• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1408-1414
• /
• 2019

Deep artificial neural network with transfer learning is applied to compressed sensing cardiovascular MRI. Transfer learning is a method that utilizes structure, filter kernels, and weights of the network used in prior learning for current learning or application. The transfer learning is useful in accelerating learning speed, and in generalization of the neural network when learning data is limited. From a cardiac MRI experiment, with 8 healthy volunteers, the neural network with transfer learning was able to reduce learning time by a factor of more than five compared to that with standalone learning. Using test data set, reconstructed images with transfer learning showed lower normalized mean square error and better image quality compared to those without transfer learning.

• Journal of the Korean Society of Radiology
• /
• v.83 no.4
• /
• pp.898-903
• /
• 2022

The incidence of ductal carcinoma in situ has increased with the rise in screening mammography; currently, ductal carcinoma in situ constitutes 20%-25% of all breast cancers, and up to half of them may become invasive. Its early detection is critical in improving the cure rate. Moreover, MRI has higher sensitivity for its detection than mammography. Herein, we report an unusual case of ductal carcinoma in situ presenting as a continuous, serpentine, linear enhancement with regional distribution on MRI.

• Proceedings of the KSMRM Conference
• /
• 2002.11a
• /
• pp.87-87
• /
• 2002

목적： Writer's cramp를 가진 환자와 정상 성인에서 뇌 활성화를 비교 하고자 한다. 대상 및 방법： Writer's cramp를 가진 1명의 환자와 정상 성인 4명에서 글을 쓰는 과제를 시행하여 관련된 대뇌영역의 활성화를 유도하였다. resting, writering을 반복하는 자극을 시행하였다. 자기공명영상은 3.0T 초전도 자기공명 영상장치에서 EPI BOLD기법을 이용하여 얻었다. 영상 후 처리는 SPM 분석 프로그램을 사용하였다.

• Proceedings of the KSMRM Conference
• /
• 2002.11a
• /
• pp.75-75
• /
• 2002

목적： 고자장 자기공명영상 시스템 등으로 신호대잡음비가 향상됨에 따라 데이터 측정에서 analog-to-digital converter (ADC)의 quantization noise 가 중요한 시스템 사양으로 부각되고 있다. 특히 자기공명영상은 공간주파수 영역에서 데이터를 측정하기 때문에 dc와 ac간의 신호 차이가 매우 크며, 이러한 dynmic range는 3-D 영상에서 더욱 커진다. 본 연구에서는 다양한 자기총명 영상기법 및 실험 파라미터에 따른 신호의 dynamic range와 ADC의 bit 수에 따른 quantization noise를 살펴봄으로써, 주어진 시스템에 적합한 ADC의 bit 수를 분석하고자 한다. 대상 및 방법： 펄스 시퀀스의 종류, 파라미터, 2D/3D 등에 따른 각 신호의 크기를 수학적으로 모델링하여 신호의 크기를 예측하였다. 또한 whole body MRI 시스템에서 실험을 통하여 신호의 크기를 비교하였다. ADC의 quantization noise를 실험과 시뮬레이션을 통하여 살펴보았다. 시뮬레이션은 test 영상을 Inverse FFT 하여 spatial frequency domain data를 만든 후, 다양한 bit 수의 ADC로 quantization을 한 후 다시 영상을 재구성하였다. 재구성된 영상과 원영상 간의 error가 quantization noise가 된다. 또한 이러한 error가 주파수 영역에서의 error 값과 일치하는지를 확인하였다.

• Journal of Biomedical Engineering Research
• /
• v.21 no.2
• /
• pp.145-156
• /
• 2000

본 논문에서는 중앙시상 두뇌 자기공명영상 분할결과를 이용한 3차원 시상 두뇌 자기공명영상의 자동분할기법을 제안한다. 제안된 알고리즘에서는 먼저 3차원 시상 두뇌 자기공명영상의 중앙영상을 분할하고, 분할된 중앙두뇌 자기공명영상을 인접하는 영상에 마스크로 적용한다. 이 때 마스크 적용으로 인하여 인접하는 영상이 절단되는 문제가 발생할 수 있다. 이러한 문제를 해결하기 위하여 절단 영역의 경계점을 검출한 후, 절단 영역에 대한 경로 재설정을 통해 절단 영역을 복원한다. 이러한 경로 재설정을 위해 connectivity-based threshold segmentation algorithm을 사용하였다. 실험결과 제안된 알고리즘의 유용성을 확인할 수 있었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.9
• /
• pp.1138-1143
• /
• 2020

MRI scans in the abdominal area are difficult to achieve optimal images due to artificial respiration. Among 45 patients (male:female = 30:15) who underwent abdominal MRI examination, a metronome-based examination method was studied for patients whose breathing is difficult and difficult to examine. The images examined without using a metronome were divided into group A, and the images examined using a metronome were divided into group B. Image quality improvement (30%) and inspection time (approximately 50 seconds) were reduced in images using metronome. During abdominal magnetic resonance imaging (ABD MRI), the images examined using a metronome had differences in quality and examination time compared to the unused images. It is more effective to use a metronome brace that controls the patient's respiratory rate during abdominal magnetic resonance imaging under respiratory induction in patients with difficulty in respiratory-gated.

