• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.37-44
• /
• 2021

Automatic classification of brain MRI images play an important role in early diagnosis of brain tumors. In this work, we present a deep learning-based brain tumor classification model in MRI images using ensemble of deep features. In our proposed framework, three different deep features from brain MR image are extracted using three different pre-trained models. After that, the extracted deep features are fed to the classification module. In the classification module, the three different deep features are first fed into the fully-connected layers individually to reduce the dimension of the features. After that, the output features from the fully-connected layers are concatenated and fed into the fully-connected layer to predict the final output. To evaluate our proposed model, we use openly accessible brain MRI dataset from web. Experimental results show that our proposed model outperforms other machine learning-based models.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.425-427
• /
• 2022

The grade of glioma is important information related to survival and thus is important to classify the grade of glioma before treatment to evaluate tumor progression and treatment planning. Glioma grading is mostly divided into high-grade glioma (HGG) and low-grade glioma (LGG). In this study, image preprocessing techniques are applied to analyze magnetic resonance imaging (MRI) using the deep neural network model. Classification performance of the deep neural network model is evaluated. The highest-performance EfficientNet-B6 model shows results of accuracy 0.9046, sensitivity 0.9570, specificity 0.7976, AUC 0.8702, and F1-Score 0.8152 in 5-fold cross-validation.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.289-290
• /
• 2022

Arteriosclerosis is a disease in which the carotid vessel wall becomes thick, and it is important to monitor the thickness of the vessel wall for diagnosis. In this study, we propose a model for extracting 324 radiomics features from carotid MRI images and diagnosing arteriosclerosis using machine learning techniques. We learned a total of four classification models: logistic regression, support vector machine, random forest, and XGBoost through radiomics features. XGBoost model, which showed the highest performance in 5-fold cross-validation, shows the results of accuracy 0.9023, sensitivity 0.9517, specificity 0.8035, AUC 0.8776.

• Speech Sciences
• /
• v.2
• /
• pp.45-55
• /
• 1997

This paper describes the procedure of implementing an articulatory speech simulator, in order to model the human articulatory organs and to synthesize speech from this model after. Images required to construct the vocal tract model were obtained from MRI, they were then used to construct 2D and 3D vocal tract shapes. In this paper 3D vocal tract shapes were constructed by spatially concatenating and interpolating sectional MRI images. 2D vocal tract shapes were constructed and analyzed automatically into a digital filter model. Following this speech sounds corresponding to the model were then synthesized from the filter. All procedures in this study were using MATLAB.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.416-418
• /
• 2022

Glioma is a type of brain tumor that occurs in glial cells and is classified into two types: high hrade hlioma with a poor prognosis and low grade glioma. Magnetic resonance imaging (MRI) as a non-invasive method is widely used in glioma diagnosis research. Studies to obtain complementary information by combining multiple modalities to overcome the incomplete information limitation of single modality are being conducted. In this study, we developed a 3D CNN-based model that applied input-level fusion to MRI of four modalities (T1, T1Gd, T2, T2-FLAIR). The trained model showed classification performance of 0.8926 accuracy, 0.9688 sensitivity, 0.6400 specificity, and 0.9467 AUC on the validation data. Through this, it was confirmed that the grade of glioma was effectively classified by learning the internal relationship between various modalities.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.172-174
• /
• 2003

화상평면내 미지호흡운동에 기인한 MRI 아티팩트를 제거하기 위한 후처리방법을 제안한다. 본 연구에서 호흡운동은 2차원의 선형확대축소운동으로 모델화 된다. 신체조직을 비압축성 유체모양의 물질로 가정할때, 화상위에서의 단위체적당 푸로톤 밀도는 일정하다고 가정한다. 사용한 모델에 따르면 호흡운동은 위상 오차와 비균일표본화 및 왜곡된 진폭변조를 MR 데이터에 부여한다. 운동 파라메타가 이미 알려져 있거나 추정 가능하다고 할 때, MRI 아티팩트를 제거하기 위하여 중첩법에 기초를 둔 재구성 알고리즘을 이용한다. 운동 파라매타가 미지인 경우 스팩트럼 이동법을 적용해서 호흡변동함수와 x 방향 확대계수 및 x 방향 확대중심을 추정한다. 다음으로 에너지 최소법을 이용해서 ｙ 방향 확대계수 및 ｙ 방향 확대중심을 추정한다. 시뮬레이션을 통해서 제안한 방법의 유효성을 확인한다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.5
• /
• pp.53-62
• /
• 2004

