• Proceedings of the Korean Society of Computer Information Conference
• /
• 2017.01a
• /
• pp.239-240
• /
• 2017

본 연구는 의료영상내 특징정보를 추출 및 평가함으로서 정밀의료 실현 가능성을 확인하고자 하였다. 영상화는 PET/CT 및 MRI 스캐너를 이용하여 암환자의 기능적 정보와 해부학적 정보를 획득하고 관심영역을 설정하였으며 각각의 영상내 특징정보를 추출하였다. 영상내 특징정보는 GLCM을 이용하여 에너지, 대비, 엔트로피, 균질성을 획득하였고, 획득된 영상 데이터에 따른 관심영역 설정 차이를 확인하였다. 영상내 특징 정보는 MRI 영상의 해부학적 정보를 이용한 분석결과에서 엔트로피 및 균질성이 PET 보다 증가 하였고 대비는 감소함을 확인하였다. 추후연구는 다양한 영상내 특징 정보를 획득하고 정밀의료를 위한 기계학습에 활용할 예정이다.

• Journal of the Korean Society of Radiology
• /
• v.5 no.3
• /
• pp.149-154
• /
• 2011

The purpose of this study was to examine the usefulness of diffusion weighted images in breast MRI by performing a quantitative comparative analysis in patients diagnosed with DCIS. On a 3.0T MR scanner, diffusion weighted images and ADC map images were obtained from 20 patients histologically diagnosed with ductal carcinoma in situ (DCIS). The findings from the quantitative image analysis are the following: The diffusion weighted images showed higher SNR and CNR at the lesion area. In addition, the ADC values were lower at the lesion area.

• Journal of IKEEE
• /
• v.12 no.3
• /
• pp.158-166
• /
• 2008

Medical ultrasound has benefits in mobility and safety than any other medical techniques such as X-ray, CT and MRI but has speckle noise which decrease the ability of an observer to distinguish the fine details in diagnostic examination. But simple removing of speckle often causes losing boundary information. Then, in this paper, we presented a novel neuro-fuzzy method which could remove speckle efficiently without loss of boundary information. Proposed method consists of image clustering by fuzzy algorithm and image processingby neural networks which was learned by back propagation. From the experiments for simulation image and real ultrasound image, we could verify the proposed method.

• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2003.09a
• /
• pp.77-77
• /
• 2003

CT와 MRI의 단면 영상을 대상으로 영상분할 (Image segmentation)과 Image registration방법을 이용하여 인체 모델을 개발 하고자 한다. 우선 인체의 Head와 Neck부분의 CT와 MR 영상을 얻어 뼈, 근육, 인대, 그리고 그 밖의 장기의 해부학적 영상 특징을 분석하였다. 인체의 Head와 Neck 부분에 대한 CT와 MR 영상에 대해 각 부위별로 ROI(region-of-interrest)를 설정하였고, 각 volxel 마다 3차원 좌표를 계산할 수 있는 소프트웨어를 개발하였다. 특히 각 해부학적 영상에서 부위별로 CT 번호를 분석하고, pulse sequence에 따른 MRI 영상의 부위별 특정을 분석하였다. 이 분석한 자료를 바탕으로 영상 분할을 하였다. 영상 분할전에 각종 잡음(noise) 제거 및 영상 분할을 효과적으로 처리하기 위해 기본적인 영상처리 (filtering)를 구현하였고, 대조도(contrast) 및 밝기(brightness)를 조절할 수 있게 프로그램을 구현하였다. 영상 분할 방법 중 선(line) 및 에지(edge) 의 검출 방법, 문턱치화(threshold) 방법, 영역확대(region growing) 방법으로 영상 분할을 해봄으로써 우리의 인체 모델링 개발에 가장 적합한 영상 분할 알고리듬 방법을 찾도록 시도하였다. 결과적으로 말하면, 한가지 방법의 알고리듬을 쓰는 것보다는 인체의 부위에 따라 두 가지 이상의 알고리듬 방법을 쓰는 것이 원하고자 하는 부위를 영상 분할하는데 더 효과적이다는 것을 알게 되었다. 우리의 연구 과제에서는 영역확대(region growing) 방법과 문턱치화 방법, 모드법(피크니스, 밸리)의 알고리듬을 이용하여 영상 분할을 한 결과 우리가 얻고자 하는 인체 부위별 중 근육과 뼈를 구별하는데는 별 무리가 없었으나, 인대 및 기타 장기를 구별하는데는 어려움을 겪게 되었다. 이후에 좀더 알고리듬을 연구하여 이번 연구에서 구별하기 어려운 장기 부분도 구별 할 수 있도록 노력하겠다.

