• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.313-315
• /
• 2023

본 논문에서는 ¹⁸F-FDG PET과 CT에서 추출한 영상인자를 이용하여 비소세포폐암의 전이를 예측하는 머신러닝 모델을 생성하였다. ¹⁸F-FDG는 종양의 포도당 대사 시 사용되며 이를 추적하여 환자의 암 세포를 진단하는데 사용되는 의료영상 기법 중 하나이다. PET과 CT 영상에서 추출한 이미지 특징은 종양의 생물학적 특성을 반영하며 해당 ROI로부터 계산되어 정량화된 값이다. 본 연구에서는 환자의 의료영상으로부터 image texture 프절 전이 예측에 있어 유의한 인자인지를 확인하기 위하여 AUC를 계산하고 단변량 분석을 진행하였다. PET과 CT에서 각각 4개(GLRLM_GLNU, SHAPE_Compacity only for 3D ROI, SHAPE_Volume_vx, SHAPE_Volume_mL)와 2개(NGLDM_Busyness, TLG_ml)의 image texture feature를 모델의 생성에 사용하였다. 생성된 각 모델의 성능을 평가하기 위해 accuracy와 AUC를 계산하였으며 그 결과 random forest(RF) 모델의 예측 정확도가 가장 높았다. 추출된 PET과 CT image texture feature를 함께 사용하여 모델을 훈련하였을 때가 각각 따로 사용하였을 때 보다 예측 성능이 개선됨을 확인하였다. 추출된 영상인자가 림프절 전이를 나타내는 바이오마커로서의 가능성을 확인할 수 있었으며 이러한 연구 결과를 바탕으로 개인별 의료 영상을 기반으로 한 비소세포폐암의 치료 전략을 수립할 수 있을 것이라 기대된다.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.1
• /
• pp.131-140
• /
• 2014

TMost of the medical images that are being used today are to confirm the existence of lesions. These kind of medical images can not only be used to confirm lesions but if we could use it for academic studies or for analysis and reference materials for surgical operations, it could offer great help in medical studies. Furthermore, we could reduce the number of errors in treatment through simulations in the surgical operations and treatments. In this paper, the necessary parts regarding the medical image visuals were extracted through the point cloud and by using the information obtained through it, it manufactures the necessary information for academic studies, diagnoses or operations. By designing and realizing the system which makes simulation possible with the manufactured information, the paper attempts to provide a more effective way to approach academic studies or giving appropriate diagnoses in the treatment of patients.

• The Journal of Korean Society for Radiation Therapy
• /
• v.22 no.2
• /
• pp.85-95
• /
• 2010

Purpose: The degradation of an image quality and error of the beam dose calculation can be caused because the metal artifact is generated during the CT simulation of head and neck patient. The usability of the gantry tilt scan for reducing the metal artifact tries to be appraised. Materials and Methods: The inferior $20^{\circ}$ gantry tilt scan was made in order to reduce the metal artifact and $0^{\circ}$ reconstruction image was acquired. The AAPM CT performance Phantom was used in order to compare the CT number of the reconstructed image and Original image. the difference of volume was compared by using the acrylic phantom. The homogeneity of the CT number was evaluated the Intensity volume Histogram (IVH) as in order to evaluate an influence by the metal artifact. A dose was evaluated as the Dose Volume Histogram (DVH). Results: in the comparison of the CT number and volume, the difference showed up less than 0.5%. As to the comparison of IVH, in the gantry tilt scan, influence by an artifact was reduced and the homogeneity of the CT number was improved. The comparison of DVH result reduced the mean dose error of the both sides parotid 0.2~6%. Conclusion: In the Head & Neck radiation therapy, It is difficult and to distinguish tumor and normal tissue and the error of dose is generated by the metal artifact. The delineation of the exact organization was possible if the Gantry tilt scan was used. The CT number homogeneity was improved and the error of dose could be reduced. The Gantry tilt scan confirmed in the Head & Neck radiation therapy to be very useful in the exact radiation therapy.

