• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.6
• /
• pp.35-43
• /
• 2009

In the Computer Aided Diagnosis(CAD) System, the efficient way of classifying nodules from chest CT images of a patient is to perform the classification of the remaining part after the pulmonary vessel extraction. During the pulmonary vessel extraction, due to the small difference between the vessel and nodule features in imaging studies such as CT scans after having an injection of contrast, nodule maybe extracted along with the pulmonary vessel. Therefore, the pulmonary vessel extraction method plays an important role in the nodule classification process. In this paper, we propose a nodule reclassification method based on vessel thickness analysis. The proposed method consist of four steps, lung region searching step, vessel extraction and thinning step, vessel topology formation and correction step and the reclassification of nodule in the vessel candidate step. The radiologists helped us to compare the accuracy of the CAD system using the proposed method and the accuracy of general one. Experimental results show that the proposed method can extract pulmonary vessels and reclassify false-positive nodules accurately.

• Journal of the Korean Society of Radiology
• /
• v.13 no.2
• /
• pp.201-207
• /
• 2019

This study has evaluated whether the method of using the combination of different risk group, according to K-TIRADS classification and K-TIRADS classification in thyroid ultrasonography is useful in a differential diagnosis of benign and malignant nodules. The subject was patients underwent thyroid ultrasonography and retrospective analysis were performed based on the results of fine needle aspiration cytology. A chi-square test was performed for the difference analysis of the score system in K-TIRADS and different risk group according to the benign and malignant of thyroid nodule. The optimized cut off value was determined by the K-TIRADS score and different risk group to predict malignant nodule through ROC curve analysis. In the differential verification result of K-TIRADS and different risk group, according to the classification of benign and malignant nodule group each showed significant difference statistically(p=.001). In the point classification according to K-TIRADS for the prediction of benign and malignant in ROC curve analysis showed AUC 0.786, Cut-off value>2(p=.001), and in the different risk group, it was decided as AUC 0.640, Cut-off value>2(p=.001). When discovering the nodule in thyroid ultrasound, it is considered that the K-TIRADAS which helps in identifying benign and malignant thyroid nodules, it is considered to be helpful in the differential diagnosis of thyroid nodules, than the classification system according to Different risk group, and when applying the classification system according to K-TIRADS, it is considered that it can reduce unnecessary fine needle aspiration cytology and could be helpful in finding the malignant nodules early.

• Journal of the Korea Computer Graphics Society
• /
• v.24 no.5
• /
• pp.31-39
• /
• 2018

In this paper, we propose an automated method for the ground-glass nodule(GGN) classification using GGN-Net based on intensity, texture, and shape-enhanced images in chest CT images. First, we propose the utilization of image that enhances the intensity, texture, and shape information so that the input image includes the presence and size information of the solid component in GGN. Second, we propose GGN-Net which integrates and trains feature maps obtained from various input images through multiple convolution modules on the internal network. To evaluate the classification accuracy of the proposed method, we used 90 pure GGNs, 38 part-solid GGNs less than 5mm with solid component, and 23 part-solid GGNs larger than 5mm with solid component. To evaluate the effect of input image, various input image set is composed and classification results were compared. The results showed that the proposed method using the composition of intensity, texture and shape-enhanced images showed the best result with 82.75% accuracy.

• Journal of the Korean Society of Radiology
• /
• v.18 no.4
• /
• pp.317-325
• /
• 2024

This study aimed to classify normal and nodule images in thyroid ultrasound images using GLCM and machine learning. The research was conducted on 600 patients who visited S Hospital in Busan and were diagnosed with thyroid nodules using thyroid ultrasound. In the thyroid ultrasound images, the ROI was set to a size of 50x50 pixels, and 21 parameters and 4 angles were used with GLCM to analyze the normal thyroid patterns and thyroid nodule patterns. The analyzed data was used to distinguish between normal and nodule diagnostic results using the SVM model and KNN model in MATLAB. As a result, the accuracy of the thyroid nodule classification rate was 94% for SVM model and 91% for the KNN model. Both models showed an accuracy of over 90%, indicating that the classification rate is excellent when using machine learning for the classification of normal thyroid and thyroid nodules. In the ROC curve, the ROC curve for the SVM model was generally higher compared to the KNN model, indicating that the SVM model has higher within-sample performance than the KNN model. Based on these results, the SVM model showed high accuracy in diagnosing thyroid nodules. This result can be used as basic data for future research as an auxiliary tool for medical diagnosis and is expected to contribute to the qualitative improvement of medical services through machine learning technology.

