• Investigative Magnetic Resonance Imaging
• /
• v.23 no.2
• /
• pp.81-99
• /
• 2019

Recently, deep learning methods have shown great potential in various tasks that involve handling large amounts of digital data. In the field of MR imaging research, deep learning methods are also rapidly being applied in a wide range of areas to complement or replace traditional model-based methods. Deep learning methods have shown remarkable improvements in several MR image processing areas such as image reconstruction, image quality improvement, parameter mapping, image contrast conversion, and image segmentation. With the current rapid development of deep learning technologies, the importance of the role of deep learning in MR imaging research appears to be growing. In this article, we introduce the basic concepts of deep learning and review recent studies on various MR image processing applications.

• Journal of Electrical Engineering and information Science
• /
• v.2 no.6
• /
• pp.36-42
• /
• 1997

This paper presents a new design technique for performing high quality low bit rate image compression. A hybrid coder(HC) which combines Mean Removed Important Coefficient Selection based JPEG(MR-ICS-JPEG) and Adaptive Vector Quantization (AVQ) is proposed. A new quantization table is developed using the Important Coefficient Selection(ICS) method; the importance of each coefficient is determined using the orthonormal property of the DCT. This quantization table is applied to standard JPEG with mean removal(MR) strategy before processing. This scheme, called MR-ICS-JPEG, produces more than 2 dB enhanced performance in terms of PSNR over standard JPEG. A set of homogeneous codebooks is generated by homogeneous training vectors. Before compression, an image is uniformly divided into 8${\times}$8 blocks. Low detail regions such as backgrounds are roughly coded by AVQ while high detail regions such as edges or curves are finely coded by the proposed MR-ICS-JPEG. This hybrid coder procuces consistently about 3 dB improved performance in terms of PSNR over standard JPEG.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.1
• /
• pp.286-295
• /
• 2000

We present a novel statistically adaptive method using the Minimum Description Length(MDL) principle for unsupervised segmentation of magnetic resonance(MR) images. In the method, Markov random filed(MRF) modeling of tissue region accounts for random noise. Intensity measurements on the local region defined by a window are modeled by a finite Gaussian mixture, which accounts for image inhomogeneities. The segmentation algorithm is based on an iterative conditional modes(ICM) algorithm, approximately finds maximum ${\alpha}$ posteriori(MAP) estimation, and estimates model parameters on the local region. The size of the window for parameter estimation and segmentation is estimated from the image using the MDL principle. In the experiments, the technique well reflected image characteristic of the local region and showed better results than conventional methods in segmentation of MR images with inhomogeneities, especially.

• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2002.09a
• /
• pp.429-431
• /
• 2002

Metabolic analysis of biological tissues, the interventional radiology in MRT (Magnetic Resonance Treatment) and for clinical diagnoses, representation of 4-Dimensional (4D) structural information (x,y,z,t) of biological tissues is required. This paper discusses image representation techniques for those 4D MR Images. We have proposed an image reconstruction method for ultra-fast 3D MRI. It is based on image interpolation and prediction of un-acquired pictorial data in both of the real and the k-space (the acquisition domain in MRI). A 4D MR image is reconstructed from only two 3D MR images and acquired a few echo signals that are optimized by prediction of the tissue motion. This prediction can be done by the phase of acquired echo signal is proportioned to the tissue motion. On the other hand, reconstructed 4D MR images are represented as a 3D-movie by using computer graphics techniques. Rendered tissue surfaces and/or ROIs are displayed on a CRT monitor. It is represented in an arbitrary plane and/or rendered surface with their motion. As examples of the proposed representation techniques, the finger and the lung motion of healthy volunteers are demonstrated.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.5
• /
• pp.311-317
• /
• 2015

The aim of this study is to develop a 3D registration algorithm for positron emission tomography/computed tomography (PET/CT) and magnetic resonance (MR) images acquired from independent PET/CT and MR imaging systems. Combined PET/CT images provide anatomic and functional information, and MR images have high resolution for soft tissue. With the registration technique, the strengths of each modality image can be combined to achieve higher performance in diagnosis and radiotherapy planning. The proposed method consists of two stages: normalized mutual information (NMI)-based global matching and independent component analysis (ICA)-based refinement. In global matching, the field of view of the CT and MR images are adjusted to the same size in the preprocessing step. Then, the target image is geometrically transformed, and the similarities between the two images are measured with NMI. The optimization step updates the transformation parameters to efficiently find the best matched parameter set. In the refinement stage, ICA planes from the windowed image slices are extracted and the similarity between the images is measured to determine the transformation parameters of the control points. B-spline. based freeform deformation is performed for the geometric transformation. The results show good agreement between PET/CT and MR images.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.405-408
• /
• 2003

