• Journal of Biomedical Engineering Research
• /
• v.38 no.6
• /
• pp.321-329
• /
• 2017

This paper presents a bone metastasis Detection algorithm on abdominal computed tomography images for early detection using fully convolutional neural networks. The images were taken from patients with various cancers (such as lung cancer, breast cancer, colorectal cancer, etc), and thus the locations of those lesions were varied. To overcome the lack of data, we augmented the data by adjusting the brightness of the images or flipping the images. Before the augmentation, when 70% of the whole data were used in the pre-test, we could obtain the pixel-wise sensitivity of 18.75%, the specificity of 99.97% on the average of test dataset. With the augmentation, we could obtain the sensitivity of 30.65%, the specificity of 99.96%. The increase in sensitivity shows that the augmentation was effective. In the result obtained by using the whole data, the sensitivity of 38.62%, the specificity of 99.94% and the accuracy of 99.81% in the pixel-wise. lesion-wise sensitivity is 88.89% while the false alarm per case is 0.5. The results of this study did not reach the level that could substitute for the clinician. However, it may be helpful for radiologists when it can be used as a screening tool.

• Journal of radiological science and technology
• /
• v.39 no.2
• /
• pp.193-198
• /
• 2016

In clinical computed tomography (CT), regular quality assurance (QA) has been required. This study is to evaluate the MTF for analyzing the spatial resolution using AAPM phantom in CT exam. The dual source somatom definition flash (siemens healthcare, forchheim, Germany), the brilliance 64 (philips medical system Netherlands) and aquilion 64 (toshiba medical system, Japan) were used in this study. The quantitative evaluation was performed using the image J (wayne rasband national institutes of health, USA) and chart method which is measurement of modulation transfer function (MTF). In MTF evaluation, the spatial frequencies corresponding to the 50% MTF for the CT systems were 0.58, 0.28, and $0.59mm^{-1}$, respectively and the 10% MTF for the CT systems were 1.63, 0.89, and $1.21mm^{-1}$, respectively. This study could evaluate the characteristic of spatial resolution of MTF using chart method, suggesting the quantitative evaluation method using the data.

• Journal of Veterinary Clinics
• /
• v.26 no.5
• /
• pp.490-494
• /
• 2009

A 2-year-old Maltese and a 5-month-old Yorkshire terrier were presented with ataxia. Tetraparesis was observed in a 9-year -old Yorkshire terrier. The localizations of the lesions suggested brain or cervical spinal cord by the neurological examination, and the following images was achieved: radiography, axial images of computed tomography (CT), reconstruction image of CT such as multi-planar reformation(MPR) and 3-dimensional(3D) reconstruction and magnetic resonance imaging (MRI). On radiography, the misalignment between atlas (C1) and axis (C2), absent dens of axis, and increased space between the dorsal arch of C1 and spinous process of C2 were found. The discontinuation between dens and body of C2 was identified through axial CT images, and the fragmentation of dens separated from axis was observed through MPR and 3D image in all case. The hyperintense lesions and the spinal cord compression on T2-weighted MR images were represented in a dog with tetraparesis, the others represented only spinal cord compression. Three dogs were diagnosed as atlantoaxial instability (AAI) by dens fracture of C2. The dog with tetraparesis was euthanized due to guarded prognosis. The others were recovered completely. It is difficult to differentiate dens fracture of C2 from abnormal dens such as agenesis and hypoplasia. We thought that CT is very useful to evaluate the dens of C2 and differentiate the causes of AAI, and the reconstruction images of CT such as MPR and 3D make the translation of the fragmented dens or axis of AAI more precisely evaluate.

