• Journal of Chest Surgery
• /
• v.22 no.5
• /
• pp.829-833
• /
• 1989

Congenital coronary arteriovenous fistula is relatively uncommon, but with widespread use of cardiac catheterization, angiography and selective coronary arteriography are being recognized with increasing frequency. Recently we experienced one case of congenital coronary artery fistula which originated from the right coronary artery. The case was 25 year-old-male, who complained of dyspnea on exertion and continuous murmur was heard, and diagnosed as right coronary artery fistula by the cardiac catheterization and aortography. On the operation field, the right coronary artery was markedly dilated from aorta to the middle segment at acute margin of the right ventricle, which the hen-egg sized aneurysm was noticed. The dilated coronary ostium and fistular site were obliterated with several horizontal mattress sutures. And the dilated tortuous right coronary artery with aneurysm was excised. Postoperative course was uneventful and discharged without problem.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.508-510
• /
• 2019

본 연구에서는 딥 러닝을 이용하여 완전 자동화된 다중 클래스 전체 심장 분할 알고리즘을 제안하였다. 제안된 방법은 recurrent convolutional block과 residual multi-dilated block을 삽입하여 기존 U-Net을 개선한 인공신경망 모델을 사용하였다. 평가는 자동화 분석 결과와 수동 평가를 비교하였다. 그 결과 96.88%의 평균 DSC, 95.60%의 정확도, 97.00%의 recall을 얻었다. 이 실험 결과는 제안된 방법이 다양한 심장 구조에서 효과적으로 구분되어 수행되었음을 알 수 있다. 본 연구에서 제안된 알고리즘이 의사와 방사선 의사가 영상을 판독하거나 임상 결정을 내리는데 보조적 역할을 할 것을 기대한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.25-28
• /
• 2020

Automatic segmentation of brain tissues such as WM, GM, and CSF from brain MRI scans is helpful for the diagnosis of many neurological disorders. Accurate segmentation of these brain structures is a very challenging task due to low tissue contrast, bias filed, and partial volume effects. With the aim to improve brain MRI segmentation accuracy, we propose an end-to-end convolutional based U-SegNet architecture designed with multi-scale kernels, which includes cascaded dilated convolutions for the task of brain MRI segmentation. The multi-scale convolution kernels are designed to extract abundant semantic features and capture context information at different scales. Further, the cascaded dilated convolution scheme helps to alleviate the vanishing gradient problem in the proposed model. Experimental outcomes indicate that the proposed architecture is superior to the traditional deep-learning methods such as Segnet, U-net, and U-Segnet and achieves high performance with an average DSC of 93% and 86% of JI value for brain MRI segmentation.

• Honam Mathematical Journal
• /
• v.32 no.1
• /
• pp.101-112
• /
• 2010

Using some interesting convolution, we find kernels recovering the given function f. By a slight change of this convolution, we obtain an identity filter related to the Fourier series in the discrete time domain. We also introduce some techniques to decompose an impulse into several dilated pieces in the discrete domain. The detail examples deal with specific constructions of those decompositions. Also we obtain localized moving averages from a decomposition of an impulse to make hybrid Bollinger bands, that might give various strategies for stock traders.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.2
• /
• pp.37-44
• /
• 2020

In this paper, we proposed that a fully automatic multi-class whole heart segmentation algorithm using deep learning. The proposed method is based on U-Net architecture which consist of recurrent convolutional block, residual multi-dilated convolutional block. The evaluation was accomplished by comparing automated analysis results of the test dataset to the manual assessment. We obtained the average DSC of 96.88%, precision of 95.60%, and recall of 97.00% with CT images. We were able to observe and analyze after visualizing segmented images using three-dimensional volume rendering method. Our experiment results show that proposed method effectively performed to segment in various heart structures. We expected that our method can help doctors and radiologist to make image reading and clinical decision.

• Applied Microscopy
• /
• v.26 no.2
• /
• pp.207-219
• /
• 1996

This study was designed to investigate the ultrastructural alterations of the hepatocyte and bile canaliculus of the fasted mice with transmission and scanning electron microscopes. The morphometry was also carried out for the caliber of the bile canaliculus and the number, length and thickness of the microvillus. The hepatocyte observed in the three day fasting group showed ultrastructural images of active function. The dilated bile canaliculi, especially of type II were increased in number as compared with those seen in the normal group. However, the hepatocyte observed in the six day fasting group showed ultrastructural images of inactive function. The bile canaliculi without dilation (type I) were increased in number. The number of microvilli were identical with one another among the different types of bile canaliculi, while their length and thickness were reduced in the dilated bile canaliculi. From the evidence, the luminal size of the bile canaliculi seems to be easily changeable according to the functional state of the hepatocyte. However, the microvilli may not be changed in number but may be changed length and thickness when the bile canaliculi are dilated.