• Journal of Korean Neurosurgical Society
• /
• v.49 no.6
• /
• pp.363-366
• /
• 2011

Cerebellar pilocytic astrocytomas (PAs) are benign gliomas predominantly found in the pediatric population. Intracranial hemorrhages are extremely rare in initial presentations of cerebellar PAs. There are no reports in the medical literature of adult cerebellar PA cases presenting with intratumoral hemorrhage. We report 2 cases of adult cerebellar pilocytic astrocytomas with intratumoral hemorrhage. The first case is a 37-year-old woman presenting with severe headache, nausea, and vomitting. Computed tomography demonstrated an acute hemorrhage adjacent to the right cerebellar hemisphere and hydrocephalus. Magnetic resonance imaging (MRI) revealed a cerebellar vermian tumor with the hemorrhage as a mixed isoin-tense area in the T2-weighted image, and as a mixed hyperintense area in the contrast-enhanced T1-weighted image. The second case is a 53-year-old man presenting with headache for 3 weeks. MRI revealed a cerebellar hemispheric tumor with the hemorrhage as a mixed hyperintense area. It had a cystic mass with a heterogeneous enhanced mural nodule in the gadolinium-enhanced T1-weighted image and a fluid-fluid level within the cyst in the T2-weighted image. Both of them underwent radical resections of their respective lesions. Histological examination of the specimens revealed typical astrocytoma, including a hemorrhagic portion. Both patients recovered postoperatively and continue to do well at present. The medical literature on hemorrhagic cerebellar PAs is also reviewed.

• Journal of Biomedical Engineering Research
• /
• v.31 no.2
• /
• pp.134-140
• /
• 2010

Readily available high resolution brain MRI scans allow detailed visualization of the brain structures. Researchers have focused on developing methods to quantify shape differences specific to diseased scans. We have developed a novel method to quantify shape information for a specific population based on Multidimensional scaling(MDS). MDS is a well known tool in statistics and here we apply this classical tool to quantify shape change. Distance measures are required in MDS which are computed from pair-wise image registrations of the training set. Registration step establishes spatial correspondence among scans so that they can be compared in the same spatial framework. One benefit of our method is that it is quite robust to errors in registrations. Applying our method to 13 brain MRI showed clear separation between normal and diseased (Cushing's syndrome). Intentionally perturbing the image registration results did not significantly affect the separability of two clusters. We have developed a novel method to quantify shape based on MDS, which is robust to image mis-registration.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.9
• /
• pp.4336-4354
• /
• 2018

Fuzzy C-means (FCM) algorithm is a most usually technique for medical image segmentation. But conventional FCM fails to perform well enough on magnetic resonance imaging (MRI) data with the noise and intensity inhomogeneity (IIH). In the paper, we propose a Gamma correction conditional FCM algorithm with spatial information (GcsFCM) to solve this problem. Firstly, the pre-processing, Gamma correction, is introduced to enhance the details of images. Secondly, the spatial information is introduced to reduce the effect of noise. Then we introduce the effective neighborhood mechanism into the local space information to improve the robustness for the noise and inhomogeneity. And the mechanism describes the degree of participation in generating local membership values and building clusters. Finally, the adjustment mechanism and the spatial information are combined into the weighted membership function. Experimental results on four image volumes with noise and IIH indicate that the proposed GcsFCM algorithm is more effective and robust to noise and IIH than the FCM, sFCM and csFCM algorithms.

