• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.4
• /
• pp.201-209
• /
• 2014

Obstructive sleep apnea syndrome(OSAS) is a respiratory disease caused by partial or complete obstruction of the upper airway during sleep. In this paper, we proposed the optical flow method to analyze the upper airway dynamic changes during respiration for children with OSAS and control subjects. We compared the absolute value of difference between inspiration and expiration for airway area analysis method and optical flow method for 5 children with OSAS and 6 control subjects. From the statistical analysis, airway area analysis method and optical flow method are statistically significant at the 0.1 (p value is 0.0977) and 0.01 (p value is 0.0011) significance level respectively. From this simulations, the optical flow method could provide more accurate information to diagnose the OSAS patients than the traditional airway area analysis method.

• The Journal of the Korea Contents Association
• /
• v.12 no.12
• /
• pp.63-70
• /
• 2012

In medical image processing field, object recognition is usually processed based on region segmentation algorithm. Region segmentation in the computing field is carried out by computerized processing of various input information such as brightness, shape, and pattern analysis. If the information mentioned does not make sense, however, many limitations could occur with region segmentation during computer processing. Therefore, this paper suggests effective region segmentation method based on R2-map information within the magnetic resonance (MR) theory. In this study, the experiment had been conducted using images including the liver region and by setting up feature points of R2-map as seed points for 2D region growing and final boundary correction to enable region segmentation even when the border line was not clear. As a result, an average area difference of 7.5%, which was higher than the accuracy of conventional exist region segmentation algorithm, was obtained.

• Journal of Acupuncture Research
• /
• v.32 no.3
• /
• pp.107-115
• /
• 2015

Objectives : The aim of this study is to investigate the state of cervical disc herniation according to the degree of cervical lordosis and age. Methods : 67 records of inpatients who were diagnosed with herniated intervertebral disc(HIVD) of the cervical spine were analyzed. Cobb's method, Jochumsen method and the Ishihara index were used to measure the degree of cervical lordosis. The state of the cervical disc was identified using magnetic resonance imaging(MRI) of the cervical spine. Then correlations among cervical lordosis, age and cervical disc herniation were analyzed. Results : Disc bulging was associated with hypolordosis and disc protrusion was associated with hyperlordosis and age. Disc extrusion was not associated with either cervical lordosis or age. The number of disc herniations in the cervical spine was correlated with age significantly, but not with cervical lordosis. Conclusions : Cervical disc herniation had a tendency to correlate with age and cervical lordosis, although this is not definite. Future studies that analyze more radiographic images of patients with HIVD of the cervical spine might be necessary to identify the influence of cervical lordosis on cervical disc herniation.

• Journal of Trauma and Injury
• /
• v.26 no.3
• /
• pp.151-156
• /
• 2013

Purpose: Traumatic cervical SCI is frequently accompanied by dural tear and the resulting cerebrospinal fluid (CSF) leak after surgery can be troublesome and delay rehabilitation with increasing morbidity. This study evaluated the incidence of intraoperative CSF leaks in patients with traumatic cervical spinal cord injury (SCI) who underwent anterior cervical surgery and described the reliable management of CSF leaks during the perioperative period. Methods: A retrospective study of medical records and radiological images was done on patients with CSF leaks after cervical spine trauma. Results: Seven patients(13.2%) were identified with CSF leaks during the intraoperative period. All patients were severely injured and showed structural abnormalities on the initial magnetic resonance image (MRI) of the cervical spine. Intraoperatively, no primary repair of dural tear was attempted because of a wide, rough defect size. Therefore, fibrin glue was applied to the operated site in all cases. Although a wound drainage was inserted, it was stopped within the first 24 hours after the operation. No lumbar drainage was performed. Postoperatively, the patients should kept their heads in an elevated position and early ambulation and rehabilitation were encouraged. None of the patients developed complications related to CSF leaks during admission. Conclusion: The incidence of CSF leaks after surgery for cervical spinal trauma is relatively higher than that of cervical spinal stenosis. Therefore, one should expect the possibility of a dural tear and have a simple and effective management protocol for CSF leaks in trauma cases established.

• Journal of Veterinary Clinics
• /
• v.31 no.5
• /
• pp.445-448
• /
• 2014

A 7-year-old, male Alaskan malamute was referred for a routine dental management. On the physical examination, the right hindlimb ataxia with a mild delay in proprioception was noted. On magnetic resonance images, an extradural ventral cystic structure compressing the spinal cord was found at the level between the first and second lumbar vertebra. The cyst showed hypointense on a T1-weighted image with rim enhancement and hyperintense on a T2-weighted image. The cystic lesion was removed through right-side hemilaminectomy. In the histopathological examination, evenly distributed fibroblasts and collagenous stroma with several cartilaginous materials were seen. The neurological signs of the right hindlimb were successfully recovered within a week in follow-up neurological examination and showed normal gait at 6 months after surgery.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.16-37
• /
• 2022

