• Journal of Yeungnam Medical Science
• /
• 제3권1호
• /
• pp.179-183
• /
• 1986

영남대학교 영남의료원 해부병리과에 최근 3년간 의뢰된 동결 절편 809예를 검토한 결과, 임파절, 위장관계, 피부의 순으로 많았으며, 동결절편 진단의 정확도는 진단연기 예(0.9%)를 포함하면 98.1%였고 위음성율은 0.5%였으며 위양성은 1예도 없었다. 위음성의 예는 임파절, 피부, 난소, 갑상선 각 1예였고 부정확한 진단 명기 및 grading error가 0.5%였다.

• 치과방사선
• /
• 제21권1호
• /
• pp.73-81
• /
• 1991

저자는 악간절기능장애의 소견을 가진 환자 75명과 악관절기능장애의 병마 및 현증이 없는 정상인 75명을 표준화된 조건하에서 두부축방향방사선촬영을 하여 과두수평각과 initial 및 final depth of cut를 측정하고 통계학적으로 분석한 결과를 다음과 같이 요약할 수 있었다. 환자군과 정상군, 환측과 비환측에 있어 과두수평각과 initial 및 final depth of cut는 통계학적으로 유의한 차가 없었으므로 두부축방향방사선사진은 임상에서 악관절기능장애의 유무 및 환측과 비환측의 감별에 이용하기에 부적합한 방법이라 사료된다. 본 연구의 계측치는 편차가 심하므로 환자마다 과두수평각을 측정하여 이각에 맞추어 악관절단층촬영을 해야 한다고 사료된다.

• 한국의학물리학회지:의학물리
• /
• 제30권2호
• /
• pp.49-58
• /
• 2019

Purpose: Automated analytical systems have begun to emerge as a database system that enables the scanning of medical images to be performed on computers and the construction of big data. Deep-learning artificial intelligence (AI) architectures have been developed and applied to medical images, making high-precision diagnosis possible. Materials and Methods: For diagnosis, the medical images need to be labeled and standardized. After pre-processing the data and entering them into the deep-learning architecture, the final diagnosis results can be obtained quickly and accurately. To solve the problem of overfitting because of an insufficient amount of labeled data, data augmentation is performed through rotation, using left and right flips to artificially increase the amount of data. Because various deep-learning architectures have been developed and publicized over the past few years, the results of the diagnosis can be obtained by entering a medical image. Results: Classification and regression are performed by a supervised machine-learning method and clustering and generation are performed by an unsupervised machine-learning method. When the convolutional neural network (CNN) method is applied to the deep-learning layer, feature extraction can be used to classify diseases very efficiently and thus to diagnose various diseases. Conclusions: AI, using a deep-learning architecture, has expertise in medical image analysis of the nerves, retina, lungs, digital pathology, breast, heart, abdomen, and musculo-skeletal system.

• 한국통신학회논문지
• /
• 제34권10B호
• /
• pp.1062-1069
• /
• 2009

비만으로 인해 대사증후군을 많이 앓고 있는 상황에서 후유증으로 신장 질환이 커다란 사회문제가 되고 있는 실정이다. 따라서 자각증상이 없는 신장의 이상 유무를 조기에 판단하여 적절한 조치를 취하는 것이 무엇보다 중요하다. 이를 위해 본 논문에서는 음성 분석을 통해 신장 질환을 무자각, 무구속, 무통종의 방법으로 진단할 수 있는 방법을 제안하였다. 구성하고자 하는 전체 시스템은 크게 음성 분석과 얼굴색을 살피는 방법을 결합시키는 시스템이 개발되고 있으며 이 중 본 논문은 입술소리를 기반으로 신장 질환을 진단하는 방법에 설계하였다. 이를 위해 본 논문에서는 첫째, 신장 질환을 앓고 있는 환자와 정상인을 대상으로 피실험자 집단을 각각 구성하고 입술소리의 수치학적 분석을 실험으로 출력하고 그 결과값에 대한 비교 분석을 수행하였으며 둘째, 한의학적 청진 이론과 언어학, 음성학과의 상관성을 분석하고 이를 기반으로 음성에 대한 신장의 특징 요소를 추출하여 제1포먼트 주파수와의 연관성을 도출하였다. 실험 결과 신장 질환자 집단이 정상인 집단보다 제1포먼트 주파수 대역폭이 넓게 형성되는 결과를 추출하였으며 최종적으로 입술소리만으로 신장 질환을 진단할 때의 오진 확률에 대해 계산하였다.