• Journal of Veterinary Clinics
• /
• v.33 no.1
• /
• pp.6-9
• /
• 2016

Bronchoarterial (BA) ratio is a commonly used criterion to define airway dilatation despite the lack of normative human and animals. The objective of our study was to compare the range of normal bronchial to accompanying arterial diameter ratio with previous reports on CT scan of the thorax in dogs and assess influence anesthetics on BA ratio in dogs. Dogs undergoing multidetector CT scan of the chest for nonpulmonary conditions at a single center were prospectively identified. High-resolution reconstruction was performed on those included and both airway lumen and vessel diameters were measured in the lobar bronchi of the left cranial (cranial and caudal parts), right cranial, right middle, left caudal, and right caudal lung lobes. Eight dog were included; Mean of the mean BA ratios was $1.43{\pm}0.24$ (95% CI = 1.36 - 1.50) in inhalation anesthetic group. In propofol group, the mean of the mean BA ratios was $1.13{\pm}0.29$ (95% CI = 1.04 - 1.22). In medetomidine group, the mean of the mean BA ratios was $0.89{\pm}0.19$ (95% CI = 0.83 - 0.95). Comparing individual lobes within anesthetic category, there was no signicant difference in mean BA ratio between lung lobes or between dog according to inhalation, propofol, and medetomidine group (P = 0.630, P = 0.878, and P = 0.508, respectively). The BA ratio in these clinically normal dogs was consistent and may be a useful tool in evaluating for bronchiectasis on CT images. However, some different criteria for bronchiectasis were applied by the anesthetic methods.

• Clinics in Shoulder and Elbow
• /
• v.16 no.2
• /
• pp.135-140
• /
• 2013

We report a case of recurrent shoulder dislocation and combined proximal humerus fracture in neurofibromatosis type I. A twenty-year-old male patient with known neurofibromatosis type I presented with right shoulder dislocation and proximal humerus fracture following a minor trauma. His injured arm also had large plexiform neurofibromas on the elbow, and the dislocation of the shoulder joint was proven to be recurrent. Bony deformation of the humerus and scapula were revealed on X-ray and computed tomography, as well as multiple neurofibromatic changes of the deltoid and periscapular muscles were noted on magnetic resonance imaging. Open reduction and capsular reconstruction were performed and humeral fracture was stabilized with a reconstruction plate. Satisfactory union and functional result were achieved and the dislocation did not recur until the 7-month followup after the procedure. Recurrent shoulder dislocation and combined proximal humerus fracture in neurofibromatosis type I are rare complications. However, it should still be considered in the orthopaedic evaluation process of the diseases.

• Journal of Veterinary Clinics
• /
• v.31 no.5
• /
• pp.449-453
• /
• 2014

A 2-year-old, intact male, Persian exotic cat, weighing 2 kg with a history of a palpated abdominal mass was admitted to Veterinary Medical Teaching Hospital, Seoul National University. On routine complete blood count (CBC) and serum biochemical analyses, there were anemia and 0.45 of albumin to globulin ratio. A feline infectious peritonitis (FIP) virus antibody test kit was negative. Radiography, ultrasonography and computed tomography (CT) were performed. A markedly enlarged abdominal mass was found. On cytologic examination of this mass, it was diagnosed as alimentary lymphoma or pyogranulomatous inflammation. Chemotherapeutic treatment for intestinal lymphoma was provided for several weeks, but the mass size was increased and clinical signs were not improved. The cat died six days after discontinuing chemotherapy. On postmortem examination, the definitive diagnosis was FIP. This case describes a cat with FIP in which an abdominal tumor had been suspected clinically.

• Journal of Veterinary Clinics
• /
• v.27 no.5
• /
• pp.593-599
• /
• 2010

A seven-year-old, castrated male Basset hound weighing 21.1 kg was presented for investigation of anorexia, lethargy, weight loss, melena and vomiting for 4 months. In laboratory findings, microcytic hypochromic anemia was identified, and other results were within a normal reference range. On the plain radiographs, soft tissue opacity was increased in the descending duodenal region; and on the contrast radiographs, ulcerative changes were identified in the entire segment of descending duodenum. Ultrasonographic findings included increased duodenal wall thickness and duodenal wall layering was lost. Endoscopy revealed irregular mucosal surface and luminal narrowing of the descending duodenum. There was concentrically thickened descending duodenum on the computed tomography. And the wall of the descending duodenum showed heterogenous enhancement after contrast agent injection. On histopathological findings, both chronic inflammation with mucosal proliferation and neoplastic changes with multiple small glandslike structures invading into the submucosa were identified. Based on these findings, presented case was diagnosed as an annular form duodenal adenocarcinoma. After 13 months of supportive medical treatment, the patient was expired. The purpose of this case report was to describe the duodenal adenocarcinoma in a dog.

