• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.7
• /
• pp.233-240
• /
• 2022

Most of the recent AI researches has focused on developing AI models. However, recently, artificial intelligence research has gradually changed from model-centric to data-centric, and the importance of learning data is getting a lot of attention based on this trend. However, it takes a lot of time and effort because the preparation of learning data takes up a significant part of the entire process, and the generation of labeling data also differs depending on the purpose of development. Therefore, it is need to develop a tool with various labeling functions to solve the existing unmetneeds. In this paper, we describe a labeling system for creating precise and fast labeling data of medical images. To implement this, a semi-automatic method using Back Projection, Grabcut techniques and an automatic method predicted through a machine learning model were implemented. We not only showed the advantage of running time for the generation of labeling data of the proposed system, but also showed superiority through comparative evaluation of accuracy. In addition, by analyzing the image data set of about 1,000 patients, meaningful diagnostic indexes were presented for men and women in the diagnosis of sarcopenia.

• Journal of the Korean Society of Radiology
• /
• v.6 no.5
• /
• pp.357-363
• /
• 2012

The purpose of this study on head computed tomography scan corporate reorganization adaptive iteration algorithm using the statistical noise, and quality assessment, reduction of dose was evaluated. Head CT examinations do not apply ASIR group [A group], ASIR 50 applies a group [B group] were divided into examinations. B group of each 46.9 %, 48.2 %, 43.2 %, and 47.9 % the measured in the phantom research result of measurement of CT noise average were reduced more than A group in the central part (A) and peripheral unit (B, C, D). CT number was measured with the quantitive analytical method in the display-image quality evaluation and about noise was analyze. There was A group and difference which the image noise notes statistically between B. And A group was high so that the image noise could note than B group (31.87 HUs, 31.78 HUs, 26.6 HUs, 30.42 HU P<0.05). The score of the observer 1 of A group evaluated 73.17 on 74.2 at the result 80 half tone dot of evaluating by the qualitative evaluation method of the image by the bean curd clinical image evaluation table. And the score of the observer 1 of B group evaluated 71.77 on 72.47. There was no difference (P>0.05) noted statistically. And the inappropriate image was shown to the diagnosis. As to the exposure dose, by examination by applying ASIR 50 % there was no decline in quality of the image, 47.6 % could reduce the radiation dose. In conclusion, if ASIR is applied to the clinical part, it is considered with the dose written much more that examination is possible. And when examination, it is considered that it becomes the positive factor when the examiner determines.

• The Korean Journal of Nuclear Medicine
• /
• v.36 no.6
• /
• pp.344-356
• /
• 2002

Purpose: RBC blood pool SPECT has been used to diagnose focal liver lesion such as hemangioma owing to its high specificity. However, low spatial resolution is a major limitation of this modality. Recently, ordered subset expectation maximization (OSEM) has been introduced to obtain tomographic images for clinical application. We compared this new modified iterative reconstruction method, OSEM with conventional filtered back projection (FBP) in imaging of liver hemangioma. Materials and Methods: Sixty four projection data were acquired using dual head gamma camera in 28 lesions of 24 patients with cavernous hemangioma of liver and these raw data were transferred to LINUX based personal computer. After the replacement of header file as interfile, OSEM was performed under various conditions of subsets (1,2,4,8,16, and 32) and iteration numbers (1,2,4,8, and 16) to obtain the best setting for liver imaging. The best condition for imaging in our investigation was considered to be 4 iterations and 16 subsets. After then, all the images were processed by both FBP and OSEM. Three experts reviewed these images without any information. Results: According to blind review of 28 lesions, OSEM images revealed at least same or better image quality than those of FBP in nearly all cases. Although there showed no significant difference in detection of large lesions more than 3 cm, 5 lesions with 1.5 to 3 cm in diameter were detected by OSEM only. However, both techniques failed to depict 4 cases of small lesions less than 1.5 cm. Conclusion: OSEM revealed better contrast and define in depiction of liver hemangioma as well as higher sensitivity in detection of small lesions. Furthermore this reconstruction method dose not require high performance computer system or long reconstruction time, therefore OSEM is supposed to be good method that can be applied to RBC blood pool SPECT for the diagnosis of liver hemangioma.