영상의 구성물질에 따른 정확한 분할은 질병의 유무를 판단하는데 매우 중요하다. 그러나 영상에서 구성물질들을 정확하게 분할하기란 쉬운 문제가 아니다. 그리고 많은 연구들이 뇌의 실질적인 량을 고려하지 못한 상태서 분할이 이루어지고 있다. 따라서 뇌의 실질적인 량과 비교할 때 가장 근접한 방법 의 개발이 필요하다고 볼 수 있다. 본 논문은 fat을 소거한 T2 영상과 T1 영상을 이용하여 조직에 따르는 명암 분포가 각각 다르게 분포되어 있는 것을 이용하여 평활화한 후 두 영상의 차로 백질, 회백질, 뇌척수액을 분리하는 방법을 제안한다. 이 방법을 이용하여 정상이의 뇌 MR 영상 이용하여 (19 Slice) 백질, 회백질, 뇌척수액을 분리하는 방법을 제시하였다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.11a
• /
• pp.159-164
• /
• 2001

본 연구에서 설계하고 구현한 뇌 MR영상 처리기에서는 뇌 MR 영상에서 진단에 필요한 정보들을 자동 추출한다. 의료영상 처리 시에는 수집된 의료영상의 특징을 분석하고 특징들을 분류해야 하며 이를 위해서는 효율적인 특징 추출 알고리즘들 필요하다. 뇌 MR 영상 처리기는 영상의 잡음제거나 영상 강화를 위한 전처리기, 영상의 특징을 추출하기 위한 영역분할기와 전역, 지역 특징 추출기로 구성된다. 뇌 MR 영상 특징 추출을 위한 효율적인 의료영상 처리기의 개발 내용을 기술한다.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.2
• /
• pp.542-551
• /
• 2000

In this paper, a novel technique is presented for automatic brain region segmentation in single channel MR image data sets for 3D visualization and analysis. The method detects brain contours in 2D and 3D processing of four steps. The first and the second make a head mask and an initial brain mask by automatic thresholding using a curve fitting technique. The stage 3 reconstructs 3D volume of the initial brain mask by cubic interpolation and generates an intermediate brain mask using morphological operation and labeling of connected components. In the final step, the brain mask is refined by automatic thresholding using curve fitting. This algorithm is useful for fully automatic brain region segmentation of T1-weighted, T2-weighted, PD-weighted, SPGR MRI data sets without considering slice direction and covering a whole volume of a brain. In the experiments, the algorithm was applied to 20 sets of MR images and showed over 0.97 in comparison with manual drawing in similarity index.

• Journal of Digital Contents Society
• /
• v.14 no.2
• /
• pp.275-282
• /
• 2013

In medical image processing field, object recognition is usually carried out by computerized processing of various input information such as brightness, shape, and pattern. If the information mentioned does not make sense, however, many limitations could occur with object recognition during computer processing. Therefore, this paper suggests effective object recognition method based on the magnetic resonance (MR) theory to resolve the basic limitations in computer processing. We propose the efficient method of robust gray-white matter segmentation by texture analysis through the Susceptibility Weighted Imaging (SWI) for contrast enhancement. As a result, an average area difference of 5.2%, which was higher than the accuracy of conventional region segmentation algorithm, was obtained.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.343-344
• /
• 2009

The aim in this paper is to show how to extract scalp of a series of brain MR images by using region growing segmentation algorithm. Most researches are all forces on the segmentation of skull, gray matter, white matter and CSF. Prior to the segmentation of these inner objects in brain, we segmented the scalp and the brain from the MR images. The scalp mask makes us to quickly exclude background pixels with intensities similar those of the skull, while the brain mask obtained from our brain surface. We make use of connected threshold method (CTM) and confidence connected method (CCM). Both of them are two implementations of region growing in Insight Toolkit (ITK). By using these two methods, the results are displayed contrast in the form of 2D and 3D scalp images.