• 대한기계학회논문집A
• /
• 제38권2호
• /
• pp.205-210
• /
• 2014

기계시스템의 결함을 진단하기 위한 방법으로 패턴인식 기법이 널리 사용되고 있다. 진동신호의 변화를 감지하여 기계시스템의 건전성을 판단하는 방법이 패턴인식 기법이다. 대표적 패턴 인식기법으로 최근 은닉 마르코프 모델과 인공신경망이 여러 분야에서 사용되고 있다. 본 연구에서는 결함진단에 은닉 마르코프 모델과 인공신경망을 혼합한 방법이 제시되었으며 결함진단 대상 구조물로는 크랙을 가진 회전하는 풍력터빈 블레이드가 선정되었다. 본 연구에서는 크랙발생 여부뿐만 아니라 그 위치 및 크기도 동시에 진단하고자 하였다. 아울러서 본 연구에서는 일정 주파수들을 갖는 모멘트를 대상 구조물에 가함으로써 외부 잡음에도 불구하고 높은 결함진단 확률을 가질 수 있도록 하였다.

• 대한임상검사과학회지
• /
• 제43권1호
• /
• pp.1-5
• /
• 2011

The infection rate of syphilis is still increasing in the world especially in developing countries and the infection is often seen in large amounts of clinical specimens. For the diagnosis of this disease, Rapid Plasma Reagin (RPR)/Venereal Disease Research Laboratory (VDRL) has still been used as one of major primary methods to diagnose syphilis even though the test readings are somewhat subjective with high false positive rates. Recently, the automatic ARCHITECT Syphilis TP, which is based on the detection of the TP-specific antibodies, has been introduced in many laboratories. Therefore, the clinical assessment of the method is needed to provide primary diagnosis of syphilis at the moment. We evaluated 3 different manual rapid kits and ARCHITECT Syphilis TP comparing with RPR/FTA-ABS and analysed their diagnostic properties. From February 2006 to April 2008, 203 positive and 250 negative specimens, obtained from Chungbuk National University Hospital were used for the evaluation. In the evaluation between manual rapid kits, their specificities were as high as 99.2 ~ 99.6% while their sensitivities were observed with little differences; 98.0% (199/203) for Kit A, 96.6% (196/203) for Kit B, and 97.4% (197/203) for Kit S. In the case of ARCHITECT Syphilis TP test, it showed 100% specificity (250/250) and 98.5% sensitivity (249/250). Kappa values comparing with RPR/FTA-ABS were 0.978 for Kit A, 0.964 for Kit B and Kit S, and 0.987 for ARCHITECT Syphilis TP. From our evaluation, we found out that manual rapid tests and ARCHITECT Syphilis TP have very good clinical accuracies and high kappa agreements with RPR/FTA-ABS. Due to its automation and quick simultaneous diagnosis with another serological markers, we suggest that the ARCHITECT Syphilis TP is one of best suitable method for the primary diagnosis of syphilis and that it might be able to replace RPR method in the laboratories.

• KNOM Review
• /
• 제24권1호
• /
• pp.13-19
• /
• 2021

최근 산업 분야에서는 공장 자동화 뿐만 아니라 장애 진단/예측을 통해 고장/사고를 사전에 방지하여 생산량을 극대화하기 위한 연구가 진행되고 있으며, 이를 구성하기 위해 많은 양의 데이터 축적을 위한 클라우드 기술, 데이터 처리를 위한 빅 데이터 기술, 그리고 데이터 분석을 쉽게 진행하기 위한 AI(Artificial Intelligence)기술이 도입되고 있다. 또한 최근에는 장애 진단/예측의 발전으로 인해 설비 유지보수(PM: Productive Maintenance) 방식도 정기적으로 설비를 유지보수 하는 방식인 TBM(Time Based Maintenance)에서 설비 상태에 따라 유지보수 하는 방식인 CBM(Condition Based Maintenance)을 조합하는 방식으로 발전하고 있다. CBM 기반 유지보수를 수행하기 위하여 설비의 상태(condition)의 정의와 분석이 필요하다. 따라서 본 논문에서는 머신 러닝(Machine Learning) 기반의 장애 진단을 위한 시스템 및 데이터 모델(Data Model)을 제안하며, 이를 기반으로 장애를 사전 예측한 사례를 제시하고자 한다.

