• Korean Journal Plant Pathology
• /
• v.3 no.2
• /
• pp.131-136
• /
• 1987

Rates of lesion development over time and disease gradients over distance for blast disease on the two rice varieties, Brazos and M-20 1 were significantly affected by two different cultural conditions, upland and flooded conditions. Flooding rice field plots lowered the rates of lesion increase and flattened the disease gradients for both varieties. Despite absence of statistically significant differences in the rate of lesion increase between four sampled distances from infection focus, rate of lesion development tended to be slightly greater as distance from the infection focus increases. Rate of lesion increase was greater with more susceptible variety M-201 than with Brazos. Disease gradient was steeper for M-201 than for Brazos. As blast disease progressed, disease gradients became flattened regardless of variety due to the infections originated from secondary foci. Between two empirical disease gradient models examined, Kiyosawa & Shiyomi model was fitted better over Gregory model. Rates of blast isopath movement under upland conditions were calculated as approximately 0.2m/day and 0.4 m/day for Brazos and M-201, respectively. The results in this study suggest that differences in varietal resistance to blast could be detected by measuring disease gradient as efficiently as by measuring infection rate.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.693-694
• /
• 2015

This paper present a decision method of middle ear disease which is developed in children and adults. In the proposed method, features are extracted from the middle ear disease images and normal images using HOG(histogram of oriented gradient) descriptor and the extracted features are learned by SVM(support vector machine) classifier. Input images are classified by SVM classifier based on the model of learning features. Experimental results show that the method yields accuracy of over 90% in decision.

• Journal of Korea Multimedia Society
• /
• v.24 no.2
• /
• pp.178-185
• /
• 2021

Three-dimensional high-resolution magnetic resonance imaging (MRI) provides fine-level anatomical information for disease diagnosis. However, there is a limitation in obtaining high resolution due to the long scan time for wide spatial coverage. Therefore, in order to obtain a clear high-resolution(HR) image in a wide spatial coverage, a super-resolution technology that converts a low-resolution(LR) MRI image into a high-resolution is required. In this paper, we propose a super-resolution technique through filter learning based on information on the surrounding gradient information in 3D space from 3D MRI images. In the learning step, the gradient features of each voxel are computed through eigen-decomposition from 3D patch. Based on these features, we get the learned filters that minimize the difference of intensity between pairs of LR and HR images for similar features. In test step, the gradient feature of the patch is obtained for each voxel, and the filter is applied by selecting a filter corresponding to the feature closest to it. As a result of learning 100 T1 brain MRI images of HCP which is publicly opened, we showed that the performance improved by up to about 11% compared to the traditional interpolation method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.3
• /
• pp.621-629
• /
• 2016

This paper presents a decision method of middle ear disease which is developed in children and adults. In the proposed method, features are extracted from the middle ear disease images and normal images using HoG (histogram of oriented gradient) descriptor and the extracted features are learned by SVM (support vector machine) classifier. To obtain an input vector into SVM, an input image is resized to a predefined size and then the resized image is partitioned into 16 blocks each of which is partitioned into 4 sub-blocks (namely cell). Finally, the feature vector with 576 components is given by using HoG with 9 bins and it is used as SVM learning and classification. Input images are classified by SVM classifier based on the model of learning features. Experimental results show that the proposed method yields the precision of over 90% in decision.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.2
• /
• pp.576-590
• /
• 2023

Machine learning is widely used in various academic fields, and recently it has been actively applied in the medical research. In the medical field, machine learning is used in a variety of ways, such as speeding up diagnosis, discovering new biomarkers, or discovering latent traits of a disease. In the respiratory field, a relative regional air volume change (RRAVC) map based on quantitative inspiratory and expiratory computed tomography (CT) imaging can be used as a useful functional imaging biomarker for characterizing regional ventilation. In this study, we seek to predict RRAVC using various regular machine learning models such as extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and multi-layer perceptron (MLP). We experimentally show that MLP performs best, followed by XGBoost. We also propose several relative coordinate systems to minimize intersubjective variability. We confirm a significant experimental performance improvement when we apply a subject's relative proportion coordinates over conventional absolute coordinates.

