• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.381-384
• /
• 2010

심장질환을 예방하기 위하여 정기적인 검진을 통한 심장 운동기능 분석과 관찰이 중요하며, 심장 기능의 분석은 좌심실의 수동윤곽분할을 통하여 혈류량과 심박구출률 계산을 통해 이루어진다. 본 논문에서는 심장단축 자기공명영상에서 좌심실을 자동분할하기 위한 연구에 대하여 설명한다. 관측자의 간섭을 최소화하고 심장기능 분석을 자동화하기 위한 자동 초기점을 추출한 후에, 그래프 탐색을 통하여 복잡한 심장 구조와 다양한 촬영환경에 적용할 수 있는 좌심실 분할 알고리즘을 제안한다. 실험 결과에 따르면 자동 초기점 추출 알고리즘의 성능은 86.8%로 나타났고, 진행 중인 그래프 탐색 알고리즘도 유용한 결과를 나타내고 있다.

• Journal of KIISE:Software and Applications
• /
• v.33 no.4
• /
• pp.418-426
• /
• 2006

This paper proposes a robust and efficient segmentation method for a cardiac ultrasound image taken from a probe inserted into the heart in surgery. The method consists of three steps: initial boundary extraction, whole boundary modification using confidence competition, and local boundary modification using the rolling spoke method. Firstly, the initial boundary is extracted with threshold regions along the global spokes emitted from the center of an ultrasound probe. Secondly, high confidence boundary edges are detected along the global spokes by competing among initial boundary candidate and new candidates achieved by edge and appearance information. finally, the boundary is modified by rolling local spokes along concave regions that are difficult to extract using the global spokes. The proposed method produces promising segmentation results for the ultrasound cardiac images acquired during surgery.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.11a
• /
• pp.561-564
• /
• 2010

의학과 기술 발달로 인해 질병과 사고에 의한 사망률은 줄어들었으나, 심장 관련 질환에 의한 사망률은 증가하였다. 심장 질환을 예방하는 데는 정기적인 검진을 통해 심장기능을 분석하고 관찰하는 것이 중요하다. 심장 기능의 분석은 이완기와 수축기 사이의 혈류량 및 심박구출률 계산을 통한 심장 운동능력 평가에 의해 이루어진다. 본 연구에서는 심장 단축 자기공명영상에서 좌심실 영역을 자동 분할하여 혈류량 및 심박 구출률을 계산하는 알고리즘을 제안한다. K평균 군집화 기법을 적용하여 좌심실을 분할하고, 그래프 탐색 기법에 기반하여 분할 오류를 수정하였다. 15명의 지원자에 대해 제안하는 알고리즘을 사용하여 혈류량과 심박구출률을 계산하였고, 수동윤곽검출 및 General Electronics 사의 MASS 소프트웨어와 비교하였다. 제안한 알고리즘의 수동윤곽검출과 차이는 혈류량의 경우 평균적으로 이완기에 $4.6mL{\pm}3.9$, 수축기에 $2.1mL{\pm}2.4$로 나타났고, 심박구출률은 $1.8%{\pm}1.7$이었다. 전반적으로 MASS소프트웨어에 비해 좋은 성능을 나타내었다.

• Journal of Chest Surgery
• /
• v.29 no.3
• /
• pp.255-262
• /
• 1996

The donor pool for heart transplants is severely limited and there is still a legal problem of brain death. This study assessed the function of hearts "absolute anoxic" for ten minutes after asphyxia by perfusing the hearts on a Langendorfr apparatus for 45 minutes with Krebs-Henseleit buffier at 37 t at 80 cm H2O. Forty isolated rat hearts were divided into four groups. Ten control hearts (group 1) were perfused on the circuit without intervening ischemia. Ten hearts (group 2) were harvested, quickly flushed with 5cc of cold University of Wisconsin solution, and stored in the same cold solution for 4 hours. Ten hearts (group 3) were excised, quickly flushed with 5 u of cold Stanford cardioplegic solution and stored in cold saline solution for 4 hours. Ten asphyxiated hearts (group 4) had warm ischemia for ten minutes and were perfused with 5u of cold Stanford cardioplegia containing 7,500 units of urokinase to dissolve intravascular clots, and stored in cold saline solution for 1.5 hours. Time of spontaneous defibrillation (TSD) after perfusion was significantly longer in group 2, group 3 and group 4 than in group 1. TSD in group 3 and group 4 was significantly longer in comparison to that of group 2. Left ventricular developed pressure(LVDP) at 15 minutes was significantly lower in group 3 and group 4 than in group 1 and group 2. In group 4, LVDP at 30 minutes and 45 minutes was significantly lower compared with that in group 1 . In conclusion, asphyxiated rat hear;ts which had absolute anoxia for 10 minutes after as hyxia showed relatively satisfactory cardiac function. function.

