• 대한기관식도과학회지
• /
• 제8권2호
• /
• pp.31-35
• /
• 2002

Background and Objectives : Salivary gland neoplasms are unique because of their infrequency, inconsistent classification, and highly variable biologic behavior. The aim of this study is to analysis the histopathologic classification of salivary glnad neoplasm and to suggest a guideline of management. Materials and Methods : The medical records of 310 patients with salivary gland neoplasm who treated at Asan medical center between 1992 and 2001 were analyzed retrospectively. Among the 310 patients, 138 patients were male and 172 patients were female. Mean age was 50.5 years. Results : Benign salivary neoplasms were 213 cases. They consisted of 153 cases (71.8%) of parotid tumor, 41 cases (19.2%) of submandibular gland tumor and 19 cases (8.9%) of minor salivary gland tumor. Pleomorphic adenoma was the most common benign neoplasm. Malignant salivary neoplasms were 97 cases. They consisted of 45 cases (46.4%) of parotid tumor, 26 cases(26.8%) of minor salivary gland tumor, 24 cases(24.7%) of submandibular gland tumor and 2 cases(2.1%) of sublingual gland tumor. Adenoid cystic carcinoma was the most common malignant neoplasm. Conclusions : The most commonly involved gland was parotid (64%) and the most frequent tumor was pleomorphic adenoma (52%). Although the majority of minor salivary gland neoplasms are malignant, three of parotid tumors are benign.

• 한국멀티미디어학회논문지
• /
• 제24권12호
• /
• pp.1614-1623
• /
• 2021

In this paper, the purpose is evaluation of the effect of using fractal feature in machine learning based pancreatic tumor classification. We used the data that Pancreas CT series 469 case including 1995 slice of benign and 1772 slice of malignant. Feature selection is implemented from 109 feature to 7 feature by Lasso regularization. In Fractal feature, fractal dimension is obtained by box-counting method, and hurst coefficient is calculated range data of pixel value in ROI. As a result, there were significant differences in both benign and malignancies tumor. Additionally, we compared the classification performance between model without fractal feature and model with fractal feature by using support vector machine. The train model with fractal feature showed statistically significant performance in comparison with train model without fractal feature.

• 한국멀티미디어학회논문지
• /
• 제25권8호
• /
• pp.1233-1241
• /
• 2022

Tumors of the brain are the deadliest, with a life expectancy of only a few years for those with the most advanced forms. Diagnosing a brain tumor is critical to developing a treatment plan to help patients with the disease live longer. A misdiagnosis of brain tumors will lead to incorrect medical treatment, decreasing a patient's chance of survival. Radiologists classify brain tumors via biopsy, which takes a long time. As a result, the doctor will need an automatic classification system to identify brain tumors. Image classification is one application of the deep learning method in computer vision. One of the deep learning's most powerful algorithms is the convolutional neural network (CNN). This paper will introduce a novel deep learning structure and image gradient to classify brain tumors. Meningioma, glioma, and pituitary tumors are the three most popular forms of brain cancer represented in the Figshare dataset, which contains 3,064 T1-weighted brain images from 233 patients. According to the numerical results, our method is more accurate than other approaches.

• Biomolecules & Therapeutics
• /
• 제30권2호
• /
• pp.179-183
• /
• 2022

Traditionally, pathologists microscopically examine tissue sections to detect pathological lesions; the many slides that must be evaluated impose severe work burdens. Also, diagnostic accuracy varies by pathologist training and experience; better diagnostic tools are required. Given the rapid development of computer vision, automated deep learning is now used to classify microscopic images, including medical images. Here, we used a Inception-v3 deep learning model to detect mouse lung metastatic tumors via whole slide imaging (WSI); we cropped the images to 151 by 151 pixels. The images were divided into training (53.8%) and test (46.2%) sets (21,017 and 18,016 images, respectively). When images from lung tissue containing tumor tissues were evaluated, the model accuracy was 98.76%. When images from normal lung tissue were evaluated, the model accuracy ("no tumor") was 99.87%. Thus, the deep learning model distinguished metastatic lesions from normal lung tissue. Our approach will allow the rapid and accurate analysis of various tissues.