Accurate liver segment segmentation based on radiological images is indispensable for the preoperative analysis of liver tumor resection surgery. However, most of the existing segmentation methods are not feasible to be used directly for this task due to the challenge of exact edge prediction with some tiny and slender vessels as its clinical segmentation criterion. To address this problem, we propose a novel deep learning based segmentation model, called Boundary-Aware Dual Attention Liver Segment Segmentation Model (BADA). This model can improve the segmentation accuracy of liver segments with enhancing the edges including the vessels serving as segment boundaries. In our model, the dual gated attention is proposed, which composes of a spatial attention module and a semantic attention module. The spatial attention module enhances the weights of key edge regions by concerning about the salient intensity changes, while the semantic attention amplifies the contribution of filters that can extract more discriminative feature information by weighting the significant convolution channels. Simultaneously, we build a dataset of liver segments including 59 clinic cases with dynamically contrast enhanced MRI(Magnetic Resonance Imaging) of portal vein stage, which annotated by several professional radiologists. Comparing with several state-of-the-art methods and baseline segmentation methods, we achieve the best results on this clinic liver segment segmentation dataset, where Mean Dice, Mean Sensitivity and Mean Positive Predicted Value reach 89.01%, 87.71% and 90.67%, respectively.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.297-299
• /
• 2022

Malignant glioma has a poor prognosis with the reported median survival of between 6 months to 14 months. Thus, it is crucial to predict the accurate survival of patients with malignant glioma. In this paper, we propose a convolutional neural network to predict the overall survival and age of the patients. A total of four MRI modalities, T1, T1-contrast enhanced, T2, and fluid-attenuated inversion recovery, which effectively capture spatial characteristics of malignant glioma, were used as input images. Age is an important factor impacting the overall survival, thus incorporating it into the model will thereby improve the performance of the proposed model. Our model successfully predicted overall survival and age of the patients with pearson correlation coefficients of 0.1748 and 0.3056, respectively.

• Journal of the Korean Society of Radiology
• /
• v.83 no.2
• /
• pp.439-443
• /
• 2022

Hibernoma is a rare benign tumor of brown adipose tissue. Herein, we report a case of axillary hibernoma in a 53-year-old female and discuss the various radiologic findings. The US revealed a 4.5-cm well-defined oval heterogenous hyperechoic mass in the right axilla with anterior displacement of the axillary vessels. Non-enhanced chest CT showed a 5.0-cm well defined, oval, and low-attenuated mass. MRI demonstrated a 5.5-cm mass with heterogeneous intermediate-to-high signal intensity on T1-and T2-weighted images and irregular enhancement at the peripheral portion. The patient underwent an US-guided core needle biopsy and the final diagnosis was hibernoma.

• Journal of the Korean Society of Radiology
• /
• v.83 no.4
• /
• pp.808-829
• /
• 2022

Hepatocellular carcinoma (HCC) can be diagnosed noninvasively on multiphasic CT and MRI based on its distinctive imaging findings. These features include arterial phase hyperenhancement and washout on portal or delayed phase images. However, radiologists face significant diagnostic challenges because some HCCs exhibit atypical imaging characteristics. In addition to many HCC-mimicking lesions, such as arterioportal shunts, combined HCC-cholangiocarcinoma, intrahepatic cholangiocarcinoma, and hemangioma present a challenge for radiologists in actual clinical practice. The ability to distinguish HCCs from mimickers on initial imaging examinations is crucial for appropriate management and treatment decisions. Therefore, this pictorial review presents the imaging findings of atypical HCCs and HCCs mimicking malignant and benign lesions and discusses important clues that may help narrow down the differential diagnosis.

• Investigative Magnetic Resonance Imaging
• /
• v.18 no.3
• /
• pp.208-218
• /
• 2014

Purpose : To evaluate the relationship between the speed of enhancement of hepatic hemangiomas on gadolinium-enhanced MRI and ADC values by using various parameters, including the D, f, $D^*$ and $ADC_{fit}$ on intravoxel incoherent motion (IVIM) MR Imaging. Materials and Methods: The institutional review board approved this retrospective study. A total of 47 hepatic hemangiomas from 39 patients were included (20 men and 19 women). The hemangiomas were classified into three types according to the enhancement speed of the hepatic hemangiomas on gadolinium-enhanced dynamic T1-weighted images: rapid (Type A), intermediate (Type B), and slow (Type C) enhancement. The D, f, $D^*$ and $ADC_{fit}$ values were calculated using IVIM MR imaging. The diffusion/perfusion parameters and ADC values were compared among the three types of hemangiomas. Results: Both the $ADC_{fit}$ and D values of type C were significantly lower than those of type A (P = 0.0022, P = 0.0085). However, for the f and $D^*$, there were no significant differences among the three types. On DWI with all b values (50, 200, 500 and $800sec/mm^2$), the ADC values of type C were significantly lower than those of the type A (P < 0.012). For b values with $800sec/mm^2$, the $ADC_{800}$ values of the type C hemangiomas were significantly lower than those of type B (P = 0.0021). We found a negative correlation between hepatic hemangioma enhancement type and $ADC_{50}$ (${\rho}=-0.357$, P = 0.014), $ADC_{200}$ (${\rho}=-0.537$, P = 0.0001), $ADC_{500}$ (${\rho}=-0.614$, P = 0.0001), and $ADC_{800}$ (${\rho}=-0.607$, P = 0.0001). Therefore, four ADC values of $ADC_{50}$, $ADC_{200}$, $ADC_{500}$, and $ADC_{800}$ were decreased with decreasing enhancement speed. Conclusion: Hepatic hemangiomas had variable ADCs according to the type of enhancement, and the reduced ADCs in slowly enhancing hemangiomas may be related to the reduced pure molecular diffusion (D).