• 한국방사선학회논문지
• /
• 제9권6호
• /
• pp.369-374
• /
• 2015

본 연구에서 제안된 질감특징분석 알고리즘은 뇌출혈환자의 CT영상을 이용하여 정상영상과 질환영상으로 구분하여, 고유영상 및 실험영상을 생성하고 제안된 컴퓨터보조진단 시스템에 적용하여 6개의 파라메타로 정량적 분석을 통해 뇌출혈 CT영상의 인식률을 도출하고 평가하였다. 결과로 뇌출혈 CT영상 40증례 중에서 각각의 질감 특징값에 대한 인식률은 평균밝기의 경우 100%, 평균대조도의 경우 100%, 평탄도의 경우 100%, 왜곡도의 경우 100%로 높게 나타났고, 균일도의 경우 95%, 엔트로피의 경우 87.5%로 다소 낮은 질환 인식률을 보였다. 따라서 본 연구의 결과를 바탕으로 의료영상의 컴퓨터보조진단 시스템으로 발전된 프로그램을 구현한다면 뇌출혈 CT영상의 질환부위 자동검출 및 정량적 진단이 가능해 컴퓨터보조진단 자료로서 활용이 가능할 것으로 판단되며 최종판독에서 정확성과 판독시간 단축에 유용하게 사용 될 것으로 사료된다.

• 콘크리트학회논문집
• /
• 제20권4호
• /
• pp.513-522
• /
• 2008

섬유복합재료의 우수한 인장 성능은 섬유가 매트릭스의 균열 면에서 가교작용을 함으로써 발현되기 때문에 섬유의 분포 특성이 복합재료의 성능에 결정적인 영향을 미치게 된다. 그러나 PVA 섬유를 보강 섬유로 사용하는 섬유복합재료의 경우 PVA 섬유와 매트릭스 사이의 낮은 명암비와 PVA의 비전도성 특징으로 인하여 섬유의 위치 및 분포특성을 정량적으로 평가히는 방법은 연구가 미흡한 실정이다. 이 연구에서는 PVA 섬유를 보강 섬유로 사용하는 섬유복합재료의 섬유 분포 특성 등을 평가할 때 가장 중요한 과정인 섬유의 검출에 대하여 검출 성능을 향상 시킬 수 있는 알고리즘을 제시하였다. 제안한 알고리즘은 형광 현미경을 사용하여 얻은 섬유 이미지를 유형별로 분류하고, 분류된 분류된 섬유 이미지의 특성에 따라 분수령 알고리즘 (watershed algorithm)과 형태학적 재구성 (morphological reconstruction)을 이용하여 보다 정확히 섬유를 검출하는 과정으로 구성된다. 이 과정에서 섬유 이미지를 총 5가지 유형으로 분류하였으며, 인공신경회로망(ANN)을 분류기로 활용하기 위하여 형상 특성을 나타내는 5가지 특징값 즉, $F_s$, $F_c$, $F_p$, $F_l$과 $F_{rl}$을 추출하였다. 추출된 특징값에 대한 데이터베이스를 구축하여 ANN을 학습하여 분류기를 구축함으로써 섬유의 유형을 자동으로 분류할 수 있도록 하였다. 또한 5가지 섬유 이미지 유형 중에서 잘못 검출된 섬유이미지를 분수령 알고리즘과 형태학적 재구성을 통하여 섬유를 정확히 검출할 수 있는 기법을 제안하였다.

• 지질공학
• /
• 제18권4호
• /
• pp.537-543
• /
• 2008

지표 근접 지질조사 및 매설물 조사에 응용되는 GPR탐사를 통해 석조문화재 기초지반을 효과적으로 탐지하기 위해 토양층과 자갈층이 교호하는 지질공학적인 모형지반(천부 지층구조)을 3 m의 깊이로 재건하여 GPR탐사 및 탄성파 굴절법 탐사를 측선별로 수행하였다. GPR탐사에 사용되는 주파수를 다양하게 적용하여 주파수별 분해능에 따른 모형지반의 경계구분을 확인하였다. 그 결과 100 MHz의 주파수 대역에서 분해능은 낮지만 가장 효과적으로 지층경계를 구분할 수 있었으며, 225 MHz, 450 MHz, 900 MHz의 고주파수 대역을 사용한 결과에서는 전자파의 감쇠로 인해 하부경계를 구분할 수 없었으나 천부의 분해능은 향상되었다. 이로부터 심부에 대한 100 MHz의 자료와 천부에 대한 고주파수 단면도를 조합하여 최종단면도를 얻었다. 탄성파 굴절법 탐사에서는 모형지반의 정확한 경계 및 지층 간 경계면은 파악할 수 없었으나, 모형지반이 위치하고 있는 구간에서의 등속도선이 주변 지역에 비해 상대적으로 깊은 심도까지 분포하는 것으로 볼 때 원지반과 모형 경계의 속도 대비를 파악할 수 있는 것으로 나타났다.

