• 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.

• The KIPS Transactions:PartB
• /
• v.12B no.5 s.101
• /
• pp.553-560
• /
• 2005

Generally, it is known that the spleen accompanied by liver cirrhosis is hypertrophied or enlarged. We examined the wave pattern of the spleen by using abdominal CT images of a patient with liver cirrhosis, and found that they are different from those of a person with a normal liver In the abdominal CT image of the patient with liver cirrhosis, there is a deep wave part on the left side of the spleen. In the case of the normal liver, there are waves on the left side, but they aren't deep. Therefore, the total area of waving parts of the spleen with liver cirrhosis is found to be greater than that of the spleen with the normal liver. Moreover, when examining circularity by abstracting the waves of the spleen from the image iO liver cirrhosis, we found they are more circular than those of the spleen accompanied by a normal liver. This paper suggests an automatic method to diagnose splenic enlargement by using the wave pattern of the spleen in abdominal CT images on the basis of the two principles. It tells us that we can judge if the abdomen has a focal splenic enlargement automatically, without the manual test of the size of spleen, only with the shape of spleen.

• Journal of KIISE:Software and Applications
• /
• v.32 no.6
• /
• pp.532-541
• /
• 2005

We examined the wave pattern of the spleen by using abdominal CT images of a patient with liver cirrhosis, and found that they are different from those of a person with a normal liver. In the abdominal CT image of the patient with liver cirrhosis, there is a deep wave part on the left side of the spleen. In the case of the normal liver, there are waves on the left side, but they aren't deep. Therefore, the total area of waving parts of the spleen with liver cirrhosis is found to be greater than that of the spleen with the normal liver. Moreover, when examining circularity by abstracting the waves of the spleen from the image with liver cirrhosis, we found they are more circular than those of the spleen accompanied by a normal liver. This paper suggests an automatic method to diagnose liver cirrhosis by using the wave pattern of the spleen in abdominal CT images on the basis of the two principles. It tells us that we can judge if the liver has liver cirrhosis automatically, without the manual test of the ratio of caudate lobe to right lobe, only with the spleen.

• Korean Journal of Radiology
• /
• v.20 no.5
• /
• pp.749-758
• /
• 2019

Objective: To investigate whether a computer-aided diagnosis (CAD) system based on a deep learning framework (deep learning-based CAD) improves the diagnostic performance of radiologists in differentiating between malignant and benign masses on breast ultrasound (US). Materials and Methods: B-mode US images were prospectively obtained for 253 breast masses (173 benign, 80 malignant) in 226 consecutive patients. Breast mass US findings were retrospectively analyzed by deep learning-based CAD and four radiologists. In predicting malignancy, the CAD results were dichotomized (possibly benign vs. possibly malignant). The radiologists independently assessed Breast Imaging Reporting and Data System final assessments for two datasets (US images alone or with CAD). For each dataset, the radiologists' final assessments were classified as positive (category 4a or higher) and negative (category 3 or lower). The diagnostic performances of the radiologists for the two datasets (US alone vs. US with CAD) were compared Results: When the CAD results were added to the US images, the radiologists showed significant improvement in specificity (range of all radiologists for US alone vs. US with CAD: 72.8-92.5% vs. 82.1-93.1%; p < 0.001), accuracy (77.9-88.9% vs. 86.2-90.9%; p = 0.038), and positive predictive value (PPV) (60.2-83.3% vs. 70.4-85.2%; p = 0.001). However, there were no significant changes in sensitivity (81.3-88.8% vs. 86.3-95.0%; p = 0.120) and negative predictive value (91.4-93.5% vs. 92.9-97.3%; p = 0.259). Conclusion: Deep learning-based CAD could improve radiologists' diagnostic performance by increasing their specificity, accuracy, and PPV in differentiating between malignant and benign masses on breast US.

• Journal of Chest Surgery
• /
• v.25 no.10
• /
• pp.989-1000
• /
• 1992

The concepts of modern type computer are so called "General purpose, stored program and digital computer" that is proposed by Charles Babbage. ENIAC, the initial operational electronic digital computer model, was produced in 1946. During the last 50 years, an epoch-making development of the personal computer was marked. The computerization of all levels of society is going on and also computerization of the general hospital and medical college is developing. But patient data management system for clinician is not used generally. We suggest the use of computer aided data management application programs for the clinical informations of the patients of the Department of Thoracic and Cardiovascular Surgery for better management and to make best of medical informations, to co-operate with the current of this times, and to prepare against the Hospital Information Systems[HIS], actively. Also, we suggest to standardize the format and structure of database files to store the clinical data of the patients By standardization of the database files, we can integrate and relate the data of the individual department or hospital, build up the regional or national statistics of the patients easily, and promote the generation of application programs. The medical network by the communication and computer would be utilized to collect the database files. And finally, we suggest the use of code system to input and search the informations about the diagnosis and operation such as the code system of International Classfication of Disease[WHO] and the table of the classfication of operation of the Ministry of Health and Social Affairs, Korea. In this article, we tried to show the new standards, the essential items for computerization of clinical informations of the patients of the Department of Thoracic and Cardiovascular Surgery.r Surgery.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.20 no.4
• /
• pp.77-83
• /
• 2016