• Journal of the Korean Society of Radiology
• /
• v.17 no.3
• /
• pp.297-303
• /
• 2023

Pediatric Computed Tomography (CT) examinations can often result in exam failures or the need for frequent retests due to the difficulty of cooperation from young patients. Deep Learning Image Reconstruction (DLIR) methods offer the potential to obtain diagnostically valuable images while reducing the retest rate in CT examinations of pediatric patients with high radiation sensitivity. In this study, we investigated the possibility of applying DLIR to reduce artifacts caused by respiration or motion and obtain clinically useful images in pediatric chest CT examinations. Retrospective analysis was conducted on chest CT examination data of 43 children under the age of 7 from P Hospital in Gyeongsangnam-do. The images reconstructed using Filtered Back Projection (FBP), Adaptive Statistical Iterative Reconstruction (ASIR-50), and the deep learning algorithm TrueFidelity-Middle (TF-M) were compared. Regions of interest (ROI) were drawn on the right ascending aorta (AA) and back muscle (BM) in contrast-enhanced chest images, and noise (standard deviation, SD) was measured using Hounsfield units (HU) in each image. Statistical analysis was performed using SPSS (ver. 22.0), analyzing the mean values of the three measurements with one-way analysis of variance (ANOVA). The results showed that the SD values for AA were FBP=25.65±3.75, ASIR-50=19.08±3.93, and TF-M=17.05±4.45 (F=66.72, p=0.00), while the SD values for BM were FBP=26.64±3.81, ASIR-50=19.19±3.37, and TF-M=19.87±4.25 (F=49.54, p=0.00). Post-hoc tests revealed significant differences among the three groups. DLIR using TF-M demonstrated significantly lower noise values compared to conventional reconstruction methods. Therefore, the application of the deep learning algorithm TrueFidelity-Middle (TF-M) is expected to be clinically valuable in pediatric chest CT examinations by reducing the degradation of image quality caused by respiration or motion.

• The Journal of Korean Academy of Prosthodontics
• /
• v.48 no.4
• /
• pp.294-300
• /
• 2010

Purpose: The template-guided implant surgery offers several advantages over the traditional approach. The purpose of this study was to evaluate the accuracy of coordinate synchronization procedure with 5-axis milling machine for surgical template fabrication by means of reverse engineering through universal CAD software. Materials and methods: The study was performed on ten edentulous models with imbedded gutta percha stoppings which were hidden under silicon gingival form. The platform for synchordination was formed on the bottom side of models and these casts were imaged in Cone beam CT. Vectors of stoppings were extracted and transferred to those of planned implant on virtual planning software. Depth of milling process was set to the level of one half of stoppings and the coordinate of the data was synchronized to the model image. Synchronization of milling coordinate was done by the conversion process for the platform for the synchordination located on the bottom of the model. The models were fixed on the synchordination plate of 5-axis milling machine and drilling was done as the planned vector and depth based on the synchronized data with twist drill of the same diameter as GP stopping. For the 3D rendering and image merging, the impression tray was set on the conbeam CT and pre- and post- CT acquiring was done with the model fixed on the impression body. The accuracy analysis was done with Solidworks (Dassault systems, Concord, USA) by measuring vector of stopping’s top and bottom centers of experimental model through merging and reverse engineering the planned and post-drilling CT image. Correlations among the parameters were tested by means of Pearson correlation coefficient and calculated with SPSS (release 14.0, SPSS Inc. Chicago, USA) ($\alpha$ = 0.05). Results: Due to the declination, GP remnant on upper half of stoppings was observed for every drilled bores. The deviation between planned image and drilled bore that was reverse engineered was 0.31 (0.15 - 0.42) mm at the entrance, 0.36 (0.24 - 0.51) mm at the apex, and angular deviation was 1.62 (0.54 - 2.27)$^{\circ}$. There was positive correlation between the deviation at the entrance and that at the apex (Pearson Correlation Coefficient = 0.904, P = .013). Conclusion: The coordinate synchronization 5-axis milling procedure has adequate accuracy for the production of the guided surgical template.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.541-546
• /
• 2006