• 대한의용생체공학회:의공학회지
• /
• 제43권5호
• /
• pp.341-352
• /
• 2022

Medical image segmentation is the most important task in radiation therapy. Especially, when segmenting medical images, the liver is one of the most difficult organs to segment because it has various shapes and is close to other organs. Therefore, automatic segmentation of the liver in computed tomography (CT) images is a difficult task. Since tumors also have low contrast in surrounding tissues, and the shape, location, size, and number of tumors vary from patient to patient, accurate tumor segmentation takes a long time. In this study, we propose a method algorithm for automatically segmenting the liver and tumor for this purpose. As an advantage of setting the boundaries of the tumor, the liver and tumor were automatically segmented from the CT image using the 2D CoordConv DeepLab V3+ model using the CoordConv layer. For tumors, only cropped liver images were used to improve accuracy. Additionally, to increase the segmentation accuracy, augmentation, preprocess, loss function, and hyperparameter were used to find optimal values. We compared the CoordConv DeepLab v3+ model using the CoordConv layer and the DeepLab V3+ model without the CoordConv layer to determine whether they affected the segmentation accuracy. The data sets used included 131 hepatic tumor segmentation (LiTS) challenge data sets (100 train sets, 16 validation sets, and 15 test sets). Additional learned data were tested using 15 clinical data from Seoul St. Mary's Hospital. The evaluation was compared with the study results learned with a two-dimensional deep learning-based model. Dice values without the CoordConv layer achieved 0.965 ± 0.01 for liver segmentation and 0.925 ± 0.04 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.927 ± 0.02 for liver division and 0.903 ± 0.05 for tumor division. The dice values using the CoordConv layer achieved 0.989 ± 0.02 for liver segmentation and 0.937 ± 0.07 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.944 ± 0.02 for liver division and 0.916 ± 0.18 for tumor division. The use of CoordConv layers improves the segmentation accuracy. The highest of the most recently published values were 0.960 and 0.749 for liver and tumor division, respectively. However, better performance was achieved with 0.989 and 0.937 results for liver and tumor, which would have been used with the algorithm proposed in this study. The algorithm proposed in this study can play a useful role in treatment planning by improving contouring accuracy and reducing time when segmentation evaluation of liver and tumor is performed. And accurate identification of liver anatomy in medical imaging applications, such as surgical planning, as well as radiotherapy, which can leverage the findings of this study, can help clinical evaluation of the risks and benefits of liver intervention.

• Korean Journal of Radiology
• /
• 제22권3호
• /
• pp.344-353
• /
• 2021

Objective: The mitotic count of gastrointestinal stromal tumors (GIST) is closely associated with the risk of planting and metastasis. The purpose of this study was to develop a predictive model for the mitotic index of local primary GIST, based on deep learning algorithm. Materials and Methods: Abdominal contrast-enhanced CT images of 148 pathologically confirmed GIST cases were retrospectively collected for the development of a deep learning classification algorithm. The areas of GIST masses on the CT images were retrospectively labelled by an experienced radiologist. The postoperative pathological mitotic count was considered as the gold standard (high mitotic count, > 5/50 high-power fields [HPFs]; low mitotic count, ≤ 5/50 HPFs). A binary classification model was trained on the basis of the VGG16 convolutional neural network, using the CT images with the training set (n = 108), validation set (n = 20), and the test set (n = 20). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated at both, the image level and the patient level. The receiver operating characteristic curves were generated on the basis of the model prediction results and the area under curves (AUCs) were calculated. The risk categories of the tumors were predicted according to the Armed Forces Institute of Pathology criteria. Results: At the image level, the classification prediction results of the mitotic counts in the test cohort were as follows: sensitivity 85.7% (95% confidence interval [CI]: 0.834-0.877), specificity 67.5% (95% CI: 0.636-0.712), PPV 82.1% (95% CI: 0.797-0.843), NPV 73.0% (95% CI: 0.691-0.766), and AUC 0.771 (95% CI: 0.750-0.791). At the patient level, the classification prediction results in the test cohort were as follows: sensitivity 90.0% (95% CI: 0.541-0.995), specificity 70.0% (95% CI: 0.354-0.919), PPV 75.0% (95% CI: 0.428-0.933), NPV 87.5% (95% CI: 0.467-0.993), and AUC 0.800 (95% CI: 0.563-0.943). Conclusion: We developed and preliminarily verified the GIST mitotic count binary prediction model, based on the VGG convolutional neural network. The model displayed a good predictive performance.

