• 한국전자파학회논문지
• /
• 제11권4호
• /
• pp.582-591
• /
• 2000

본 논문에서는 휴대전화기에 의한 인체 영향을 FDTD (시간영역 유한차분법) 해석할 수 있도록 한국인 표준 에 알맞는 인체 두부의 FDTD 모텔 제작 방법을 소개하였다. 한국인 표준에 알맞은 사람의 두부를 MRI 촬영한 다음.2차원 MRI 영상 데이터를 이용하여 2차원 segmentation을 하였다. segmentation은 반자동법을 적용하였 으며 제작된 2차원 se밍nentation 데이터를 토대로 $1mm\times1mm\times1mm$크기의 3차원 고해상도 segmentation 데이터를 제작하였다. 3차원 고해상도 segmentation 데이터를 이용하여 휴대전화기의 사용 상황에 어올리도록 다양한 각도로 기울인 인체 두부의 FDTD 모델을 제작하였다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2020년도 추계학술대회
• /
• 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.

• International Journal of Computer Science & Network Security
• /
• 제22권8호
• /
• pp.343-351
• /
• 2022

A brain tumor forms when some tissue becomes old or damaged but does not die when it must, preventing new tissue from being born. Manually finding such masses in the brain by analyzing MRI images is challenging and time-consuming for experts. In this study, our main objective is to detect the brain's tumorous part, allowing rapid diagnosis to treat the primary disease instantly. With image processing techniques and deep learning prediction algorithms, our research makes a system capable of finding a tumor in MRI images of a brain automatically and accurately. Our tumor segmentation adopts the U-Net deep learning segmentation on the standard MICCAI BRATS 2018 dataset, which has MRI images with different modalities. The proposed approach was evaluated and achieved Dice Coefficients of 0.9795, 0.9855, 0.9793, and 0.9950 across several test datasets. These results show that the proposed system achieves excellent segmentation of tumors in MRIs using deep learning techniques such as the U-Net algorithm.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.55-58
• /
• 2009

Automatic segmentation of brain MRI data usually leaves some segmentation errors behind that are to be subsequently removed interactively, using computer graphics tools. This interactive removal is normally performed by operating on individual 2D slices. It is very tedious and still leaves some segmentation errors which are not visible on the slices. We have proposed to perform a novel 3D interactive correction of brain segmentation errors introduced by the fully automatic segmentation algorithms. We have developed the tool which is based on 3D semi-automatic propagation algorithm. The paper describes the implementation principles of the proposed tool and illustrates its application.

• Korean Journal of Radiology
• /
• 제20권1호
• /
• pp.102-113
• /
• 2019

Objective: To assess the accuracy and potential bias of computed tomography (CT) ventricular volumetry using semiautomatic three-dimensional (3D) threshold-based segmentation in repaired tetralogy of Fallot, and to compare them to those of two-dimensional (2D) magnetic resonance imaging (MRI). Materials and Methods: This retrospective study evaluated 32 patients with repaired tetralogy of Fallot who had undergone both cardiac CT and MRI within 3 years. For ventricular volumetry, semiautomatic 3D threshold-based segmentation was used in CT, while a manual simplified contouring 2D method was used in MRI. The indexed ventricular volumes were compared between CT and MRI. The indexed ventricular stroke volumes were compared with the indexed arterial stroke volumes measured using phase-contrast MRI. The mean differences and degrees of agreement in the indexed ventricular and stroke volumes were evaluated using Bland-Altman analysis. Results: The indexed end-systolic (ES) volumes showed no significant difference between CT and MRI (p > 0.05), while the indexed end-diastolic (ED) volumes were significantly larger on CT than on MRI (93.6 ± 17.5 mL/m² vs. 87.3 ± 15.5 mL/m² for the left ventricle [p < 0.001] and 177.2 ± 39.5 mL/m² vs. 161.7 ± 33.1 mL/m² for the right ventricle [p < 0.001], respectively). The mean differences between CT and MRI were smaller for the indexed ES volumes (2.0-2.5 mL/m²) than for the indexed ED volumes (6.3-15.5 mL/m²). CT overestimated the stroke volumes by 14-16%. With phase-contrast MRI as a reference, CT (7.2-14.3 mL/m²) showed greater mean differences in the indexed stroke volumes than did MRI (0.8-3.3 mL/m²; p < 0.005). Conclusion: Compared to 2D MRI, CT ventricular volumetry using semiautomatic 3D threshold-based segmentation provides comparable ES volumes, but overestimates the ED and stroke volumes in patients with repaired tetralogy of Fallot.

• 응용통계연구
• /
• 제16권2호
• /
• pp.305-319
• /
• 2003

본 연구에서는 바이어스 필드에 의해 왜곡된 MRI 영상에 대한 분할을 위해 확장된 EM 알고리즘을 기반으로 한 통계적 접근법을 제시한다. 영상의 명암값을 자료로 하는 분할기법들은 고주파 성분의 잡음 뿐만 아니라 영상을 불균질하게 만드는 바이어스 필드라는 저주파 성분의 왜곡에 특히 취약하다. 이 문제를 해결하기 위해 본 논문에서는 잡음을 효과적으로 제어하기 위해 마코프랜덤필드가 적용된 정규혼합모형을 고려하며, 효과적인 바이어스 필드의 보정을 위해 페널티-우도를 도입하여 추정하는 방법으로 고안되었다.

