• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2022.06a
• /
• pp.301-301
• /
• 2022

본 연구는 운영 조건이 달라짐에 따라 회전수가 변하는 기기의 정상적 가동 여부와 고장 종류를 판별하기 위한 인공지능 알고리즘의 적용을 다루고 있다. 회전수가 변하는 장비로부터 계측된 상태 모니터링 센서의 신호는 비정상(non-stationary)적 특성이 있으므로, 상태 신호의 한계치가 고장 판별의 기준이 되기 어렵다는 점을 해결하고자 하였다. 정상 가동 여부는 이상 감지에 효율적인 오토인코더 및 기계학습 알고리즘을 적용하였으며, 고장 종류 판별에는 기계학습법과 합성곱 기반의 심층학습 방법을 적용하였다. 변하는 회전수와 연계된 주파수의 비정상적 시계열도 적절한 고장 특징 (Feature)로 대변될 수 있도록 시간 및 주파수 영역에서 특징 벡터를 구성할 수 있음을 예제로 설명하였다. 차원 축소 및 카이 제곱 기법을 적용하여 최적의 특징 벡터를 추출하여 기계학습의 분류 알고리즘이 비정상적 회전 신호를 가진 장비의 고장 예측에 활용될 수 있음을 보였다. 이 과정에서 k-NN(k-Nearest Neighbor), SVM(Support Vector Machine), Random Forest의 기계학습 알고리즘을 적용하였다. 또한 시계열 기반의 오토인코더 및 CNN (Convolution Neural Network) 적용하여 이상 감지와 고장진단을 수행한 결과를 비교하여 제시하였다.

• Proceedings of the Korean Nutrition Society Conference
• /
• 2000.06a
• /
• pp.127-127
• /
• 2000

• Magazine of the Korea Institute for Structural Maintenance and Inspection
• /
• v.18 no.3
• /
• pp.62-67
• /
• 2014

• Clinical and Experimental Pediatrics
• /
• v.50 no.1
• /
• pp.47-51
• /
• 2007

Purpose : The early diagnosis of Kawasaki disease (KD) is difficult sometimes, especially in atypical or incomplete cases presenting 4 or less principal clinical features without coronary arterial lesions (coronary arterial ectasia or aneurysm). The authors investigated the incidence of echocardiographic abnormalities in patients with KD to discover whether abnormal echocardiographic findings might be helpful in the early diagnosis of KD. Methods : Echocardiography was done in the acute stage of 103 patients with KD (Kawasaki group) and 40 patients with other acute febrile illnesses (control group). Abnormal echocardiographic findings were classified into 4 categories and defined as follows; 1) significant pericardial effusion, 2) significant valvular dysfunctions, 3) left ventricular systolic dysfunction, 4) coronary arterial ectasia or aneurysm. Results : In the Kawasaki group, significant pericardial effusion was present in 24 patients (23.3 percent), significant valvular dysfunctions in 30 patients (29.1 percent), left ventricular systolic dysfunction in 10 patients (9.7 percent), and coronary arterial lesions in 27 patients (26.2 percent). In the control group, significant pericardial effusion was present in only one patient (2.5 percent). The number of patients with any one of echocardiographic abnormalities was 57 (55.3 percent) in Kawasaki group and one (2.5 percent) in control group. The sensitivity of echocardiography in the diagnosis of KD was 55.3 percent, the specificity 97.5 percent, the positive predictive value 98.3 percent, and the negative predictive value 45.9 percent. Conclusion : If abnormal echocardiographic findings, even if other than coronary arterial lesions, are confirmed in patients in whom KD is suspected, it seems desirable to inifiate specific treatment for KD.

• The Journal of the Korea Contents Association
• /
• v.9 no.12
• /
• pp.339-347
• /
• 2009

Purpose : This study was to investigate the influence of heart rate and coronary calcification on diagnostic accuracy of 64-slice multidetector CT(MDCT) in coronary artery disease. Methods : 178 patients(84 men, 94 women) undergoing cardiac CT were included in this study. 3 coronary arteries(LAD, LCX, RCA) were assessed the presence of significant stenosis($\geq50%$) and the results compared with those of coronary angiography. Results : On a patient-based analysis, the diagnostic accuracy of 64-slice MDCT was 96.6%. The diagnostic accuracy on left anterior descending, left circumflex, right coronary artery were 86.5%, 84.3%, 92.1% respectively. Body mass index and blood pressure were not influenced on diagnostic accuracy of 64-slice MDCT. In less than 60/min of heart rate, accuracy was 90.1% and $\kappa$ value was 0.78. While in more than 70/min of heart rate, accuracy was 75.8% and $\kappa$ value was 0.52. In less than 100 of coronary calcification, accuracy was 91.3% and $\kappa$ value was 0.81. While in more than 400, accuracy was 68.6% and $\kappa$ value was 0.33. Conclusion : 64-slice MDCT shows similar diagnostic accuracy as coronary angiography. But in the context of more than 70/min of heart rate and 400 of coronary calcification, diagnostic accuracy was decreased. So there needs to identify heart rate and coronary calcification in cardiac CT, and if heart rate shows more than 70/min, use beta-blocker to regulate it.