• Journal of the Korean Society of Radiology
• /
• v.12 no.7
• /
• pp.813-819
• /
• 2018

To demonstrate the 3D usefulness of MDCT, a 73-year-old male patient with subclavian thrombosis was obtained 3D images of maximum intensity projection (MIP), volume rendering, and multiplanar reformation (MPR) to clearly detect and locate the subclavian artery. The data will be provided to the patient for diagnosis and treatment. The scan data were acquired as 3D CT images MIP, volume rendering, curved MPR, and virtual endoscopy images. In the 3D program, the ascending aorta was measured as 364.28 HU, the left carotid artery was 413.77 HU, and the left subclavian artery was 15.72 HU. MIP coronal image shows the closure of the subclavian artery in the left side. Three-dimensional volume images were obtained with 100% permeability and 87-1265 HU. The coronal curved MPR and sagittal curved MPR images show the closure of the subclavian artery due to thrombus using 3D image processing. In the case of subclavian arterial occlusion due to thrombosis, the patient is scanned with MDCT and 3D image processing can be used to confirm occlusion of subclavian artery.

• Korean Journal of Head & Neck Oncology
• /
• v.29 no.2
• /
• pp.75-78
• /
• 2013

영아의 설근부에 발생하는 악성 종양은 매우 드물다. 저자들은 연하곤란과 폐쇄성 수면 무호흡증으로 의뢰된 17개월 남아의 설근부에 발생한 횡문근육종을 경험하였기에 문헌고찰과 함께 보고하는 바이다. 외래에서 실시한 이학적 검사 상 설근부 전체가 돌출되어 있었으며 단단한 종괴가 촉지 되었다. 조직검사와 기관절개술을 시행하였으며 컴퓨터단층촬영, 양전자 방출 단층 촬영 그리고 뼈 스캔과 함께 염색체 분석을 실시하였다. 조직검사 결과 폐포성 횡문근육종이 확인되었으며 전이의 증거는 없었다. 염색체 분석상 폐포성 횡문근육종에 상응하는 PAX7-FKHR 유전자 전좌가 발견되었다. 8 회의 항암화학요법과 방사선 치료 후 촉지되는 설근부의 종괴는 없었으며 환자가 호소하는 증상도 개선되었다. 추적 관찰 시 시행된 자기공명영상 결과 확연한 종괴 크기의 감소를 확인할 수 있었다. 횡문근육종은 매우 드문 악성 종양으로 수술과 함께 항암화학요법, 방사선치료 등 여러 치료 방법이 동원되지만 전이나 재발이 있을 시 예후는 매우 불량하다. 그러므로 영아에서 연하곤란, 호흡 곤란 등의 증상이 있을 시에는 설근부를 포함한 상부호흡소화관을 적극적으로 검사하여야 하며 악성 종양의 가능성을 염두에 두어야 하겠다.

• The Journal of Korean Institute of Information Technology
• /
• v.16 no.11
• /
• pp.35-41
• /
• 2018

Electrical impedance tomography (EIT) is a kind of nondestructive testing technique that obtains the internal resistivity distribution from the voltages measured at the electrodes located outside the area of interest. However, an image reconstruction problem in EIT has innate non-linearity and ill-posedness, so that it is difficult to obtain satisfactory reconstructed results. In general, a neural network can efficiently model the input and output relationships of a non-linear system. This paper proposes a method for estimating the position and size of a circular target using a multi-layer neural network. To verify the performance of the proposed method, neural network was trained and various computer simulations were performed and satisfactory performance was verified.

• Applied Chemistry for Engineering
• /
• v.31 no.6
• /
• pp.585-590
• /
• 2020

Successful early diagnosis of cancer diseases such as lung, prostate, liver and adrenocortical carcinoma is a key step in determining the cost of treatment, survival rate, and cure rate. Most of current cancer diagnosis systems including biopsy, computed tomography (CT), positron emission tomography (PET)-CT, magnetic resonance imaging (MRI), ultrasonography, etc., require expensive and complicated equipment with highly trained human resources. Global medical and scientific communities have thus made numerous efforts on developing effective but rapid disease management system via introducing a wide spectrum of point-of-care medical diagnosis devices. Among them, a lateral flow immunoassay strip technique has gained a great attention due to many advantages such as cost-effectiveness, short inspection time, and user friendly accessibility. In this mini-review, we will highlight recent research trend on the development of colorimetry based LFIA strips for cancer diagnosis and discuss the future research direction and potential applications.