• Journal of KIISE:Software and Applications
• /
• v.36 no.10
• /
• pp.825-835
• /
• 2009

The advance in medical technology has decreased death rates from diseases such as tubercle, pneumonia, malnutrition, and hepatitis. However, death rates from cardiac diseases are still increasing. To prevent cardiac diseases and quantify cardiac function, magnetic resonance imaging not harmful to the body is used for calculating blood volumes and ejection fraction(EF) on routine clinics. In this paper, automatic left ventricle(LV) segmentation is presented to segment LV and calculate blood volume and EF, which can replace labor intensive and time consuming manual contouring. Radial threshold determination is designed to segment LV and blood volume and EF are calculated. Especially, basal slices which were difficult to segment in previous researches are segmented automatically almost without user intervention. On short axis cardiac MRI of 36 subjects, the presented algorithm is compared with manual contouring and General Electronic MASS software. The results show that the presented algorithm performs in similar to the manual contouring and outperforms the MASS software in accuracy.

• 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.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.2
• /
• pp.560-569
• /
• 2016

Purpose The purpose of this study was not only to explore the factors associated with the survival of OHCA(Out-of-hospital Cardiac Arrest), but to provide ideas for improving the operation of emergency medical system in Korea. Method 90,734 OHCAs(Out-of-hospital Cardiac Arrest) with a cardiac etiology, who had been transported by 119 EMS ambulances for seven years from 2006 to 2012 in Korea, were analyzed. The data had a multilevel structure in that patient's survival in the same region is interrelated, so two-level (patient-region) logistic regression analysis was applied to adjust this correlation. Results The adjusted OR in group who were given CPR(Cardiopulmonary Resuscitation) by a bystander were 1.40 for survival to discharge. In addition, the adjusted OR in the group with an implementation of AED (automated external defibrillator) before arriving in hospital was 2.98 for survival to discharge. we categorized some continuous variables (number of emergency physician, OHCAs volume fo hospital, area deprivation level) into five quintiles. The adjusted OR in the number of emergency physician compared with Q1(lowest) was 1.29(Q2), 2.89(Q3), 3.39(Q4), 4.07(Q5), respectively. the adjusted OR in OHCAs volume of each hospital compared with Q1(lowest) was 2.06(Q2), 3.06(Q3), 3.46(Q4), 4.36(Q5), respectively. Lastly, the adjusted OR in deprivation level compared with Q1(least deprived area) was 0.72(Q4), 0.64(Q5) so that the adjusted OR of survival to discharge tended to decrease in more deprived districts. Conclusion The survival to discharge was better significantly in group given CPR by a bystander and with the implementation of AED before arriving in hospital. The survival to discharge tended to be significantly better in hospitals with a larger number of emergency physicians and higher volume of OHCAs in less deprived districts.

• The Monthly Diabetes
• /
• s.256
• /
• pp.22-24
• /
• 2011

동맥경화가 심장의 관상동맥에 일어나면, 혈관이 좁아져 피가 잘 흐르지 못하고, 심장근육에 산소부족이 일어나 가슴이 조여 들고 무거운 듯한 통증을 느끼게 된다. 평소에 느끼지 못했던 통증이 가슴 혹은 복부 등에 나타나면 병원을 찾는 것이 좋겠고 특히 통증이 15분 이상 진행될 때는 가까운 응급실에서 곧바로 심장을 검사하는 것이 심근경색 등에 의한 치사율 감소에 큰 도움이 된다.

• Journal of Chest Surgery
• /
• v.31 no.3
• /
• pp.255-261
• /
• 1998

A clinical evaluation was performed on 11 cases of the cardiac rupture by blunt chest trauma at the department of thoracic and cardiovascular surgery, Yeungnam University Hospital during the period from July, 1993 to May, 1995. The results were as follows ; The sex distribution was 8 mem and one women, and mean age was 41 years old. The causes of accident was traffic accident in most cases. And then one case was cultivator accident and another one was fall down. The average times from trauma to admission was 139 minutes and the patients that transferred via other hospitals have relatively long average times to 227 minutes. The average times from admission to operation was 117 minutes and we consumed 25 minutes for the preoperative preparation. The sites of injury were 7 cases in the right heart and 3 cases in the left heart. There were symptoms and signs of the cardiac tamponade(dyspnea, chest pain, nausea/vomitus, neck vein distention & hypovolemic shock) at admission and in most of them typical symptoms and signs of a tamponade appeared. Surgical approach was performed with median sternotomy in 10 cases. Subxiphoid pericardial window was created in one case. Another case which was very difficult in surgical procedure was operated under cardiopulmonary bypass and the result was good.

• 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.