• Journal of Broadcast Engineering
• /
• v.23 no.5
• /
• pp.669-681
• /
• 2018

CNN (Convolution Neural Network) is one of the most important techniques to identify the kind of objects in the captured pictures. Whereas the conventional models have been used for low resolution images, the technique to recognize the high resolution images becomes crucial in the field of artificial intelligence. In this paper, we proposed an efficient CNN model based on dilated convolution and thresholding techniques to increase the recognition ratio and to decrease the computational complexity. The simulation results show that the proposed algorithm outperforms the conventional method and the thresholding technique enhances the performance of the proposed model.

• Advances in pediatric surgery
• /
• v.1 no.1
• /
• pp.27-32
• /
• 1995

Jejunal and ileal atresias are the most common cause of congenital intestinal obstruction and accounts for about 1/3 of all cases of intestinal obstruction in newborns. Despite the relative frequency of this anomaly, its survival rate was less than 10% up to 1950, more recently the survival rate has risen rapidly to 90% with the introduction of modern surgical techniques and the use of total parenteral nutrition. In 1969 Thomas described a tapering jejunoplasty to manage the discrepancy in the size of the proximal dilated lumen & contracted distal lumen, and to preserve absorptive surface when the dilated jejunum involved a long length, and Grosfeld et al.(1979) facilitated this method by using GIA staplers. Author have also used GIA stapler to resect the antimesenteric portion of the dilated proximal bowel in 8 cases of jejunoileal atresias with good results. The following results were obtained ; 1. There we 3 jejunal atresias & 5 ileal atresias, and male to female sex ratio was 5 : 3. 2. The type of atresia was as follows ; type IIIa was 3 cases, type IIIb was 4 cases, type IIIb+IV was 1 case. 3. In non-complication cases(5 cases), the mean hospital day was 16 days, and oral feeding was feasible from 6.2 days after operation. 4. The complications(anastomotic leakage, pneumonia) were frequently occurred in type IIIb cases and in low birth weight cases(75%). 5. Mortality rate was 25% including DAMA(discharge against medical advice) discharge case.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.40 no.5
• /
• pp.240-245
• /
• 2014

Angiokeratoma is a benign cutaneous lesion of the capillaries, presenting as dilated vessels in the upper part of the dermis. Although this disorder is classified into various types and has been occasionally reported in the skin of the scrotum or extremities, the involvement of the oral cavity mucosa has been rarely reported. The present study reports a case of angiokeratoma circumscriptum in the buccal mucosa. The expression of vascular endothelial growth factor (VEGF) and both of its receptors (VEGFR-1 and VEGFR-2) was demonstrated by immunohistochemistry in the endothelial cells lining the dilated vessels. The expression of VEGFR-2 was higher than that of VEGFR-1 in the endothelial cells in the lesion, indicating an increased rate of endothelial cell proliferation within the lesion. Interestingly, some of the endothelial cells co-expressed VEGF and its two receptors. These results suggest that endothelial cells in the pathologically dilated vessels possess VEGF autocrine growth activity involved in vasculogenesis and maintenance in angiokeratoma lesions. To our knowledge, this is the second report published on isolated oral angiokeratoma confined to the buccal mucosa and the first case report on angiokeratoma circumscriptum involving the buccal mucosa.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.2_1
• /
• pp.217-223
• /
• 2023

With the rapid advancement of technologies, need for different research fields where this technology can be used is also increasing. One of the most researched topic in computer vision is object detection, which has widely been implemented in various fields which include healthcare, video surveillance and education. The main goal of object detection is to identify and categorize all the objects in a target environment. Specifically, methods of object detection consist of a variety of significant techniq ues, such as image processing and patterns recognition. Anomaly detection is a part of object detection, anomalies can be found various scenarios for example crowded places such as subway stations. An abnormal event can be assumed as a variation from the conventional scene. Since the abnormal event does not occur frequently, the distribution of normal and abnormal events is thoroughly imbalanced. In terms of public safety, abnormal events should be avoided and therefore immediate action need to be taken. When abnormal events occur in certain places, real time detection is required to prevent and protect the safety of the people. To solve the above problems, we propose a modified YOLOv5 object detection algorithm by implementing dilated convolutional layers which achieved 97% mAP50 compared to other five different models of YOLOv5. In addition to this, we also created a simple mobile application to avail the abnormal event detection on mobile phones.