• Nuclear Engineering and Technology
• /
• v.38 no.4
• /
• pp.327-342
• /
• 2006

Radiation therapy is an important part of cancer treatment in which cancer patients are treated using high-energy radiation such as x-rays, gamma rays, electrons, protons, and neutrons. Currently, about half of all cancer patients receive radiation treatment during their whole cancer care process. The goal of radiation therapy is to deliver the necessary radiation dose to cancer cells while minimizing dose to surrounding normal tissues. Success of radiation therapy highly relies on how accurately 1) identifies the target and 2) aim radiation beam to the target. Both tasks are strongly dependent of imaging technology and many imaging modalities have been applied for radiation therapy such as CT (Computed Tomography), MRI (Magnetic Resonant Image), and PET (Positron Emission Tomogaphy). Recently, many researchers have given significant amount of effort to develop and improve imaging techniques for radiation therapy to enhance the overall quality of patient care. For example, advances in medical imaging technology have initiated the development of the state of the art radiation therapy techniques such as intensity modulated radiation therapy (IMRT), gated radiation therapy, tomotherapy, and image guided radiation therapy (IGRT). Capability of determining the local tumor volume and location of the tumor has been significantly improved by applying single or multi-modality imaging fur static or dynamic target. The use of multi-modality imaging provides a more reliable tumor volume, eventually leading to a better definitive local control. Image registration technique is essential to fuse two different image modalities and has been In significant improvement. Imaging equipments and their common applications that are in active use and/or under development in radiation therapy are reviewed.

• Journal of Magnetics
• /
• v.22 no.2
• /
• pp.238-243
• /
• 2017

In this research, we report the preliminary results of an insertional type parallel transmission (pTx) array that has 7-elements that are placed in the space above a patient table as a transmit (Tx) coil to give an RF transmission ($B_1{^+}$) field for the body object of a 3 Tesla (T) MRI system. In previous research, we have tried to compare the performances of different coil elements and array geometries for a pTx body image. Based on these results, we attempt to obtain a human image with the proposed pTx array. Through the simulation and experimental results, we introduce a possible structure of multi-channel Tx array and verify the utility of a multi-channel Tx body image using $B_1{^+}$ shimming. The insertional pTx array, combined with a receiver (Rx) array coil, provides an enhanced $B_1{^+}$ field homogeneity in a large ROI image as a result of $B_1{^+}$ shimming applied over the full body size object. Through this research, we hope to determine the usefulness of the proposed insertional type RF coil combination for 3 T body imaging.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2788-2808
• /
• 2023

Brain tumors are one of the most threatening malignancies for humans. Misdiagnosis of brain tumors can result in false medical intervention, which ultimately reduces a patient's chance of survival. Manual identification and segmentation of brain tumors from Magnetic Resonance Imaging (MRI) scans can be difficult and error-prone because of the great range of tumor tissues that exist in various individuals and the similarity of normal tissues. To overcome this limitation, the Amended Convolutional Neural Network (ACNN) model has been introduced, a unique combination of three techniques that have not been previously explored for brain tumor detection. The three techniques integrated into the ACNN model are image tissue preprocessing using the Kalman Bucy Smoothing Filter to remove noisy pixels from the input, image tissue segmentation using the Isotonic Regressive Image Tissue Segmentation Process, and feature extraction using the Marr Wavelet Transformation. The extracted features are compared with the testing features using a sigmoid activation function in the output layer. The experimental findings show that the suggested model outperforms existing techniques concerning accuracy, precision, sensitivity, dice score, Jaccard index, specificity, Positive Predictive Value, Hausdorff distance, recall, and F1 score. The proposed ACNN model achieved a maximum accuracy of 98.8%, which is higher than other existing models, according to the experimental results.