• Korean Journal of Radiology
• /
• v.21 no.10
• /
• pp.1115-1125
• /
• 2020

Combined hepatocellular-cholangiocarcinoma (cHCC-CCA) is a primary liver cancer (PLC) with both hepatocytic and cholangiocytic phenotypes. Recently, the World Health Organization (WHO) updated its histological classification system for cHCC-CCA. Compared to the previous WHO histological classification system, the new version no longer recognizes subtypes of cHCC-CCA with stem cell features. Furthermore, some of these cHCC-CCA subtypes with stem cell features have been recategorized as either hepatocellular carcinomas (HCCs) or intrahepatic cholangiocarcinomas (ICCs). Additionally, distinctive diagnostic terms for intermediate cell carcinomas and cholangiolocarcinomas (previous cholangiolocellular carcinoma subtype) are now recommended. It is important for radiologists to understand these changes because of its potential impact on the imaging-based diagnosis of HCC, particularly because cHCC-CCAs frequently manifest as HCC mimickers, ICC mimickers, or as indeterminate on imaging studies. Therefore, in this review, we introduce the 2019 WHO classification system for cHCC-CCA, illustrate important imaging features characteristic of its subtypes, discuss the impact on imaging-based diagnosis of HCC, and address other important considerations.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.7
• /
• pp.1236-1241
• /
• 2016

Cancer has been the most frequent in Korea, and pathogenesis and progression of cancer have been known to be occurred through various causes and stages. Recently, the research of chromosomal and genetic disorder and the research about prognostic factor to predict occurrence, recurrence and progress of chromosomal and genetic disorder have been performed actively. In this paper, we analyzed DNA methylation data downloaded from TCGA (The Cancer Genome Atlas), open database, to research bladder cancer which is the most frequent among urinary system cancers. Using three level of methylation data which had the most preprocessing, 59 candidate CpG island were extracted from 480,000 CpG island, and then we analyzed extracted CpG island applying data mining technique. As a result, cg12840719 CpG island were analyzed significant, and in Cox's regression we can find the CpG island with high relative risk in comparison with other CpG island. Shown in the result of classification analysis, the CpG island which have high correlation with bladder cancer are cg03146993, cg07323648, cg12840719, cg14676825 and classification accuracy is about 76%. Also we found out that positive predictive value, the probability which predicts cancer in case of cancer was 72.4%. Through the verification of candidate CpG island from the result, we can utilize this method for diagnosing and treating cancer.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.3
• /
• pp.360-366
• /
• 2008

The initial fuzzy partitions in fuzzy rule-based classification systems are determined by considering the domain region of each attribute with the given data, and the optimal classification boundaries within the fuzzy partitions can be discovered by tuning their parameters using various learning processes such as neural network, genetic algorithm, and so on. In this paper, we propose a selection method for fuzzy partition based on statistical information to maximize the performance of pattern classification without learning processes where statistical information is used to extract the uncertainty regions (i.e., the regions which the classification boundaries in pattern classification problems are determined) in each input attribute from the numerical data. Moreover the methods for extracting the candidate rules which are associated with the partition intervals generated by statistical information and for minimizing the coupling problem between the candidate rules are additionally discussed. In order to show the effectiveness of the proposed method, we compared the classification accuracy of the proposed with those of conventional methods on the IRIS and New Thyroid Cancer data. From experimental results, we can confirm the fact that the proposed method only considering statistical information of the numerical patterns provides equal to or better classification accuracy than that of the conventional methods.

• Journal of Korea Multimedia Society
• /
• v.19 no.4
• /
• pp.736-746
• /
• 2016

In this paper, we propose an automatic prostate cancer detection method using position, signal intensity and texture feature based on SVM in multi-parametric MR images. First, to align the prostate on DWI and ADC map to T2wMR, the transformation parameters of DWI are estimated by normalized mutual information-based rigid registration. Then, to normalize the signal intensity range among inter-patient images, histogram stretching is performed. Second, to detect prostate cancer areas in T2wMR, SVM classification with position, signal intensity and texture features was performed on T2wMR, DWI and ADC map. Our feature classification using multi-parametric MR imaging can improve the prostate cancer detection rate on T2wMR.

• Journal of Multimedia Information System
• /
• v.8 no.3
• /
• pp.175-182
• /
• 2021

Oral cancer is ranked second most diagnosed cancer among Indian population and ranked sixth all around the world. Oral cancer is one of the deadliest cancers with high mortality rate and very less 5-year survival rates even after treatment. It becomes necessary to detect oral malignancies as early as possible so that timely treatment may be given to patient and increase the survival chances. In recent years deep learning based frameworks have been proposed by many researchers that can detect malignancies from medical images. In this paper we have proposed a deep learning-based framework which detects oral cancer from histopathology images very efficiently. We have designed our model to split the color channels and extract deep features from these individual channels rather than single combined channel with the help of Efficient NET B3. These features from different channels are fused by using feature fusion module designed as a layer and placed before dense layers of Efficient NET. The experiments were performed on our own dataset collected from hospitals. We also performed experiments of BreakHis, and ICML datasets to evaluate our model. The results produced by our model are very good as compared to previously reported results.