• 한국의학물리학회지:의학물리
• /
• 제26권1호
• /
• pp.28-35
• /
• 2015

본 논문에서는 반복적 구성 기법과 관전류 노출자동조절 기법이 영상의 화질과 방사선량에 미치는 영향을 관상동맥 전산화단층촬영 혈관조영 영상(coronary computed tomography angiography, CCTA)을 대상으로 팬텀 실험에 기반하여 평가하고자 한다. 이를 위하여 미국 의학물리학회(American Association of Physics in Medicine) 표준의 성능 평가 팬텀을 320 다중검출열 CT로써 촬영하였다. 80 kVp, 100 kVp, 120 kVp의 관전압에 있어서, 관전류 노출자동조절 기법은 저선량 목표 표준편차(SD=44)와 고선량(목표 표준편차=33)의 두 가지 설정으로써 촬영하였다. 재구성 변수로서는 필터보정 역투영(FBP)와 반복적 재구성 방법을 설정하여, 전부 12개의 재구성 영상을 획득하였다(12=3 (80, 100, 120 kVp)${\times}2$ (저선량(목표SD=44), 고선량(목표SD=33))${\times}2$ (필터보정역투영, 반복적 재구성). 영상의 화질은 잡음의 세기(표준편차), 변조전달함수, 대조대잡음비(CNR)에 의하여 평가하였으며, 관전압과 관전류 노출자동조절 기법에서의 목표 선량과 대소 및 재구성 기법의 선택이 화질과 방사선량에 미치는 영향을 관찰하였다. 반복적 재구성 기법은 필터보정역투영 기법보다 영상 잡음을 대폭 감소시켰으며 이는 저선량의 경우 더욱 뚜렷하였다. 즉, 잡음의 세기는 관전류 노출자동조절의 설정이 고선량 (목표SD=33)과 저선량(목표SD=44)인 경우, 각각 평균 38%와 평균 46% 감소하였다. 반복적 재구성 기법에 의하여, 변조전달 함수에 의한 공간적 해상도의 평가에 있어서 미약한 감소를 보였으나, 이로써 잡음 저감과 대조대잡음비(CNR)에 있어서의 현저한 개선을 상쇄할 정도의 영향에는 미치지 못 하였다. 결과적으로, 관상동맥 전산화단층촬영 혈관조영 영상의 획득에서 있어서, 반복적 재구성 기법과 관전류 노출자동조정 기법을 동시에 사용하는 것은 영상의 화질을 개선하면서 공간적 해상도의 저하 등 그 부작용은 최소화함으로써, 합리적으로 획득 가능한 한 최소한의 선량 (ALARA)의 원칙에 충실한 실제 임상적 효과를 의미한다고 기대할 수 있다.