• Journal of the Korean Arthroscopy Society
• /
• v.12 no.1
• /
• pp.40-44
• /
• 2008

Purpose: The purpose of this study was to report the real geometry of Resident's ridge doing in anterior cruciate ligament reconstruction Materials and Methods: From Jan 2007 to Aug 2007, 48 cases which had normal distal femoral condyle analyzed with Multidetector-Row Computed Tomography. Resident's ridge was defined as change of height above 1 mm in lateral wall of intercondylar notch. Anterior-posterior length of intercondylar notch, length and height of Resident's ridge, distance of Resident's ridge from posterior cortex were estimated with 3-D reconstruction using $Lucion^{(R)}$ program. Results: Cases were $59{\pm}16$ years olds and male was 16 cases, female was 32 cases. 9 cases had no Resident's ridge, anterior-posterior length of intercondylar notch was $25.4{\pm}3.5$ mm, average of length and height of the Resident's ridge was $8.2{\pm}2.6,\;3.5{\pm}1.5$ mm. Distance of the Resident's ridge from posterior cortex was $7.6{\pm}2.6$ mm. Conclusion: Resident's ridge was used as landmark in anterior cruciate ligament reconstruction, which presented in many cases and which had distinct length and height.

• Journal of the Korean Orthopaedic Association
• /
• v.54 no.4
• /
• pp.293-301
• /
• 2019

Soft tissue masses of the extremities and torso are a common problem encountered by orthopaedic surgeons. Although these soft tissue masses are often benign, orthopaedic surgeons need to recognize the key features differentiating benign and malignant masses. An understanding of the epidemiology and clinical presentation of soft tissue masses is needed to develop a practical approach for evaluation and surgical management. Size and depth are the two most important factors on which triage decisions should be based. In a differential diagnosis of a tumor, it is important to know the characteristics of the soft tissue mass through detailed history taking and physical examinations before the diagnostic procedures. A variety of imaging studies, such as simple radiography, ultrasound, magnetic resonance imaging, positron emission tomography, computed tomography, bone scan, and angiography can be used to diagnose tumors. Know the ledge of advantages and disadvantages of each imaging study is essential for confirming the characteristics of the tumor that can be observed in the image. In particular, ultrasonography is convenient because it can be performed easily in an outpatient clinic and its cost is lower than other image studies. On the other hand, the accuracy of the test is affected by the skill of the examiner. A biopsy should be performed to confirm the tumor and be performed after all imaging studies have been done but before the final treatment of soft tissue tumors. When a biopsy is to be performed, careful attention to detail with respect to multidisciplinary coordination beforehand, cautious execution of the procedure to minimize complications, and expedient follow-up and referral to a musculoskeletal oncologist when appropriate, are essential.

• Journal of radiological science and technology
• /
• v.34 no.4
• /
• pp.305-314
• /
• 2011

The purpose of this study was to compare the findings on the chest low-dose CT (LDCT) images between the negative and positive groups for pneumoconiosis in the group exposed to inorganic dust. From May 30, 2007 to August 31, 2008, total 328 subjects were examined by a LDCT. LDCT images were read by a chest radiologist who has much experience for reading of pneumoconiosis. All subjects were classified into two groups based on digital images after consensus reading of two radiologists according to the ILO 2000 guidelines; negative group (87, 26.5%) without pneumoconiosis and positive group (241, 73.5%). Statistical analysis was performed using a SPSS 14.0. There were significant differences in age (60.9 vs. 65.0, p<0.001), and in dust expose duration (17.0 vs. 19.2, p=0.024) between two groups, but no significant difference in smoking (p=0.784). Of the 328 subjects, 13 diagnosis were extracted from 245 subjects (74.7%). Coronary artery calcification (CAC) was significantly higher in positive group than that in negative group (36.9% vs. 25.3%, p=0.049). Honeycombing showed higher frequency in positive group than in negative group (6.2% vs. 1.2%, p=0.079). Pneumoconiosis findings caused by inorganic dusts exposure showed the significant relation with CAC on LDCT images. Future studies need to prove that pneumoconiosis finding is independent risk factor for CAC using a coronary artery angiography.