• Transactions on Electrical and Electronic Materials
• /
• v.15 no.4
• /
• pp.230-234
• /
• 2014

Automatic detection of disease helps medical institutions that are introducing digital images to read images rapidly and accurately, and is thus applicable to lesion diagnosis and treatment. The aim of this study was to apply a symmetry contribution algorithm to unsharp mask filter-applied MR images and propose an analysis technique to automatically recognize brain tumor and edema. We extracted the skull region and drawed outline of the skull in database of images obtained at P University Hospital and detected an axis of symmetry with cerebral characteristics. A symmetry contribution algorithm was then applied to the images around the axis of symmetry to observe intensity changes in pixels and detect disease areas. When we did not use the unsharp mask filter, a brain tumor was detected in 60 of a total of 95 MR images. The disease detection rate for the brain was 63.16%. However, when we used the unsharp mask filter, the tumor was detected in 87 of a total of 95 MR images, with a disease detection rate of 91.58%. When the unsharp mask filter was used in the pre-process stage, the disease detection rate for the brain was higher than when it was not used. We confirmed that unsharp mask filter can be used to rapidly and accurately to read many MR images stored in a database.

• Journal of Biomedical Engineering Research
• /
• v.30 no.2
• /
• pp.153-161
• /
• 2009

Mammogram is one of the important techniques for mass detection, which is the early diagnosis stage of a breast cancer. Especially, the CAD(Computer Aided Diagnosis) using mammogram improves the working performance of radiologists as it offers an effective mass detection. There are two types of CAD systems using mammogram; automatic and semi-automatic CAD systems. However, the automatic segmentation is limited in performance due to the difficulty of obtaining an accurate segmentation since mass occurs in the dense areas of the breast tissue and has smoother boundaries. Semi-automatic CAD systems overcome these limitations, however, they also have problems including high FP (False Positive) rate and a large amount of training data required for training a classifier. The proposed system which overcomes the aforementioned problems to detect mass is composed of the suspected area selection, the level set segmentation and SVM (Support Vector Machine) classification. To assess the efficacy of the system, 60 test images from the FFDM (Full-Field Digital Mammography) are analyzed and compared with the previous semi-automatic system, which uses the ANN classifier. The experimental results of the proposed system indicate higher accuracy of detecting mass in comparison to the previous systems.

• Journal of Multimedia Information System
• /
• v.9 no.1
• /
• pp.21-32
• /
• 2022

Early-stage diagnosis of Alzheimer's Disease (AD) from Cognitively Normal (CN) patients is crucial because treatment at an early stage of AD can prevent further progress in the AD's severity in the future. Recently, computer-aided diagnosis using magnetic resonance image (MRI) has shown better performance in the classification of AD. However, these methods use a traditional machine learning algorithm that requires supervision and uses a combination of many complicated processes. In recent research, the performance of deep neural networks has outperformed the traditional machine learning algorithms. The ability to learn from the data and extract features on its own makes the neural networks less prone to errors. In this paper, a dense neural network is designed for binary classification of Alzheimer's disease. To create a classifier with better results, we studied result of different activation functions in the prediction. We obtained results from 5-folds validations with combinations of different activation functions and compared with each other, and the one with the best validation score is used to classify the test data. In this experiment, features used to train the model are obtained from the ADNI database after processing them using FreeSurfer software. For 5-folds validation, two groups: AD and CN are classified. The proposed DNN obtained better accuracy than the traditional machine learning algorithms and the compared previous studies for AD vs. CN, AD vs. Mild Cognitive Impairment (MCI), and MCI vs. CN classifications, respectively. This neural network is robust and better.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.425-427
• /
• 1999

본 논문에서는 시간 간격을 두고 활영한 흉부의 단순 x선 영상의 차영상을 이용하여 컴퓨터 도움 진단에 활용할 수 있는 방법에 대해 연구하였다. 시간 간격을 두고 촬영한 흉부 단순 x선 영상의 차영상은 시간에 따른 변화를 명확히 보여줌으로써 질병의 조기진단 및 질병의 전개과정 등을 알아보는데 유용하게 쓰일 수 있다. 특히, 이 방법은 폐암과 같이 조기진단이 매우 어려운 질병에 대하여 정기검진 등에서 정기적으로 촬영한 단순 x선 영상을 이용하여 조기진단을 할 수 있는 방법으로 활용될 수 있다. 그러나, 촬영시의 여러 가지 조건들, x선의 세기와 조영시간, 환자의 촬영 자세 및 호흡 상태 등에 따라 단순 x선 영상이 크게 달라져 단순한 뺄셈에 의한 차영상은 진단에 도움이 되지 못한다. 진단에 도움을 주기 위해서는 두 영상 사이의 전체적인 밝기와 대조도를 맞추고 늑골, 쇄골 등 해부학적 구조물의 위치와 크기를 서로 맞추어 차영상을 얻는 영상처리 방법이 필요하다. 또한, 폐의 크기와 위치도 서로 맞추어 차영상을 얻어야 한다. 그러나, 이러한 방법도 늑골과 폐의 크기와 위치 변화가 서로 일치하지 않는 문제점을 가지고 있다. 본 논문에서는 이러한 영상처리를 통하여 차영상을 얻는 방법에 대하여 논하고 방법상의 문제점과 해결 방법을 제시한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.2352-2355
• /
• 2003