• 대한원격탐사학회지
• /
• 제39권6_1호
• /
• pp.1353-1369
• /
• 2023

해상에서는 재난·재해 사고가 발생했을 시 바람 등에 의한 기상영향과 해류, 조류와 같은 해상영향에 의해 피해 규모가 달라지게 되며, 빠른 현장 파악을 통해 적합한 방제 방안을 세워 피해 규모를 최소화할 의무가 있다. 특히, 해상에 유출되는 오염물질 중 상대적으로 낮은 점도와 표면장력으로 인해 해수면에서 얇은 막으로 존재하는 오염물질은 육안으로 식별하기 어렵다. 따라서 본 연구에서는 현장에서 쉽게 활용 가능한 촬영장비를 활용하여 RGB 이미지에서 해수면의 부유성 오염물질을 탐지하는 알고리즘을 개발하고, 실 해역에서 획득된 입력자료를 활용하여 알고리즘의 성능을 평가하고자 한다. 개발된 알고리즘은 영상 강화 기법을 활용하여 오염물질과 일반 해수면의 강도값 대비를 향상시키고, 히스토그램(Histogram) 분석을 통해 배경 임계값을 찾아 오염물질 이외의 부유물질을 제거하여 최종적으로 오염물질을 분류한다. 본 연구에서는 개발된 알고리즘의 성능평가를 위해서 대체물질을 이용한 실 해역 테스트를 수행하였으며, 대부분의 부유성 해양오염물질은 탐지되었으나 파도가 강한 곳에서는 오탐지 영역이 발생하였다. 그러나 기존 알고리즘에서 단일 임계값을 사용한 탐지 방법보다 약 3배 이상의 개선된 탐지 결과를 보여준다. 본 연구개발 결과를 통해 기존 현장에서 육안으로 식별이 어려웠던 부유성 해양오염물질을 탐지함으로써 현장에서의 방제 대응 활동에 유용하게 사용될 것으로 기대된다.

• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2008년도 International Meeting on Information Display
• /
• pp.993-994
• /
• 2008

Electrochromic (EC) devices are capable of reversibly changing their optical properties upon charge injection and extraction induced by the external voltage. The characteristics of the EC device, such as low power consumption, high coloration efficiency, and memory effects under open circuit status, make them suitable for use in a variety of applications including smart windows and electronic papers. Coloration due to reduction or oxidation of redox chromophores can be used for EC devices (e-paper), but the switching time is slow (second level). Recently, with increasing demand for the low cost, lightweight flat panel display with paper-like readability (electronic paper), an EC display technology based on dye-modified $TiO_2$ nanoparticle electrode was developed. A well known organic dye molecule, viologen, was adsorbed on the surface of a mesoporous $TiO_2$ nanoparticle film to form the EC electrode. On the other hand, ZnO is a wide bandgap II-VI semiconductor which has been applied in many fields such as UV lasers, field effect transistors and transparent conductors. The bandgap of the bulk ZnO is about 3.37 eV, which is close to that of the $TiO_2$ (3.4 eV). As a traditional transparent conductor, ZnO has excellent electron transport properties, even in ZnO nanoparticle films. In the past few years, one-dimension (1D) nanostructures of ZnO have attracted extensive research interest. In particular, 1D ZnO nanowires renders much better electron transportation capability by providing a direct conduction path for electron transport and greatly reducing the number of grain boundaries. These unique advantages make ZnO nanowires a promising matrix electrode for EC dye molecule loading. ZnO nanowires grow vertically from the substrate and form a dense array (Fig. 1). The ZnO nanowires show regular hexagonal cross section and the average diameter of the ZnO nanowires is about 100 nm. The cross-section image of the ZnO nanowires array (Fig. 1) indicates that the length of the ZnO nanowires is about $6\;{\mu}m$. From one on/off cycle of the ZnO EC cell (Fig. 2). We can see that, the switching time of a ZnO nanowire electrode EC cell with an active area of $1\;{\times}\;1\;cm^2$ is 170 ms and 142 ms for coloration and bleaching, respectively. The coloration and bleaching time is faster compared to the $TiO_2$ mesoporous EC devices with both coloration and bleaching time of about 250 ms for a device with an active area of $2.5\;cm^2$. With further optimization, it is possible that the response time can reach ten(s) of millisecond, i.e. capable of displaying video. Fig. 3 shows a prototype with two different transmittance states. It can be seen that good contrast was obtained. The retention was at least a few hours for these prototypes. Being an oxide, ZnO is oxidation resistant, i.e. it is more durable for field emission cathode. ZnO nanotetropods were also applied to realize the first prototype triode field emission device, making use of scattered surface-conduction electrons for field emission (Fig. 4). The device has a high efficiency (field emitted electron to total electron ratio) of about 60%. With this high efficiency, we were able to fabricate some prototype displays (Fig. 5 showing some alphanumerical symbols). ZnO tetrapods have four legs, which guarantees that there is one leg always pointing upward, even using screen printing method to fabricate the cathode.