• Journal of the Korean Society of Radiology
• /
• v.84 no.1
• /
• pp.92-109
• /
• 2023

In patients with thyroid nodules, ultrasonography (US) has been established as a primary diagnostic imaging method and is essential for treatment decision. The Korean Thyroid Imaging Reporting and Data System (K-TIRADS) is a pattern-based, US malignancy risk stratification system that can easily diagnose nodules during real-time ultrasound examinations. The 2021 K-TIRADS clarified the US criteria for nodule classification and revised the size thresholds for nodule biopsy, thereby reducing unnecessary biopsies for benign nodules while maintaining the appropriate sensitivity to detect malignant tumors in patients without feature of high risk thyroid cancer. Thyroid radiology practice has an important clinical role in the diagnosis and non-surgical treatment of patients with thyroid nodules, and should be performed according to standard practice guidelines for proper and effective clinical care.

• Proceedings of the KOR-BRONCHOESO Conference
• /
• 1987.05a
• /
• pp.23.1-23
• /
• 1987

1985년 1월부터 1986년 12월까지 전북의대 부속병원 이비인후과에 입원하여 수술치료를 받은 34예의 갑상선 결절을 가진 환자를 대상으로 임상분석을 실시하여 다음과 같은 결과를 얻었다. 1) 34례중 32례(94.1%)가 양성, 그리고 2례(5.9%)가 악성 결절이었다. 2) 성비는 33 : 1로 여성에서 월등히 많았다. 3) 20-40대가 25명(73.5%)였고, 40대가 11명(32.4%)로 가장 많았다. 4) 이병기간은 3개월 이내가 14명(41.1%)으로 가장 많았다. 5) 임상증상에서 결절 촉진 34례(100%), 피로감 6례(18%) 심계항진 5례(15%)순이었다. 6) 발생부위에서는 우엽 21례(62%), 좌엽 10례(29%), 양엽 2례(6%) isthmus 1례(3%)순이었다. 7) $I^{131}$섭취검사에서 73.9%가 정상범위였으며, 갑상선주사소견에서는 cold결절이 91.3%였다. 8) 병리조직학적 분류를 보면 양성에서는 adenema 24례(75%), adenomatous goiter 5례(16%), cyst 3례(9%)순이었고 악성 2례는 papillary earcinoma 였다. 9) 수술방법은 일측성 편엽절제술이 22례(64.7%)로 가장 많았다. 10) 수술후 합병증은 경도의 출혈이 5례(14.7%)였고, 다음이 일시적 사성 3례(8.8%)이었다.

• Pediatric Gastroenterology, Hepatology & Nutrition
• /
• v.3 no.2
• /
• pp.151-159
• /
• 2000

Purpose: Recently, a wide application of gastrofiberscopy in the pediatric group have revealed that nodular duodenitis is not an uncommon disease in children and is suspected to be associated with H. pylori infection. The aim of this retrospective study was to investigate the clinical and histopathologic features in children with nodular duodenitis, and to assess the correlations beween both. Methods: During a period of 5 years (Jan. 1995~Dec. 1999), we investigated clinical, endoscopic and histopathologic features of 39 consecutive patients diagnosed as having nodular duodenitis at Pediatric department of Seoul Red Cross Hospital. In 35 children with nodular duodenitis endoscopic biopsy specimens were stained with Hematoxylin & Eosin and Giemsa's stain, and were graded according to the criteria outlined by Triadafilopoulos, Whitehead et al., and Prieto et al.. Statistical analyses were performed with Graph PAD InStat. Results: The prevalence rate of nodular duodenitis was 17.1% and the most frequent chief complaint was abdominal pain (69.2%). Endoscopically grade 1 was the most common (45.7%) and nodular gastritis was coexistent in 28.3%. The most common histology of the duodenum was grade 2 (54.3%), and the most common histologic score of the stomach was 2 (42.9%). H. pylori was found in the duodenum in 37.1%, and in the stomach in 31.4%. The correlation coefficient between the endoscopic grade and the histologic grade of nodular duodenitis was 0.3983 (p=0.0178). And the correlation coefficient between the histologic grade and the grade of H. pylori colonization in the duodenum was 0.5154 (p=0.0018). Conclusion: There was significant correlation between the endoscopic grade and the histologic grade of nodular duodenitis, and was also significant correlation between the histologic grade and the grade of H. pylori colonization in the duodenum. Therfore H. pylori infection should be regarded as an etiologic factor of nodular duodenitis.

