• Journal of Korean Neurosurgical Society
• /
• 제66권3호
• /
• pp.308-315
• /
• 2023

Recent updates in genomic-integrated glioma classification have caused confusion in current clinical practice, as management protocols and health insurance systems are based on evidence from previous diagnostic classifications. The Korean Brain Tumor Society conducted an electronic questionnaire for society members, asking for their ideas on risk group categorization and preferred treatment for each individual diagnosis listed in the new World Health Organization (WHO) classification of gliomas. Additionally, the current off-label drug use (OLDU) protocols for glioma management approved by the Health Insurance Review and Assessment Service (HIRA) in Korea were investigated. A total of 24 responses were collected from 20 major institutes in Korea. A consensus was reached on the dichotomic definition of risk groups for glioma prognosis, using age, performance status, and extent of resection. In selecting management protocols, there was general consistency in decisions according to the WHO grade and the risk group, regardless of the individual diagnosis. As of December 2022, there were 22 OLDU protocols available for the management of gliomas in Korea. The consensus and available options described in this report will be temporarily helpful until there is an accumulation of evidence for effective management under the new classification system for gliomas.

• Journal of Information Processing Systems
• /
• 제19권5호
• /
• pp.652-662
• /
• 2023

Identifying highly discriminating genes is a critical step in tumor recognition tasks based on microarray gene expression profile data and machine learning. Gene selection based on tree models has been the subject of several studies. However, these methods are based on a single-tree model, often not robust to ultra-highdimensional microarray datasets, resulting in the loss of useful information and unsatisfactory classification accuracy. Motivated by the limitations of single-tree-based gene selection, in this study, ensemble gene selection methods based on multiple-tree models were studied to improve the classification performance of tumor identification. Specifically, we selected the three most representative tree models: ID3, random forest, and gradient boosting decision tree. Each tree model selects top-n genes from the microarray dataset based on its intrinsic mechanism. Subsequently, three ensemble gene selection methods were investigated, namely multipletree model intersection, multiple-tree module union, and multiple-tree module cross-union, were investigated. Experimental results on five benchmark public microarray gene expression datasets proved that the multiple tree module union is significantly superior to gene selection based on a single tree model and other competitive gene selection methods in classification accuracy.

• Journal of Yeungnam Medical Science
• /
• 제28권1호
• /
• pp.1-12
• /
• 2011

Primary liver carcinomas have classified classified into hepatocellular carcinoma, cholangiocarcinoma, and combined hepatocellular-cholangiocarcinoma (CHC). CHC is a tumor containing unequivocal, intimately mixed elements of both hepatocellular carcinoma and cholangiocarcinoma. It forms a small but significant proportion of primary liver carcinomas. The origin and pathogenesis of CHC have not been well established. According to the 2010 WHO classification, CHCs are categorized into 2 groups: the classical type and a subtype with stem cell features. This review describes recent progress in pathology and classification of CHC.