• Clinics in Shoulder and Elbow
• /
• v.12 no.2
• /
• pp.173-179
• /
• 2009

Purpose: On the hypothesis that the acromion morphology is changed according to the its site, we identified the morphology of the acromion on the oblique slices of MRI and we investigated the association of the acromial shape with the clinical stages of rotator cuff disorder on the magnetic resonance (MR) images. In addition, we compared the acromion morphology on MRI and simple X-rays. Material and Methods: The MR images of seventy one patients with rotator cuff disorder and who underwent arthroscopic surgery were compared with that of a control group of sixteen patients who didn't have rotator cuff disorder on MRI. On three subsequent oblique sagittal slices from the lateral edge of the acromion (S1, S2 and S3), each acromion morphology on the MRI slices was classified according to Epstein et al: flat, curved or hooked. We investigated the changing parttern of the acromion shape and we compared the acromion shape on MRI and that on simple X-rays. We classified the rotator cuff tear by the severity: bursitis, partial thickness tear or full thickness tear. We investigated which acromial type on the MRI oblique slice was associated with the severity of rotator cuff disease. Results: Changes of the acromial shape occurred in 54 patients (76.1%). The most frequent pattern was that the types are same on S1 and S2 and different on S3 (22 cases, 31.0%). The acromial type on S1 and S2 was significantly associated with the severity of rotator cuff disorder (p=0.001 and 0.022), respectively. There was no reliability of the acromial shape on MRI and roentgenography (p>0.05) Conclusion: The type of acromion changed from lateral to medial. Among the three positions, the shape of the acromion on S1 and S2 had meaningful correlation with the clinical stage of rotator cuff disorder. There was no statistical correlation of the acromial shape between MRI and simple X-ray.

• Journal of Science and Technology Studies
• /
• v.14 no.2
• /
• pp.199-222
• /
• 2014

Functional Magnetic Resonance Angiography (fMRA) was a technoscientific innovation that allows scientists to directly view the changes made in the blood vessels of a brain. fMRA was first developed at Neuroscience Research Institute (NRI) in Korea. fMRA mainly utilizes 7 Tesla MRI technology, and NRI is equipped with the instrument. First article on fMRA was published in 2008, and two more papers in 2010 and 2012 consecutively had been published on the newly developed technique. However, fMRA is a competitive technology with existing fMRI. Both techniques capture microvascular changes in a brain, and by doing it, both techniques visualize the cognitive and affective changes. fMRI technology was introduced by Seiji Ogawa in the early 1990's and has been widely used since then. In contrast, fMRA was a newer technology and rather unknown. Developers of fMRA in NRI used series of signifiant-politics in order to make it better known to scientific community as well as public. By resetting the boundaries of existing concept of fMRI, they tried to lower the threshold of a new concept/technique. This case study shows how technoscientists use semiotic strategies governing new technology.

• Journal of KIISE:Software and Applications
• /
• v.27 no.8
• /
• pp.815-826
• /
• 2000

This study is for the segmentation and volume calculation of the white matter and gray matter from brain MRI. In general, the volume of white and gray matter is reduced by contraction of each components in the case of mental retardation which are Alzheimer's disease and Down's syndrome. As results, it is useful for diagnostic and early detection for various mental retardation through the tracing of variation for its volume from the brain MRI. But, until now, it was very difficult to calculate the partial volume of each components existing in some thickness, because MR image was represented by single gray value after scanning by MR scanner. Accordingly, new segmentation algorithm proposed in this paper is to calculate the partial volume of the white and gray matter existing in some thickness through the analysis of the blurred gray value, and is to determine the threshold for segmentation of white and gray matter, and is to calculate the volume of each segmented component. And finally, proposed algorithm was applied the models which was created manually, and then acquired results was compared with that of original model.

• Investigative Magnetic Resonance Imaging
• /
• v.16 no.2
• /
• pp.136-141
• /
• 2012

Purpose : To investigate the pharmacologic modulation of motor task-dependent physiologic responses by antiplatelet agent, clopidogrel, during hand motor tasks in healthy subjects. Materials and Methods: Ten healthy, right-handed subjects underwent three functional magnetic resonance (fMRI) sessions: one before drug administration, one after high dose drug administration and one after reaching drug steady state. For the motor task fMRI, finger flexion-extension movements were performed. Blood oxygenation level dependent (BOLD) contrast was collected for each subject using a 3.0 T VHi (GE Healthcare, Milwaukee, USA) scanner. $T2^*$-weighted echo planar imaging was used for fMRI acquisition. The fMRI data processing and statistical analyses were carried out using SPM2. Results: Second-level analysis revealed significant increases in the extent of activation in the contralateral motor cortex including primary motor area (M1) after drug administration. The number of activated voxels in motor cortex was 173 without drug administration and the number increased to 1049 for high dose condition and 673 for steady-state condition respectively. However, there was no significant difference in the magnitude of BOLD signal change in terms of peak T value. Conclusion: The current results suggest that cerebral motor activity can be modulated by clopidogrel in healthy subjects and that fMRI is highly senstive to evidence such changes.