In recent yeas, studies on computer aided diagnosis (CAD) using image analysis on CT images have been conducted with respect to various diseases. Extracting ground-glass opacities (GGO) on lung CT images is one of such subjects, though it has not found an established method yet. If the region of ground-glass opacities is large on CT images, it can be detected without much difficulty. On the other hand, if the region is small, it is still difficult to find it exactly. In the latter case, increasing overlooking possibility cannot be avoided according to smaller size of the region. To solve this difficulty, this paper proposes an automatic technique for extracting ground-glass opacities on lung CT images employing some statistical parameters of a gray level histogram and a differential histogram. The proposed technique is applied to some lung CT images in the performed experiment. The results are shown with discussion on future work.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.553-555
• /
• 2002

컴퓨터지원진단(Computer Aided Diagnosis; CAD) 시스템은 방사선 의사들이 흉부 X-ray 영상에서 결절을 탐지하는데 있어 실제적으로 발생할 수 있는 오진율을 줄이고, 폐 결절이 존재하는 폐야에서 결절의 존재 유무를 판단하여 검출을 표시함으로써 진단율을 개선시킬 수 있도록 하였다. 본 논문은 흉부 X-ray 영상에서의 폐 결절을 추출하는데 유전자 알고리즘(Genetic Algorithm)을 이용한 템플릿 매칭(Template Matching) 방법을 제안한다. 제안한 방법은 흉부 X-ray 영상에 존재하는 결절과 레퍼런스 이미지를 매칭시켜 적합도를 계산한 후, 그 값을 통하여 수치가 낮은 개체를 선택하여 높은 개체와 교차시킨다. 그리고 레퍼런스 이미지는 결절이 존재하는 환자 X-ray 영상에서 샘플 노듈을 추출한 후 가우시안 분포를 갖는 512개의 레퍼런스 이미지를 생성하였다. 본 논문에서 사용된 영상은 결절 50개, 비결절 30개와 흉부 X-ray 영상에서 육안으로 판별이 가능한 결절 영상을 20개를 포함하여 총 100개 영상을 사용하였다. 실험 결과 83%의 결절을 자동 추출 하였으며, 가장 적절한 레퍼런스 이미지를 발견하고 이를 흉부영상에 매칭시켜 정확한 결절의 위치를 확인하였다.

• Clinical Endoscopy
• /
• v.51 no.6
• /
• pp.547-551
• /
• 2018

Unlike wired endoscopy, capsule endoscopy requires additional time for a clinical specialist to review the operation and examine the lesions. To reduce the tedious review time and increase the accuracy of medical examinations, various approaches have been reported based on artificial intelligence for computer-aided diagnosis. Recently, deep learning-based approaches have been applied to many possible areas, showing greatly improved performance, especially for image-based recognition and classification. By reviewing recent deep learning-based approaches for clinical applications, we present the current status and future direction of artificial intelligence for capsule endoscopy.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.12
• /
• pp.1678-1684
• /
• 2018

Dental caries is one of the most popular oral disease. The aim of automatic dental cavity detection system is helping dentist to make accurate diagnosis. It is very important to separate cavity from the teeth in the detection system. In this paper, We compared two active contour algorithms, Snake and DRLSE(Distance Regularized Level Set Evolution). To improve performance, image is selected ROI(region of interest), then applied bilateral filter, Canny edge. In order to evaluate the algorithms, we applied to 7 tooth phantoms from incisor to molar. Each teeth contains two cavities of different shape. As a result, Snake is faster than DRLSE, but Snake has limitation to compute topology of objects. DRLSE is slower but those of performance is better.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.04a
• /
• pp.106-109
• /
• 2009

Recently, image matching becomes important in Computer Aided Diagnosis (CAD) due to the huge amount of medical images. Specially, texture feature is useful in medical image matching. However, texture features such as co-occurrence matrices can't describe well the spatial distribution of gray levels of the neighborhood pixels. In this paper we propose a frequency domain-based texture feature extractor that describes the local spatial distribution for medical image retrieval. This method is based on 2D Local Discrete Fourier transform of local images. The features are extracted from local Fourier histograms that generated by four Fourier images. Experimental results using 40 classes Brodatz textures and 1 class of Emphysema CT images show that the average accuracy of retrieval is about 93%.

• Journal of Korea Multimedia Society
• /
• v.13 no.5
• /
• pp.722-729
• /
• 2010

Prediction problems are widely used in medical domains. For example, computer aided diagnosis or prognosis is a key component in a CDSS (Clinical Decision Support System). SVMs with nonlinear kernels like RBF kernels, have shown superior accuracy in prediction problems. However, they are not preferred by physicians for medical prediction problems because nonlinear SVMs are difficult to visualize, thus it is hard to provide intuitive interpretation of prediction results to physicians. Nomogram was proposed to visualize SVM classification models. However, it cannot visualize nonlinear SVM models. Localized Radial Basis Function (LRBF) was proposed which shows comparable accuracy as the RBF kernel while the LRBF kernel is easier to interpret since it can be linearly decomposed. This paper presents a new tool named VRIFA, which integrates the nomogram and LRBF kernel to provide users with an interactive visualization of nonlinear SVM models, VRIFA visualizes the internal structure of nonlinear SVM models showing the effect of each feature, the magnitude of the effect, and the change at the prediction output. VRIFA also performs nomogram-based feature selection while training a model in order to remove noise or redundant features and improve the prediction accuracy. The area under the ROC curve (AUC) can be used to evaluate the prediction result when the data set is highly imbalanced. The tool can be used by biomedical researchers for computer-aided diagnosis and risk factor analysis for diseases.