• Imaging Science in Dentistry
• /
• 제53권1호
• /
• pp.61-67
• /
• 2023

Purpose: The aim of this study was to objectively detect simulated tooth ankylosis using a novel method involving cone-beam computed tomography (CBCT). Materials and Methods: Tooth ankylosis was simulated in single-rooted human permanent teeth, and CBCT scans were acquired at different current levels (5, 6.3, and 8 mA) and voxel sizes (0.08, 0.125, and 0.2). In axial reconstructions, a line of interest was perpendicularly placed over the periodontal ligament space of 21 ankylosed and 21 non-ankylosed regions, and the CBCT grey values of all voxels along the line of interest were plotted against their corresponding X-coordinates through a line graph to generate a profile. The image contrast was increased by 30% and 60% and the profile assessment was repeated. The internal area of the resulting parabolas was obtained from all images and compared between ankylosed and non-ankylosed regions under different contrast enhancement conditions, voxel sizes, and mA levels using multi-way analysis of variance with the Tukey post hoc test(α=0.05). Results: The internal area of the parabolas of all non-ankylosed regions was significantly higher than that of the ankylosed regions(P<0.05). Contrast enhancement led to a significantly greater internal area of the parabolas of non-ankylosed regions (P<0.05). Overall, voxel size and mA did not significantly influence the internal area of the parabolas(P>0.05). Conclusion: The proposed novel method revealed a relevant degree of applicability in the detection of simulated tooth ankylosis; increased image contrast led to greater detectability.

• Journal of Microbiology
• /
• 제42권3호
• /
• pp.211-215
• /
• 2004

Toxoplasmosis has been well known as an important human infection to consider especially in pregnant women. Although many serologic methods are available, the diagnosis of toxoplasmosis can be extremely difficult. The presence of increased levels of Toxoplasma-specific IgG antibodies indicates an infection, but it does not differentiate between a recent and past infection. The purpose of our study was to compare the performance of the ELISA T. gondii IgG/IgM test, a widely used enzyme-linked immunosorbent assay, to the ELISA IgG avidity method. One hundred and four serum samples (from 38 males and 66 females) were tested and evaluated from symptomatic patients (chorioretinitis, lymphadenopathy), and from women in their first trimester of pregnancy who were suspected of having toxoplasmosis, The high IgG avidity and ELISA IgG antibody levels were in agreement for 51 of the specimens (49.0％). Thirty-eight discrepant (borderline) results from the IgG avidity method were positive for IgM (3 specimens) and IgG (37 specimens). Interestingly, out of the eight serum samples that were positive for both IgG and IgM antibodies, two samples were low IgG avidity, and three samples were borderline. There was no statistically significant relation observed between the results of the IgG avidity method and the ELISA IgG test, and the IgG avidity method and ELISA IgM test (X$^2$=1.987; p=0.370 and X$^2$=2.152; p=0.341, respectively). The IgG avidity method was considered easy to perform and an acceptable approach for the differentiation of discrepant results (recent/chronic) and for the current detection of T. gondii antibodies. We concluded that the determination of IgG avidity is a helpful tool for the diagnosis of the ocular form of toxoplasmosis and it is a safe method for screening this disease in the first trimester of pregnancy.

• 제어로봇시스템학회논문지
• /
• 제22권4호
• /
• pp.276-280
• /
• 2016

Due to the increasing interest in safety and consistent product quality over a past few decades, demand for effective quality monitoring and safe operation in the modern industry has propelled research into statistical based fault detection and diagnosis methods. This paper describes the application of Hotelling $T^2$ index based Principal Component Analysis (PCA) method for fault detection and diagnosis in industrial processes. Multivariate statistical process control techniques are now widely used for performance monitoring and fault detection. Conventional methods such as PCA are suitable only for steady state processes. These conventional projection methods causes false alarms or missing data for the systems with transient values of processes. These issues significantly compromise the reliability of the monitoring systems. In this paper, a reliable method is used to overcome false alarms occur due to varying process conditions and missing data problems in transient states. This monitoring method is implemented and validated experimentally along with matlab. Experimental results proved the credibility of this fault detection method for both the steady state and transient operations.