Recently, interests for maintenance of transmission tower are increasing to extend life of structures and reduce maintenance cost. However, existing classical diagnosis method of corrosion deteriorated degree on the transmission tower steel members, visual inspection, has a problem that error often due to difference of inspector's individual knowledge and experience. In order to solve the problem, this study carried out to develop the corrosion deterioration inspection tool for transmission tower steel members. This tool is composed of camera equipment and computer-aided diagnosis system. We standardized the photographing method by camera equipment to obtain suitable pictures for image processing. Diagnosis system was designed to evaluate automatically degree of corrosion deterioration for member of transmission tower on the basis of the RGB color image processing techniques. It is anticipated that developed the corrosion deterioration inspection tool will be very helpful in decision of optimal maintenance time for transmission tower corrosion.

• Journal of Biomedical Engineering Research
• /
• v.25 no.5
• /
• pp.421-430
• /
• 2004

Clinically, it is almost impossible for a physician to distinguish subtle changes of frequency spectrum by using a stethoscope alone especially in the early stage of thrombus formation. Considering that reliability of mechanical valve is paramount because the failure might end up with patient death, early detection of valve thrombus using noninvasive technique is important. Thus the study was designed to provide a tool for early noninvasive detection of valve thrombus by observing shift of frequency spectrum of acoustic signals with computer aid diagnosis system. A thrombus model was constructed on commercialized mechanical valves using polyurethane or silicon. Polyurethane coating was made on the valve surface, and silicon coating on the sewing ring of the valve. To simulate pannus formation, which is fibrous tissue overgrowth obstructing the valve orifice, the degree of silicone coating on the sewing ring varied from 20％, 40％, 60％ of orifice obstruction. In experiment system, acoustic signals from the valve were measured using microphone and amplifier. The microphone was attached to a coupler to remove environmental noise. Acoustic signals were sampled by an AID converter, frequency spectrum was obtained by the algorithm of spectral analysis. To quantitatively distinguish the frequency peak of the normal valve from that of the thrombosed valves, analysis using a neural network was employed. A return map was applied to evaluate continuous monitoring of valve motion cycle. The in-vivo data also obtained from animals with mechanical valves in circulatory devices as well as patients with mechanical valve replacement for 1 year or longer before. Each spectrum wave showed a primary and secondary peak. The secondary peak showed changes according to the thrombus model. In the mock as well as the animal study, both spectral analysis and 3-layer neural network could differentiate the normal valves from thrombosed valves. In the human study, one of 10 patients showed shift of frequency spectrum, however the presence of valve thrombus was yet to be determined. Conclusively, acoustic signal measurement can be of suggestive as a noninvasive diagnostic tool in early detection of mechanical valve thrombosis.

• The Journal of the Korea Contents Association
• /
• v.10 no.10
• /
• pp.54-67
• /
• 2010

Applying the conventional machine-learning method which has been frequently used in health-care area has several fundamental problems for modern U-health service analysis. First of all, we are still lack of application examples of the traditional method for our modern U-health environment because of its short term history of U-health study. Second, it is difficult to apply the machine-learning method to our U-health service environment which requires real-time management of disease because the method spends a lot of time in the process of learning. Third, we cannot implement a personalized U-health diagnosis system using the conventional method because there is no way to assign weights on the disease-related variables although various kinds of machine-learning schemes have been proposed. In this paper, a novel diagnosis scheme PCADP is proposed to overcome the problems mentioned above. PCADP scheme is a personalized diagnosis method and it makes the bio-data analysis just a 'process' in the U-health service system. In addition, we offer a semantics modeling of the U-health ontology framework in order to describe U-health data and service specifications as meaningful representations based on this PCADP. The PCADP scheme is a kind of statistical diagnosis method which has characteristics of flexible structure, real-time processing, continuous improvement, and easy monitoring of decision process. Upto the best of authors' knowledge, the PCADP scheme and ontology framework proposed in this paper reveals one of the best characteristics of flexible structure, real-time processing, continuous improvement, and easy monitoring among recently developed U-health schemes.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.10 no.4
• /
• pp.259-265
• /
• 2016

In this paper, new method was proposed to classify lung tissues such as Broncho vascular, Emphysema, Ground Glass Reticular, Ground Glass, Honeycomb, Normal for early lung disease detection. 459 Statistical features was extraced from joint histogram matrix based on multi resolution analysis, volumetric LBP, and CT intensity, then dominant features was selected by using adaboost learning. Accuracy of proposed features and 3D AMFM was 90.1% and 85.3%, respectively. Proposed joint histogram based features shows better classification result than 3D AMFM in terms of accuracy, sensitivity, and specificity.