• Journal of KSNVE
• /
• v.25 no.1
• /
• pp.16-21
• /
• 2015

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.05a
• /
• pp.280-280
• /
• 2004

기계의 소음 및 진동은 환경요인과 함께 기계의 정밀도 향상 및 고장진단과 관련하여 가장 중요한 분야중 하나이다. 특히 기계의 구동원으로 가장 많이 사용되고 모터는 기계설비이면서도 전기적인 현상으로 작동하기 때문에 설비보전의 사각지대로 나타나고 있다. 또한, 진동으로 모터의 전기적인 결함을 분석하기에는 미흡한 부분이 나타나기 때문에 모터 전용분석 기법을 사용하여 전동기의 전기적 결함을 분석하면 보다 정확한 진단이 가능하다. 현재 일부 생산 현장에서는 모터를 진단하기 위해 진동, 온도, 전류 등의 측정기법 들을 이용하여 분석하고 있다.(중략)

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10c
• /
• pp.199-201
• /
• 2002

본 논문에서는 심장질환의 진단 알고리즘의 개발에 있어서 필수적으로 요구되는 심장질환별 ECG 데이터의 수집에 관하여 기술한다. 또한, 진단 알고리즘을 개발하기 위한 전단계로서 심전도 신호에서 각 특징들을 검출하는 알고리즘에 관하여 설명하고, 이를 MITDB와 수집한 ECG 신호에 적용한 결과를 보인다. QRS-complex의 검출은 99% 이상의 정확도를 보이나, P-wave와 T-wave의 검출에서는 아직까지 보완할 점이 많은 것으로 나타난다. 심장질환별 12-채널 ECG 데이터베이스의 구축은 보다 정확하고 현실적인 진단 알고리즘을 개발하는 데 크게 기여할 것으로 기대한다.

• Database Research
• /
• v.34 no.3
• /
• pp.58-68
• /
• 2018

Distributed computing helps to efficiently store and process large data on a cluster of multiple machines. The performance of distributed computing is greatly influenced depending on the state of the servers constituting the distributed system. In this paper, we propose a self-diagnosis system that collects log data in a distributed system, detects anomalies and visualizes the results in real time. First, we divide the self-diagnosis process into five stages: collecting, delivering, analyzing, storing, and visualizing stages. Next, we design a real-time self-diagnosis system that meets the goals of real-time, scalability, and high availability. The proposed system is based on Apache Flume, Apache Kafka, and Apache Storm, which are representative real-time distributed techniques. In addition, we use simple but effective moving average and 3-sigma based anomaly detection technique to minimize the delay of log data processing during the self-diagnosis process. Through the results of this paper, we can construct a distributed real-time self-diagnosis solution that can diagnose server status in real time in a complicated distributed system.

• Journal of Yeungnam Medical Science
• /
• v.14 no.2
• /
• pp.393-398
• /
• 1997

An accurate early diagnosis of congenital midgut malrotation is essential to prevention of catastrophic effects of volvulus. To evaluate the usefulness of radiologic examinations in diagnosing intestinal malrotation, we retrospectively analyzed radiologic findings and operation records of 17 intestinal malrotation patients, who were radiologically diagnosed. The age range of the patients studied were from 1day to 12years. The presenting symptoms were vomiting, vomiting with abdominal pain, abdominal distention, diarrhea and failure to thrive. The viewpoints of this analysis were the location of duodeno-jejunal flexure on barium meal and cecal location on barium enema. Sixteen of 17 patients, who were radiologically diagnosed, were surgically proven, but one patient with annular pancreas was false positive. In the case of 3 surgically proved patients, malrotation was suspected on barium meal prior to the barium enema, but final diagnosis was determined on barium enema examination. We concluded that a barium enema should be performed on all children with suspected malrotation where the initial upper gastro-intestinal study was normal or suspicious on account of the small incidence of false positive and false negative barium meals.