In this study, an improved post-processing technique for correcting MRI artifact due to the unknown respiratory motion in the imaging plane is presented. Respiratory motion is modeled by a two-Dimensional linear expending-shrinking movement. Assuming that the body tissues are incompressible fluid like materials, the proton density per unit volume of the imaging object is kept constant. According to the introduced model, respiratory motion imposes phase error, non-uniform sampling and amplitude modulation distortions on the acquired MRI data. When the motion parameters are known or can be estimatead a reconstruction algorithm based on biliner superposition method was used to correct the MRI artifact. In the case of motion parameters are unknown, first, the spectrum shift method is applied to find the respiratory fluctuation function, x directional expansion coefficient and x directional expansion center. Next, y directional expansion coefficient and y directional expansion center are estimated by using the minimum energy method. Finally, the validity of this proposed method is shown to be effective by using the simulated motion images.

• Journal of the Korean Society of Radiology
• /
• v.8 no.7
• /
• pp.389-395
• /
• 2014

The research was carried out to develop a animal model of malignant brain tumor for the researches in glioblastoma multiform (GBM) diagnosis and therapy. C6 cells were transplanted into the right striatum of SD rat using stereotactic instrument for the development. The developed animal model was verified by MRI and H&E stain assay of anatomicohistological examination. The MRI observations showed that the tumor developed at the injection site at the 7 days after glioblastoma inoculation. At 14 days post inoculation, the tumor grew to a large volume occupying almost a half of the right cerebral hemisphere. It was confirmed that the expression of excessive mitosis and pleomorphism in anatomicohistological examination. The developed animal model must be necessary and useful tool for the in vivo level research in the development of the new modality for the diagnosis and therapy of brain cancer.

• Journal of the Korean Society of Radiology
• /
• v.11 no.3
• /
• pp.139-146
• /
• 2017

At the Brain MRI examination, RF Pulse are irradiated on the human head in order to acquire MR images. At this time, a considerable part of the irradiated RF Pulse energy is absorbed in our body and the temperature of the human head will rise depending on the degree of exposure, so it will affect the human head. Even if the same RF Pulse energy is given, if the metal is inserted in the human head, the conductivity of the human head is greatly increased by the metal, so the SAR value increases and the temperature also rises. Therefore, we started this research with the question as to whether there is difference between the change in SAR value and temperature displayed on the head of the human according to use or not of the dental implant. Experiments were using the XFDTD program on a 128 MHz RF Pulse frequency by a 3.0 tesla MRI. We can see that both are increasing that the average value of SAR and temperature that absorbed by the human head model used the dental implant. In addition, the average maximum SAR value and the maximum temperature rise in the brain part are shown below the international safety standard value but the influence can not be ignored because of the result may change according to the increase in the number of dental implant. And as future tasks. we need to the simulation of temperature rise and SAR due to an increase in the number of implants and volumes of teeth, dental implant material.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.13-14
• /
• 2023

뇌의 전기적 신경활동을 측정하는 뇌전도(EEG)는 저렴하게 취득할 수 있고 높은 시간 해상도를 갖는 반면 공간적 정보를 제공하지는 않는다. 기능적 자기공명영상(fMRI)은 혈류변화를 감지하여 뇌활동을 측정하는 방식으로서 높은 공간 분해능을 갖지만 고가의 비용과 설비를 요구한다. 최근 저렴하게 취득할 수 있는 EEG 데이터로부터 딥러닝을 사용하여 fMRI 합성영상을 생성하는 기술이 제안되었지만, 저주파수 대역에서 EEG와 fMRI 간의 뇌과학적 상관관계를 반영하지는 않는다. 본 연구에서는 휴식상태에서 취득된 EEG 데이터를 스펙트로그램으로 변환한 후 저주파수 특성을 사용하여 fMRI 합성영상을 생성하는 U-net 기반의 크로스 모달리티 변환 모델의 실현가능성을 평가하였다.