• Radiation Oncology Journal
• /
• v.28 no.2
• /
• pp.99-105
• /
• 2010

Purpose: To compare radiation dose of the brain and lens among various conventional whole brain radiotherapy (WBRT) techniques. Materials and Methods: Treatment plans for WBRT were generated with planning computed tomography scans of 11 patients. A traditional plan with an isocenter located at the field center and a parallel anterior margin at the lateral bony canthus was generated (P1). Blocks were automatically generated with a 1 cm margin on the brain (5 mm for the lens). Subsequently, the isocenter was moved to the lateral bony canthus (P2), and the blocks were replaced into the multileaf collimator (MLC) with a 5 mm leaf width in the craniocaudal direction (P3). For each patient plan, 30 Gy was prescribed at the isocenter of P1. Dose volume histogram (DVH) parameters of the brain and lens were compared by way of a paired t-test. Results: Mean values of $D_{max}$ and $V_{105}$ of the brain in P1 were 111.9% and 23.6%, respectively. In P2 and P3, $D_{max}$ and $V_{105}$ of the brain were significantly reduced to 107.2% and 4.5~4.6%, respectively (p<0.001). The mean value of $D_{mean}$ of the lens was 3.1 Gy in P1 and 2.4~2.9 Gy in P2 and P3 (p<0.001). Conclusion: WBRT treatment plans with an isocenter located at the lateral bony canthus have dosimetric advantages for both the brain and lens without any complex method changes.

• Journal of the Korean Society of Radiology
• /
• v.10 no.3
• /
• pp.161-169
• /
• 2016

In this study, the proposal to seek ways to reduce the amount of radiation is drawn by comparing and analyzing CT Dose Index(CTDI) on the pediatric head CT which was performed at the busan regional hospitals, to the national diagnostic reference levels. As a result, it was appeared to exceed the amount of the dose recommendation in order of hospital, general hospital and senior general hospital in the hospital-specific classification and from 2 to 5 year, from 1 month to 1 year and from 6 to 10 year in the age-specific classification. In addition, the amount of the dose recommendation was exceed in order of helical, axial and volume in the scan-specific classification. As the results of the scan range reset to match the diagnostic reference level, the dose reduction showed 11.68%, 15.79% and 20.66% in senior general hospital, general hospital and hospital respectively. In the results of analysing patient average scan ranges which does not deviate from the guideline of patient dose recommendation, there was age of 1 month to 1 year, 2 to 5 year and 6 to 10 year of $03.2{\pm}11.8mm$, $110.5{\pm}14.5mm$, and $117.8{\pm}17.2mm$ respectively.

• Journal of Yeungnam Medical Science
• /
• v.18 no.1
• /
• pp.123-137
• /
• 2001

Background: The purpose of this study is to evaluate the accuracy of measurements obtained from 3-dimensional computerized tomography and 3-dimensional cephalogram constructed by using the frontal and lateral cephalogram of six human dry skulls. Materials and Methods: After CT scans and each cephalograms were taken, 3-dimensional coordinates (X, Y, Z) of landmarks were obtained using computer programs. In this study, the accuracy of both methods were determined by means of 14 linear measurements compare with caliper measurements. Results: The standard deviation of landmarks of 3-dimensional CT and 3-dimensional cephalogram were 0.23 mm, and 0.30 mm in X axis, 0.27 mm and 0.25 mm in Y axis, and 0.27 mm and 0.31 mm in Z axis. In both methods, the standard deviation were less than 0.5 mm in all landmarks, and the most of landmarks showed less than 1 mm in range. Concerning the accuracy, the mean difference between 3-dimensional CT and manual measurements was 0.33 mm, and 1.13 mm between 3-dimensional cephalogram and manual measurements. The distance between RGo and LGo showed the largest difference (2.03 mm). There were highly significant, and large correlation with manual measurements in both methods (p<0.01). Conclusion: It is concluded that closeness of repeated measures to each skulls reveal the precision of both methods. Computerized tomography and cephalogram for 3-dimensional measurement of maxillofacial structure are equivalent in quality to caliper measurements.