• Korean Journal of Radiology
• /
• 제25권1호
• /
• pp.103-112
• /
• 2024

Objective: To investigate the association of ultrasound (US) features of follicular thyroid carcinoma (FTC) with tumor invasiveness and prognosis based on the World Health Organization (WHO) classification and telomerase reverse transcriptase (TERT) promoter mutations. Materials and Methods: This retrospective study included 54 surgically confirmed FTC patients with US images and TERT promoter mutations (41 females and 13 males; median age [interquartile range], 40 years [30-51 years]). The WHO classification consisted of minimally invasive (MI), encapsulated angioinvasive (EA), and widely invasive (WI) FTCs. Alternative classifications included Group 1 (MI-FTC and EA-FTC with wild type TERT), Group 2 (WI-FTC with wild type TERT), and Group 3 (EA-FTC and WI-FTC with mutant TERT). Each nodule was categorized according to the US patterns of the Korean Thyroid Imaging Reporting and Data System (K-TIRADS) and American College of Radiology-TIRADS (ACR-TIRADS). The Jonckheere-Terpstra and Cochran-Armitage tests were used for statistical analysis. Results: Among 54 patients, 29 (53.7%) had MI-FTC, 16 (29.6%) had EA-FTC, and nine (16.7%) had WI-FTC. In both the classifications, lobulation, irregular margins, and final assessment categories showed significant differences (all Ps ≤ 0.04). Furthermore, the incidences of lobulation, irregular margin, and high suspicion category tended to increase with increasing tumor invasiveness and worse prognosis (all Ps for trend ≤ 0.006). In the WHO groups, hypoechogenicity differed significantly among the groups (P = 0.01) and tended to increase in proportion as tumor invasiveness increased (P for trend = 0.02). In the alternative group, punctate echogenic foci were associated with prognosis (P = 0.03, P for trend = 0.03). Conclusion: Increasing tumor invasiveness and worsening prognosis in FTC based on the WHO classification and TERT promoter mutation results were positively correlated with US features that indicate malignant probability according to both K-TIRADS and ACR-TIRADS.

• 대한방사선기술학회지:방사선기술과학
• /
• 제42권1호
• /
• pp.9-17
• /
• 2019

Although digital mammography is a representative method for breast cancer detection. It has a limitation in detecting and classifying breast tumor due to superimposed structures. Machine learning, which is a part of artificial intelligence fields, is a method for analysing a large amount of data using complex algorithms, recognizing patterns and making prediction. In this study, we proposed a technique to improve the diagnostic accuracy of energy-selective mammography by training data using the machine learning algorithm and using dual-energy measurements. A dual-energy images obtained from a photon-counting detector were used for the input data of machine learning algorithms, and we analyzed the accuracy of predicted tumor thickness for verifying the machine learning algorithms. The results showed that the classification accuracy of tumor thickness was above 95% and was improved with an increase of imput data. Therefore, we expect that the diagnostic accuracy of energy-selective mammography can be improved by using machine learning.

• BMB Reports
• /
• 제53권12호
• /
• pp.664-669
• /
• 2020

Breast cancer is one of the most frequently diagnosed cancers. Although biomarkers are continuously being discovered, few specific markers, rather than classification markers, representing the aggressiveness and invasiveness of breast cancer are known. In this study, we used samples from canine mammary tumors in a comparative approach. We subjected 36 fractions of both canine normal and mammary tumor plasmas to high-performance quantitative proteomics analysis. Among the identified proteins, LCAT was selectively expressed in mixed tumor samples. With further MRM and Western blot validation, we discovered that the LCAT protein is an indicator of aggressive mammary tumors, an advanced stage of cancer, possibly highly metastatic. Interestingly, we also found that LCAT is overexpressed in high-grade and lymph-node-positive breast cancer in silico data. We also demonstrated that LCAT is highly expressed in the sera of advanced-stage human breast cancers within the same classification. In conclusion, we identified a possible common plasma protein biomarker, LCAT, that is highly expressed in aggressive human breast cancer and canine mammary tumor.

• 융합신호처리학회논문지
• /
• 제20권2호
• /
• pp.70-77
• /
• 2019

조기 유방암을 진단하기 위해서는 유방초음파 판독이 매우 중요하다. 초음파 검사는 초음파장비에 따라 화질의 차이가 심하게 나타날 뿐만 아니라 검사자의 경험과 숙련 정도에 따라 진단의 차이가 크게 나타난다. 따라서 정확한 진단과 치료를 위하여 객관적인 판단기준이 필요하다. 이에 본 연구에서는 GLCM(Gray Level Co-occurrence Matrix) 알고리듬을 적용하여 질감 특성을 분석하고 특징파라미터들을 추출하여 신경망분류기를 이용하여 유방암을 진단하였다. 유방초음파 영상은 정상 조직과 양성, 악성 종양으로 분류하여 질감 특성 파라미터 6가지를 추출하였다. 유방초음파검사로 진단된 정상 영상, 악성 및 양성종양 영상 각각 14증례를 대상으로 추출된 6개의 파라미터들을 적용하여 다층 퍼셉트론 신경망구조 역전파 학습방법으로 학습을 시켰다. 학습된 모델에 정상 유방 영상 51증례, 양성종양 영상 62증례, 악성종양 영상 74증례의 영상을 사용하여 분류한 결과 95.2%의 분류율을 나타내었다.