• 한국통신학회논문지
• /
• 제39C권6호
• /
• pp.490-496
• /
• 2014

무릎 관절 연골은 두께가 얇아 대부분 무릎 질환의 원인이 되고 있다. 그러므로 무릎 자기공명영상에서 관절 연골 분할은 무릎 질환의 정확한 진단을 위한 필수조건이다. 특히 수동이 아닌 전자동 방식으로 무릎 관절 연골을 분할하여야만 효과적인 무릎 질환 진단을 할 수 있다. 본 논문에서는 뇌 자기공명영상에서 대표적으로 사용되는 레벨 셋 기반의 영상 분할 기법을 분석하여 무릎 자기공명영상에 적용 시 문제점을 파악하고 이를 해결함으로써, 무릎 자기공명영상에 레벨 셋 기반 영상분할 방식을 적용하였다. 이는 본 논문에서 제안하는 분할기법을 사용할 경우 무릎 관절 연골 분할에 대한 모든 과정이 전자동화 되어 기존 반자동화 방식보다 빠른 처리가 가능하며, 3차원 형상화를 통해 보다 정확한 진단에 도움을 줄 수 있다. 또한 우리는 제안하고 있는 분할기법이 기존 대표적인 무릎 관절 분할보다 더 높은 정확도를 갖는 것을 실험을 통해 확인할 수 있었다.

• 한국멀티미디어학회논문지
• /
• 제17권5호
• /
• pp.566-572
• /
• 2014

The hippocampus has been known as one of the most important structure related to many neurological disorders, such as Alzheimer's disease. This paper presents the snake model to segment hippocampus from brain MRI. The snake model or active contour model is widely used in medical image processing fields, especially image segmentation they look onto nearby edge, localizing them accurately. We applied a snake model on brain MRI. Then we compared our results with an active shape approach. The results show that hippocampus was successfully segmented by the snake model.

• 정보과학회 논문지
• /
• 제42권10호
• /
• pp.1222-1230
• /
• 2015

무릎 자기공명영상에서 전방십자인대의 분할은 밝기값의 불균일성 및 주변 조직들과의 유사 밝기값 특성으로 인해 기존 분할기법의 적용에 한계가 있다. 본 논문에서는 지역적 정렬을 통한 확률아틀라스 생성 및 반복적 그래프 컷을 통한 다중아틀라스 기반 전방십자인대 분할기법을 제안한다. 첫째, 전역 및 지역적 다중아틀라스 강체정합을 통해 전방십자인대의 확률아틀라스를 생성한다. 둘째, 생성된 확률아틀라스를 이용하여 최대사후추정 및 그래프 컷을 통하여 전방십자인대 초기 분할을 수행한다. 셋째, 마스크 기반 강체정합을 통한 형상정보 개선 및 반복적 그래프 컷을 통해 전방십자인대 분할 개선을 수행한다. 제안방법의 성능평가를 위하여 육안평가 및 정확성평가를 수행하였으며, 평가 결과 제안방법의 Dice 유사도는 75.0%, 평균표면거리는 1.7화소, 제곱근표면거리는 2.7화소로서 기존 그래프 컷 방법에 비하여 전방 십자인대의 분할정확도가 각각 12.8%, 22.7%, 및 22.9% 향상된 것으로 나타났다.

• International Journal of Computer Science & Network Security
• /
• 제24권2호
• /
• pp.43-52
• /
• 2024

Medical imaginings assume a important part in the analysis of tumors and cerebrospinal fluid (CSF) leak. Magnetic resonance imaging (MRI) is an image segmentation technology, which shows an angular sectional perspective of the body which provides convenience to medical specialists to examine the patients. The images generated by MRI are detailed, which enable medical specialists to identify affected areas to help them diagnose disease. MRI imaging is usually a basic part of diagnostic and treatment. In this research, we propose new techniques using the 4D-MRI image segmentation process to detect the brain tumor in the skull. We identify the issues related to the quality of cerebrum disease images or CSF leakage (discover fluid inside the brain). The aim of this research is to construct a framework that can identify cancer-damaged areas to be isolated from non-tumor. We use 4D image light field segmentation, which is followed by MATLAB modeling techniques, and measure the size of brain-damaged cells deep inside CSF. Data is usually collected from the support vector machine (SVM) tool using MATLAB's included K-Nearest Neighbor (KNN) algorithm. We propose a 4D light field tool (LFT) modulation method that can be used for the light editing field application. Depending on the input of the user, an objective evaluation of each ray is evaluated using the KNN to maintain the 4D frequency (redundancy). These light fields' approaches can help increase the efficiency of device segmentation and light field composite pipeline editing, as they minimize boundary artefacts.