• Journal of Intelligence and Information Systems
• /
• v.11 no.1
• /
• pp.17-33
• /
• 2005

This paper presents the proposed a classifier of liver cirrhotic step using MR(magnetic resonance) imaging and hierarchical neural network. The data sets for classification of each stage, which were normal, 1type, 2type and 3type, were analysis in the number of data was 231. We extracted liver region and nodule region from T1-weight MR liver image. Then objective interpretation classifier of liver cirrhotic steps. Liver cirrhosis classifier implemented using hierarchical neural network which gray-level analysis and texture feature descriptors to distinguish normal liver and 3 types of liver cirrhosis. Then proposed Neural network classifier learned through error back-propagation algorithm. A classifying result shows that recognition rate of normal is $100\%$, 1type is $82.8\%$, 2type is $87.1\%$, 3type is $84.2\%$. The recognition ratio very high, when compared between the result of obtained quantified data to that of doctors decision data and neural network classifier value. If enough data is offered and other parameter is considered this paper according to we expected that neural network as well as human experts and could be useful as clinical decision support tool for liver cirrhosis patients.

• Proceedings of the KSLP Conference
• /
• 1997.11a
• /
• pp.263-263
• /
• 1997

성대결절, 폴립, 부종 등은 성대의 남용이나 과용등의 성대손상이 그 공통된 주된 원인으로 거론되고 있다. 하지만 음성치료를 비롯한 보존적 치료에 대한 반응이 서로 상이하며, H＆E 염색을 이용한 병리조직학적인 감별이 곤란하여 진단에 혼돈이 있으며, 치료의 방침을 결정하거나 예후를 예측함에 있어서도 어려움이 있다. 양성성대질환은 기저막부 위와 세포외 간질에 주된 변화가 발생함이 알려져 있고, collagen type IV의 발현양상이 성대결절과 폴립에서 서로 다름에 대하여는 보고된 바 있으나 기타 점막하층의 골격유지를 주기능으로 하는 대표적 세포외간질인 collagen subtype에 대하여는 아직 보고된 바가 없는 실정이다. Collagen 발현의 차이를 연구하는 것은 상기질환의 병인을 이해하고 질환분류의 guideline을 제시하며 나아가 적절한 치료방범을 제시하는 데에 큰 의미가 있을 것으로 기대된다. Paraffin에 고정되어 있는 5례 이상씩의 성대결절과 성대폴립, 육아 종 및 라인케씨 부종 조직을 collagen type I부터 VII에 대하여 peroxidase kit를 사용하여 염색한 후 각 군간에 collagen 분포양상과 발현정도에 차이가 있는가 비교하였다.

• Journal of Digital Convergence
• /
• v.13 no.9
• /
• pp.489-495
• /
• 2015

Image availability evaluated by the degree of agreement and sensitive using the process improve visualization according to the Algorithm modification in Image Post-Processing. Reliability measured by the Breast Imaging Reporting and Data System. 172 patients visit same period divided by BI-RADS, category five stages, and contents of breast parenchyma into Calcification, Nodule and Mass. Evaluated the TE/PV image reliability, visualization sensitive, agreement of diagnosis. Convergence analysis was an in various fields. According to the result of this research, PV has higher sensitive and accuracy about lesions than TE visual and there is a difference insensitive by contents of breast parenchyma. Therefore, practical use of Algorithm Modification(Tissue Equalization: TE, Premium View: PV) is expected to improve more accurate, useful diagnosis, which has not been easy until now.