• International Journal of Costume and Fashion
• /
• 제7권1호
• /
• pp.49-61
• /
• 2007

The subject of study concerns the color characteristics of clothing/accessories from �Chosun�a nd modern Korea. It particularly focuses on the colors of the �Jeogori? traditional jacket) which represents Korean traditional clothing. Color data were collected from 353 woman�s �Jegori�s from �Chosun�a nd modern Korea and divided into the predominant colors & sub colors, and analyzed with HV/C and PCCS. According to the analysis, the representative the predominant color of 'Chosun Jeogori' for woman is YR, Y and side one is R, YR. The p, dp, sf, ltg colors are frequently found and the freshness of most of them is medium or low which shows a calm-feeling. The p, and ltg tones are often found as a the predominant color and p, dp are also often found as a the side color. In the 'Chosun Jeogori' for woman, the predominant colors and the side colors spread in a similar frequency and most of them have medium or low freshness of the color that shows a calm-feeling. The side color is one of the R colors and it shows characteristics of traditional coloring such as �Jajoo Goreum? Among the �Jeogori�f or modern women, the R, YR, Y, and B colors are often used for the single-color �Jeogori? the Y, GY, N, G colors are for the colorful �Jeogori? The P, lt, and b tons of the color are often used for a single-color �Jeogori�a nd the p, lt, and W are for the colorful �Jeogori? For the colorful �Jeogori? the side color have a high freshness which shows that a strong image was used frequently as the dp, s, dk tons of the color were often used. According to an analysis of the coloring, the predominant color and the side color of �Chosun Jeogori�a re applied to create a harmony of analogy & contrast in one aspect of Moon & Spencer�s view. It is found that the predominant color and the side color are similar or contrasting in one view of the color harmony, except that the traditional the �Jeogori�i s colored by ideological and symbolic meanings. The predominant color and the side color of modern the �Jeogori�a re complementary colors which are applied to the indistinct relationship or contrasting harmony in one aspect of Moon & Spencer�s view of the color harmony. The characteristics of the color as above show that traditional the �Jeogori�r egards the side color as both the harmony of the colors in general and also of symbolic meaning, while modern the �Jeogori�f or woman has a tendency toward a cultural transition and is mixed with new colors focusing more on its the predominant color than on the general harmony and character of ideological & traditional coloring.

• Korean Journal of Radiology
• /
• 제20권4호
• /
• pp.621-630
• /
• 2019

Objective: To study the prevalence and clinical characteristics of decreased myocardial blood flow (MBF) quantified by dynamic computed tomography (CT) myocardial perfusion imaging (MPI) in symptomatic patients without in-stent restenosis. Materials and Methods: Thirty-seven (mean age, 71.3 ± 10 years; age range, 48-88 years; 31 males, 6 females) consecutive symptomatic patients with patent coronary stents and without obstructive de novo lesions were prospectively enrolled to undergo dynamic CT-MPI using a third-generation dual-source CT scanner. The shuttle-mode acquisition technique was used to image the complete left ventricle. A bolus of contrast media (50 mL; iopromide, 370 mg iodine/mL) was injected into the antecubital vein at a rate of 6 mL/s, followed by a 40-mL saline flush. The mean MBF value and other quantitative parameters were measured for each segment of both stented-vessel territories and reference territories. The MBF_ratio was defined as the ratio of the mean MBF value of the whole stent-vessel territory to that of the whole reference territory. An MBF_ratio of 0.85 was used as the cut-off value to distinguish hypoperfused from non-hypoperfused segments. Results: A total of 629 segments of 37 patients were ultimately included for analysis. The mean effective dose of dynamic CT-MPI was 3.1 ± 1.2 mSv (range, 1.7-6.3 mSv). The mean MBF of stent-vessel territories was decreased in 19 lesions and 81 segments. Compared to stent-vessel territories without hypoperfusion, the mean MBF and myocardial blood volume were markedly lower in hypoperfused stent-vessel territories (77.5 ± 16.6 mL/100 mL/min vs. 140.4 ± 24.1 mL/100 mL/min [p < 0.001] and 6.4 ± 3.7 mL/100 mL vs. 11.5 ± 4 mL/100 mL [p < 0.001, respectively]). Myocardial hypoperfusion in stentvessel territories was present in 48.6% (18/37) of patients. None of clinical parameters differed statistically significantly between hypoperfusion and non-hypoperfusion subgroups. Conclusion: Decreased MBF is commonly present in patients who are symptomatic after percutaneous coronary intervention, despite patent stents and can be detected by dynamic CT-MPI using a low radiation dose.