• Journal of Gastric Cancer
• /
• 제15권3호
• /
• pp.183-190
• /
• 2015

Purpose: The Lauren classification system is a very commonly used pathological classification system of gastric adenocarcinoma. A recent study proposed that the Lauren classification should be modified to include the anatomical location of the tumor. The resulting three types were found to differ significantly in terms of genomic expression profiles. This retrospective cohort study aimed to evaluate the clinical significance of the modified Lauren classification (MLC). Materials and Methods: A total of 677 consecutive patients who underwent curative gastrectomy from January 2005 to December 2007 for histologically confirmed gastric cancer were included. The patients were divided according to the MLC into proximal non-diffuse (PND), diffuse (D), and distal non-diffuse (DND) type. The groups were compared in terms of clinical features and overall survival. Multivariate analysis served to assess the association between MLC and prognosis. Results: Of the 677 patients, 48, 358, and 271 had PND, D, and DND, respectively. Their 5-year overall survival rates were 77.1%, 77.7%, and 90.4%. Compared to D and PND, DND was associated with significantly better overall survival (both P<0.01). Multivariate analysis showed that age, differentiation, lympho-vascular invasion, T and N stage, but not MLC, were independent prognostic factors for overall survival. Multivariate analysis of early gastric cancer patients showed that MLC was an independent prognostic factor for overall survival (odds ratio, 5.946; 95% confidence intervals, 1.524~23.197; P=0.010). Conclusions: MLC is prognostic for survival in patients with gastric adenocarcinoma, in early gastric cancer. DND was associated with an improved prognosis compared to PND or D.

• 한국컴퓨터정보학회논문지
• /
• 제26권7호
• /
• pp.37-44
• /
• 2021

뇌 MRI 영상의 자동 분류는 뇌종양의 조기 진단을 하는 데 있어 중요한 역할을 한다. 본 연구에서 우리는 심층 특징 앙상블을 사용한 MRI 영상에서의 딥 러닝 기반 뇌종양 분류 모델을 제안한다. 우선 사전 학습된 3개의 합성 곱 신경망을 사용하여 입력 MRI 영상에 대한 심층 특징들을 추출한다. 그 이후 추출된 심층 특징들은 완전 연결 계층들로 구성된 분류 모듈의 입력 값으로 들어간다. 분류 모듈에서는 우선 3개의 서로 다른 심층 특징들 각각에 대해 먼저 완전 연결 계층을 거쳐 특징 차원을 줄인다. 그 이후 3개의 차원이 준 특징들을 결합하여 하나의 특징 벡터를 생성한 뒤 다시 완전 연결 계층의 입력값으로 들어가서 최종적인 분류 결과를 예측한다. 우리가 제안한 모델을 평가하기 위해 웹상에 공개된 뇌 MRI 데이터 셋을 사용하였다. 실험 결과 우리가 제안한 모델이 다른 기계학습 기반 모델보다 더 좋은 성능을 나타냄을 확인하였다.

• Journal of Korean Neurosurgical Society
• /
• 제61권3호
• /
• pp.292-301
• /
• 2018

Medulloblastoma is the most common malignant brain tumor of childhood and remains a major cause of cancer related mortality in children. Significant scientific advancements have transformed the understanding of medulloblastoma, leading to the recognition of four distinct clinical and molecular subgroups, namely wingless (WNT), sonic hedgehog, group 3, and group 4. Subgroup classification combined with the recognition of subgroup specific molecular alterations has also led to major changes in risk stratification of medulloblastoma patients and these changes have begun to alter clinical trial design, in which the newly recognized subgroups are being incorporated as individualized treatment arms. Despite these recent advancements, identification of effective targeted therapies remains a challenge for several reasons. First, significant molecular heterogeneity exists within the four subgroups, meaning this classification system alone may not be sufficient to predict response to a particular therapy. Second, the majority of novel agents are currently tested at the time of recurrence, after which significant selective pressures have been exerted by radiation and chemotherapy. Recent studies demonstrate selection of tumor sub-clones that exhibit genetic divergence from the primary tumor, exist within metastatic and recurrent tumor populations. Therefore, tumor resampling at the time of recurrence may become necessary to accurately select patients for personalized therapy.