• Mass Spectrometry Letters
• /
• v.6 no.4
• /
• pp.116-119
• /
• 2015

Globotriaosylsphingosine (lyso-Gb3) is considered as one of the biological marker for Fabry disease. To date, a reliable biomarker that reflects disease severity and progression has not been discovered to guide the management of Fabry disease. A new method included a simple protein precipitation with acetonitrile in 100 μL of plasma following analyte separation on an Phenomenex Kintex- C18 column using a gradient elution (0.1% formic acid in 5-90% acetonitrile). Total run time was within 12 min including sample preparation and MS/MS analysis. The limit of detection and limit of quantitation were 1 ng/mL and 2 ng/mL, respectively. The calibration curve was linear over the concentration range of 2.0-200.0 ng/mL (r² = 0.9999). Inter-day accuracy and precision at 7 level were 93.4-100.7% with RSD of 0.55-5.97%. Absolute recovery was 97.6-98.6%. The method was applied to human and mice plasma, proved the suitability for quantification of lyso-Gb3 for screening, diagnosis and therapeutic monitoring of Fabry disease patients.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1233-1241
• /
• 2022

Tumors of the brain are the deadliest, with a life expectancy of only a few years for those with the most advanced forms. Diagnosing a brain tumor is critical to developing a treatment plan to help patients with the disease live longer. A misdiagnosis of brain tumors will lead to incorrect medical treatment, decreasing a patient's chance of survival. Radiologists classify brain tumors via biopsy, which takes a long time. As a result, the doctor will need an automatic classification system to identify brain tumors. Image classification is one application of the deep learning method in computer vision. One of the deep learning's most powerful algorithms is the convolutional neural network (CNN). This paper will introduce a novel deep learning structure and image gradient to classify brain tumors. Meningioma, glioma, and pituitary tumors are the three most popular forms of brain cancer represented in the Figshare dataset, which contains 3,064 T1-weighted brain images from 233 patients. According to the numerical results, our method is more accurate than other approaches.

• Clinical and Experimental Pediatrics
• /
• v.55 no.8
• /
• pp.297-300
• /
• 2012

Symptomatic pulmonary arterial hypertension (PAH) in patients with isolated atrial septal defect (ASD) is rare during infancy. We report a case of isolated ASD with severe PAH in an infant who developed airway obstruction as cardiomegaly progressed. The patient presented with recurrent severe respiratory insufficiency and failure to thrive before the repair of the ASD. Echocardiography confirmed volume overload on the right side of heart and severe PAH (tricuspid regurgitation [TR] with a peak pressure gradient of 55 to 60 mmHg). The chest radiographs demonstrated severe collapse of both lung fields, and a computed tomography scan showed narrowing of the main bronchus because of an intrinsic cause, as well as a dilated pulmonary artery compressing the main bronchus on the left and the intermediate bronchus on the right. ASD patch closure was performed when the infant was 8 months old. After the repair of the ASD, echocardiography showed improvement of PAH (TR with a peak pressure gradient of 22 to 26 mmHg), and the patient has not developed recurrent respiratory infections while showing successful catch-up growth. In infants with symptomatic isolated ASD, especially in those with respiratory insufficiency associated with severe PAH, extrinsic airway compression should be considered. Correcting any congenital heart diseases in these patients may improve their symptoms.

• Journal of Chest Surgery
• /
• v.25 no.9
• /
• pp.925-929
• /
• 1992

Supravalvular aortic stenosis is relatively uncommon form of congenital heart disease and the most important lesion of this anomaly is various narrowing of the aortic lumen just above the sinus of Valsalva. We experienced a case of hourglass type of supravalvular aortic stenosis involving lcm from length from lcm above the sinus of Valsalva. The patient was associated with mental retardation, peculiar facies and dental anomaly. The diagnosis was confirmed preoperatively by retrograde left heart catheterization and left ventriculography. An incision was made in the ascending aorta and into the right coronary and noncornary sinus. Care was taken to protect the right coronary artery. A Y-shaped patch of Dacron was made to enlarge the stenotic portion of aorta. Postoperative pressure gradient between the aorta and left ventricle markedly reduced 36 mmHg in comparison with preoperative pressure gradient 150mmHg. The boy was discharged without any event.

• Proceedings of the KSME Conference
• /
• 2003.04a
• /
• pp.1915-1918
• /
• 2003

Atherosclerosis, which is a degenerate disease, is believed to occur in the vascular system due to deposition of cholesterol and low density lipoprotein(LDL) or thrombosis on the blood vessel. Atherosclerosis narrows arterial lumen, which is known as stenosis phenomenon of blood vessel. Pathogenesis of atherosclerosis is thought to occur mainly by aging. Restenosis phenomenon is observed in the same site of insertion of a stent and balloon angioplasty after treatment of interventional theraphy. Several hypothetical theories related to the generation of atherosclerosis have been reported: high shear stress theory, low shear stress theory, high shear stress gradient theory, flow separation and turbulence theory and high pressure theory. However, no one theory clearly explains the causes of atherosclerosis. In the present study the generation of atherosclerosis in the left coronary artery is investigated. The hypotheses are verified by using the computer simulation.