• ETRI Journal
• /
• v.41 no.3
• /
• pp.358-370
• /
• 2019

The main aim of this study is to select the optimal set of genes from microarray cancer datasets that contribute to the prediction of specific cancer types. This study proposes the enhancement of the feature selection filter algorithm based on Joe's normalized mutual information and its use for gene selection. The proposed algorithm is implemented and evaluated on seven benchmark microarray cancer datasets, namely, central nervous system, leukemia (binary), leukemia (3 class), leukemia (4 class), lymphoma, mixed lineage leukemia, and small round blue cell tumor, using five well-known classifiers, including the naive Bayes, radial basis function network, instance-based classifier, decision-based table, and decision tree. An average increase in the prediction accuracy of 5.1% is observed on all seven datasets averaged over all five classifiers. The average reduction in training time is 2.86 seconds. The performance of the proposed method is also compared with those of three other popular mutual information-based feature selection filters, namely, information gain, gain ratio, and symmetric uncertainty. The results are impressive when all five classifiers are used on all the datasets.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.12
• /
• pp.1172-1180
• /
• 2004

Recently, the gene expression data, product of high-throughput technology, appeared in earnest and the studies related with it (so-called bioinformatics) occupied an important position in the field of biological and medical research. The microarray is a revolutionary technology which enables us to monitor several thousands of genes simultaneously and thus to gain an insight into the phenomena in the human body (e.g. the mechanism of cancer progression) at the molecular level. To obtain useful information from such gene expression measurements, it is essential to analyze the data with appropriate techniques. However the high-dimensionality of the data can bring about some problems such as curse of dimensionality and singularity problem of matrix computation, and hence makes it difficult to apply conventional data analysis methods. Therefore, the development of method which can effectively treat the data becomes a challenging issue in the field of computational biology. This research focuses on the gene selection and classification for cancer subtype discrimination based on gene expression (microarray) data.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.20
• /
• pp.8651-8656
• /
• 2014

Purpose: To study the expression of angiogenin-2 (Ang-2) and its receptor Tie-2 in colorectal cancer and discuss the possible mechanisms behind this process. Materials and Methods: Using the streptavidin-peroxidase (SP) immunohistochemical method, paraffin sections from 100 colorectal cancer samples and 10 samples from tumor-adjacent normal tissue (> 2 cm from the edge of the gross tumor) were tested for protein expression of Ang-2, Tie-2, PI3K, and AKT. Reverse transcription-polymerase chain reaction and Western blots were further used to measure expression of the 4 genes and proteins in 20 freshly-resected colorectal cancer samples and tumor-adjacent normal tissues. Results: In colorectal cancer tissues, the expression of the Ang-2, Tie-2, PI3K, and AKT genes and their proteins was significantly higher than in tumor-adjacent normal tissues. Protein expression in poorly-differentiated adenocarcinoma was higher than that in well and moderately differentiated adenocarcinoma. According to Duke's classification, the protein expression in Stages C and D was significantly higher than that in Stages A and B. In the group with lymphatic metastasis, the protein expression was higher than that without lymphatic metastasis. Conclusions: In colorectal cancer, the expression of the Ang-2, Tie-2, PI3K, and AKT genes and their proteins is markedly higher than those in tumor-adjacent normal tissues. No correlation was observed between protein expression and gender, location, or histologic type. Correlations did exist between protein expression and differentiation level, stage of Duke's classification, and lymphatic metastasis; in colorectal cancer tissues with lower differentiation levels, higher stages of Duke's classification, and lymphatic metastasis, the expression of all 4 proteins was higher. The study of their expression patterns and relationships with aggression and metastasis will provide a valuable experimental foundation for assessing prognosis and targeted therapy of colorectal cancer.

• Journal of Cancer Prevention
• /
• v.23 no.4
• /
• pp.183-190
• /
• 2018

Background: As the number of big-cohort studies increases, validation becomes increasingly more important. We aimed to validate administrative database categorized as colorectal cancer (CRC) by the International Classification of Disease (ICD) 10th code. Methods: Big-cohort was collected from Clinical Data Warehouse using ICD 10th codes from May 1, 2003 to November 30, 2016 at Seoul National University Bundang Hospital. The patients in the study group had been diagnosed with cancer and were recorded in the ICD 10th code of CRC by the National Health Insurance Service. Subjects with codes of inflammatory bowel disease or tuberculosis colitis were selected for the control group. For the accuracy of registered CRC codes (C18-21), the chart, imaging results, and pathologic findings were examined by two reviewers. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for CRC were calculated. Results: A total of 6,780 subjects with CRC and 1,899 control subjects were enrolled. Of these patients, 22 subjects did not have evidence of CRC by colonoscopy, computed tomography, magnetic resonance imaging, or positron emission tomography. The sensitivity and specificity of hospitalization data for identifying CRC were 100.00% and 98.86%, respectively. PPV and NPV were 99.68% and 100.00%, respectively. Conclusions: The big-cohort database using the ICD 10th code for CRC appears to be accurate.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.15
• /
• pp.6177-6180
• /
• 2014

Background: Our study objectives were to evaluate the medical economics of cervical cancer prevention and thereby contribute to cancer care policy decisions in Japan. Methods: Model creation: we created presence-absence models for prevention by designating human papillomavirus (HPV) vaccination for primary prevention of cervical cancer. Cost classification and cost estimates: we divided the costs of cancer care into seven categories (prevention, mass-screening, curative treatment, palliative care, indirect, non-medical, and psychosocial cost) and estimated costs for each model. Cost-benefit analyses: we performed cost-benefit analyses for Japan as a whole. Results: HPV vaccination was estimated to cost $291.5 million, cervical cancer screening $76.0 million and curative treatment $12.0 million. The loss due to death was $251.0 million and the net benefit was -$128.5 million (negative). Conclusion: Cervical cancer prevention was not found to be cost-effective in Japan. While few cost-benefit analyses have been reported in the field of cancer care, these would be essential for Japanese policy determination.