• Journal of Acupuncture Research
• /
• 제32권3호
• /
• pp.229-236
• /
• 2015

Objectives : The purpose of this study is to report the clinical effect of Korean medical treatment on a spinal cord tumor. Methods : We treated a patient who was diagnosed with a spinal cord tumor. We used acupuncture, bee venom pharmacopuncture, herbal medicine, moxibustion and physical therapy; the patient was evaluated using the visual analogue scale(VAS) and given an International Standards for Neurological Classification of Spinal Cord Injury(ISNCSCI) score. Results : VAS decreased and ISNCSCI score increased meaningfully. Conclusions : According to these results, this report possibly suggests that Korean medical treatment could be a helpful choice for treating a spinal cord tumor.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• 제38권4호
• /
• pp.212-220
• /
• 2012

Objectives: Fluorine-18 fluorodeoxyglucose positron emission tomography ($^{18}F$-FDG PET) is a non-invasive diagnostic tool for many human cancers wherein glucose uptake transporter-1 (GLUT-1) acts as a main transporter in the uptake of $^{18}F$-FDG in cancer cells. Increased expression of glucose transporter-1 has been reported in many human cancers. In this study, we investigated the correlation between $^{18}F$-FDG accumulation and expression of GLUT-1 in oral cancer. Materials and Methods: We evaluated 42 patients diagnosed with oral squamous cell carcinoma (OSCC) and malignant salivary gland tumor as confirmed by histology. 42 patients underwent pre-operative $^{18}F$-FDG PET, with the maximum standardized uptake value ($SUV_{max}$) measured in each case. Immunohistochemical staining was done for each histological specimen, and results were evaluated post-operatively according to the percentage (%) of positive area, intensity, and staining score. Results: For OSCC, $SUV_{max}$ significantly increased as T stage of tumor classification increased. For malignant salivary gland tumor, $SUV_{max}$ significantly increased as T stage of tumor classification increased. For OSCC, GLUT-1 was expressed in all 36 cases. GLUT-1 staining score (GSS) increased as T stage of tumor classification increased, with the difference statistically significant. For malignant salivary gland tumor, GLUT-1 expression was observed in all 6 cases; average GSS was significantly higher in patients with cervical lymph node metastasis than that in patients without cervical lymph node metastasis. Average GSS was higher in OSCC ($11.11{\pm}1.75$) than in malignant salivary gland tumor ($5.33{\pm}3.50$). No statistically significant correlation between GSS and $SUV_{max}$ was observed in OSCC or in malignant salivary gland tumor. Conclusion: We found no statistically significant correlation between GSS and $SUV_{max}$ in OSCC or in malignant salivary gland tumor. Studies on the various uses of GLUT during $^{18}F$-FDG uptake and SUV and GLUT as tumor prognosis factor need to be conducted through further investigation with large samples.

• 한국생물정보학회:학술대회논문집
• /
• 한국생물정보시스템생물학회 2005년도 BIOINFO 2005
• /
• pp.57-62
• /
• 2005

Microarray gene expression profiling technology is one of the most important research topics in clinical diagnosis of disease. Given thousands of genes, only a small number of them show strong correlation with a certain phenotype. To identify such an optimal subset from thousands of genes is intractable, which plays a crucial role when classify multiple-class genes express models from tumor samples. This paper proposes an efficient classifier design method to simultaneously select the most relevant genes using an intelligent genetic algorithm (IGA) and design an accurate classifier using Support Vector Machine (SVM). IGA with an intelligent crossover operation based on orthogonal experimental design can efficiently solve large-scale parameter optimization problems. Therefore, the parameters of SVM as well as the binary parameters for gene selection are all encoded in a chromosome to achieve simultaneous optimization of gene selection and the associated SVM for accurate tumor classification. The effectiveness of the proposed method IGA/SVM is evaluated using four benchmark datasets. It is shown by computer simulation that IGA/SVM performs better than the existing method in terms of classification accuracy.