• 대한임상검사과학회지
• /
• 제43권2호
• /
• pp.48-56
• /
• 2011

Gastrointestinal stromal tumor (GIST) is a mesenchymal tumor and is associated with a specific immunophenotype index. It is very important to identify the specific immunophenotype and the diagnosis for the treatment GIST patients. Ninety two cases of GIST analyzed in this study were immuno-stained for c-kit, DOG1, CD34, PKC-${\theta}$, PDGFR-${\alpha}$. The rate of positive staining and statistical significance were then compared. In addition, the GISTs were analyzed as followings: very low risk, low risk, intermediate risk and high risk according to tumor size and nuclear division, and later correlated with clinical parameters. The results of the GIST positive stainings were: DOG1 (95.7%), PKC-${\theta}$ (90.2%), PDGFR-${\alpha}$ (88.0%), c-kit (87.0%) and CD34 (71.7%). Only DOG1 staining showed a statistical significance of p<0.05. It was identified in the classification system of histologic risk that staining expression of DOG1, PKC-${\theta}$, PDGFR-${\alpha}$ were significantly increased as histologic risk increases (p<0.05). However, clinical parameters such as age and sex of patients have no correlations with the classification system of histologic risk (p>0.05). Therefore, in this study, the expression of DOG1 showed statistical significance and DOG1, PKC-${\theta}$, PDGFR-${\alpha}$ staining increased significantly as the histologic risk increases in histologic classification system. Taken together, the DOG1 staining should be very effective for the diagnosis of GIST patients.

• Journal of Korean Neurosurgical Society
• /
• 제64권4호
• /
• pp.608-618
• /
• 2021

Objective : The endoscopic endonasal transsphenoidal approach is a widely-used method for the surgical treatment of pituitary adenomas. We aimed to evaluate the results of endoscopic surgery by comparing preoperative classification methods and investigating their relationship with postoperative resection and remission rates and complications. Methods : We retrospectively reviewed the medical records of 236 patients (118 males) who underwent surgery for pituitary adenomas. Preoperative Knosp classification, tumor size (TS), suprasellar extension (SSE), postoperative resection and remission rates, and complications were evaluated. Results : The follow-up period was 3 months to 6 years. The patients' ages ranged between 16 and 84 years. Endocrinologically, 114 patients (48.3%) had functional adenoma (FA), and 122 patients (51.7%) had non-functional adenoma (NFA). Among the FA group, 92 (80.7%) showed remission. A statistically significant difference was found between patients with and without remission in terms of the Knosp, TS, and SSE classifications (p<0.01). Knosp, TS, and SSE classification grades were found to be correlated with the resection rates (p<0.01). Meningitis was seen in seven patients (3.0%), diabetes insipidus in 16 (6.9%; permanently in two [0.9%]), and rhinorrhea in 19 (8.1%). Thirty-six patients (15.3%) developed pituitary insufficiency and received hormone replacement therapy. Conclusion : The resection categories and remission rates of FAs were directly proportional to the adenoma sizes and Knosp grades, while the degree of suprasellar growth further complicated resection and remission rates. Adenoma sizes less than 2 cm and SSEs less than 1 cm are associated with favorable remission and resection rates.

• Communications for Statistical Applications and Methods
• /
• 제13권3호
• /
• pp.567-576
• /
• 2006

In this paper, we applied the multiblock dimension reduction methods to the classification of tumor based on microarray gene expressions data. This procedure involves clustering selected genes, multiblock dimension reduction and classification using linear discrimination analysis and quadratic discrimination analysis.