognosis of GBM remains poor. Epileptic seizure is one of the most common symptoms in patients with GBM. Valproic acid (VPA), a histone deacetylase inhibitor, is often used as an anti-epileptic drug in patients with brain neoplasms due to its effectiveness and low toxicity profile. Several in vivo and in vitro studies have indicated that VPA has radiosensitizing effects for gliomas and radioprotective influence on normal brain tissue or hippocampal neurons. The results of several retrospective studies have also indicated potential benefit to improve survival of patients with GBM. Moreover, the promising treatment results of a phase 2 trial of concurrent radiation therapy, temozolomide, and VPA for patients with GBM have been recently reported. The use of VPA in patients with GBM has thus recently receiving more attention. In this article, we review the role of VPA in radiation therapy for GBM, focusing on the clinical evidence.

• BMB Reports
• /
• v.51 no.10
• /
• pp.481-483
• /
• 2018

Glioblastoma (GBM) is the most common and aggressive form of human adult brain malignancy. The identification of the cell of origin harboring cancer-driver mutations is the fundamental issue for understanding the nature of GBM and developing the effective therapeutic target. It has been a long-term hypothesis that neural stem cells in the subventricular zone (SVZ) might be the origin-of-cells in human glioblastoma since they are known to have life-long proliferative activity and acquire somatic mutations. However, the cell of origin for GBM remains controversial due to lack of direct evidence thereof in human GBM. Our recent study using various sequencing techniques in triple matched samples such as tumor-free SVZ, tumor, and normal tissues from human patients identified the clonal relationship of driver mutations between GBM and tumor-free SVZ harboring neural stem cells (NSCs). Tumor-free SVZ tissue away from the tumor contained low-level GBM driver mutations (as low as 1% allelic frequency) that were found in the dominant clones in its matching tumors. Moreover, via single-cell sequencing and microdissection, it was discovered that astrocyte-like NSCs accumulating driver mutations evolved into GBM with clonal expansion. Furthermore, mutagenesis of cancer-driving genes of NSCs in mice leads to migration of mutant cells from SVZ to distant brain and development of high-grade glioma through the aberrant growth of oligodendrocyte precursor lineage. Altogether, the present study provides the first direct evidence that NSCs in human SVZ is the cell of origin that develops the driver mutations of GBM.

• Journal of Life Science
• /
• v.17 no.8 s.88
• /
• pp.1039-1045
• /
• 2007

Glioblastoma multiforme (GBM) is the most frequently occurring brain cancer. Although the existence of cancer stem cells (CSCs) in GBM has been established, there is little evidence to explain the link between CSCs and chemoresistance. In this study, we investigated that only a few cells of A172 and established GBM2 survived after 1,3-bis(2chloroethyl)-1-nitrosourea (BiCNU) exposures and these sur-vived cells resist the subsequent BiCNU treatment. In addition, these BiCNU-resistant small pop-ulations derived from GBM cells increased the phosphorylations of Erk and Akt and highly expressed CD133 stem cell surface marker. Furthermore, we observed that the BiCNU-resistant cancer cells de-rived from GBM have grown tumors when transplanted into severe combined immuno-deficient (SCID) mouse brain. These results demonstrate that BiCNU-resistant subpopulation cells derived from GBM have cancer stem-like cell properties. Therefore, it may provide provide further evidence that CSCs in GBM have chemotherapeutic drug resistance.

• Journal of Korean Neurosurgical Society
• /
• v.38 no.6
• /
• pp.475-477
• /
• 2005

The tendency of glioblastoma multiforme[GBM] to metastasize to the cerebrospinal fluidis well documented. However, symptomatic intradural extramedullary metastasis of GBM in the spinal cord are rarely reported. A 31-year-old female with a previously treated supratentorial GBM presented with back pain and lower extremities weakness. Magnetic resonance imaging of the thoracic spine demonstrated an intradural extramedullary mass at levels of T2-T4 and arachnoid membrane enhancement. The patient underwent an operation. Pathologic diagnosis was confirmed as spinal metastases of GBM. We present a case of spinal metastases from supratentorial GBM presented with paraparesis.