3차원 관상동맥을 분석하기 위해서는 혈관의 분기점, 극단점, 혈관의 계층적 구조 관계를 함축적으로 표현하는 것이 매우 중요하다. 본 논문에서는3차원 CT 혈관 조영 영상으로부터 관상동맥의 3차원 골격을 자동으로 추출하는 방법을 개발하였다. 먼저, CT혈관 조영술에 의해 획득된 슬라이스 이미지로부터 3차원 조작 및 수술 시뮬레이션 등을 위하여 혈관의 3차원 표면에 대한 메쉬 모델을 생성한다. 생성된 메쉬 모델이 임의로 변형된 후에도 자동으로 골격을 쉽게 추출할 수 있도록 메쉬 모델을 복셀화하는 단계를 거친다. 이렇게 얻어진 복셀 모델로부터 표면복셀을 결정하고 표면 복셀로부터 객체 복셀까지의 유클리드 거리값를 계산하여 유클리드 거리맵(EDM)을 계산한다. 계산된 EDM 으로부터 객체 복셀이 가지게 되는 최대 내접 구를 계산하여 Discrete Medial Surface을 생성하게 되는데 이것은 골격의 후보가 된다. 골격의 후보집합 복셀에 대하여 Dijkstra 최단 경로 결정 알고리즘을 적용하여 골격을 자동으로 추출하게 된다. 이렇게 추출된 3차원 골격은 관상동맥 수술 시뮬레이션 등의 다양한 형상 분석에 유용하게 사용될 수 있다.

• Journal of the Korean Society of Radiology
• /
• v.13 no.5
• /
• pp.675-681
• /
• 2019

The purpose of this study was to investigate the effect of increasing the amount of 3D volume imaging on the hand, knee, and foot human phantom in CT, After analyzing the data, three - dimensional volumetric images were implemented using MMWP program to evaluate reproducibility. First, the data amount of three human phantoms according to each increment was analyzed. Secondly, the reproducibility evaluation and the measured length were compared. As a result of analyzing the amount of image data for each phantom according to the increment, it was confirmed that the amount of data is reduced to about 1/10 when the increment is set to 1.0 mm as compared with the case where the increment is set to 0.1 mm. In the evaluation of the feasibility, gap was generated from 0.7mm for hand phantom, 0.6mm for knee phantom and foot phantom, and it was confirmed that even when the actual phantom and actual length were compared, the length was much different and the implementation was lowered. As the increment is closer to 1.0mm, the number of images is small and the 3D implementation time is small. Therefore, it is best to determine the increase before the gap of the image is generated and to apply the Increment for preoperative diagnosis. We hope that this study will be an indicator of the accurate increment setting when implementing 3D image through VRT Rendering after CT scan.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.11
• /
• pp.3054-3059
• /
• 2009

Natural black diamonds of single crystal, polycrystalline, and agglomerated roughs become important for their industrial and gem stone application. We performed the conventional gemological tests of thermal diffusion, apparent density, scratch test, and magnification test as well as the advanced tests of Raman spectroscopy, X-ray diffraction test and Lang topography. We conclude that scratch test with SiC paper was the most efficient method in view point of speed and cost. Raman spectroscopy and XRD were useful for identification of diamond while Lang topography offered a good visualization method of the grain structure of polycrystalline black diamond roughs.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2000.04a
• /
• pp.429-432
• /
• 2000

일반사용자들이 쉽게 웹을 통하여 컴퓨터 그래픽스, 이미지(CT, MARI) 컴퓨터공학, 지리정보, 수치적 통계, 데이터분석 데이터를 그래픽으로 표현하고자 한다. VRML Visualization Server 구축하여 웹 기반 Visualizer 설계한다. 궁극적으로는 가시화의 실시간 표현에 있다. 모델링 기법에 있어서는 객체지향 그래픽 라이브러인 Visualization Toolkit을 이용하여 데이터의 Surface, Contour, Plane, Streamline, Probe 등을 구현하며, VRML 파일형태로 변환하여 클라이언트 웹으로 보내지게된다. 클라이언트 측 웹은 자신이 원하는 형태의 가시화 항목들을 선택하면 HTTP 에 의해 Visualization Server로 전송되어지고 Server의 ISAP는 전송되어져온 가시화항목을 가지고 가시화하여 VRML화일 형식으로 클라이언트 측으로 보내게된다.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.25 no.3
• /
• pp.211-224
• /
• 2009

The objective of this study is to examine the condylar surfaces in order to find out the types, the incidence and common occurrence area of the osseous abnormalities of the condyles according to the age, genders and the purpose of CT taking, and to compare those between the groups for TMD diagnosis and the other groups. 3D CT images of 199 patients which were scanned with the $i-CAT^{TM}$ Cone Beam Computed Tomography were collected from Sanbon Dental Hospital of Wonkwang university and the MPR images were transfered to the TMJ mode to be showed serial sagittal images and coronal images. The images were macroscopically examined by three independent observers for the types and incidences of the osseous abnormalities, their common occurrence area and general shapes of the condyles. As a result, type F is most common ever than type N. The common occurrence area in sagittal images is antero-superior and superior area except for type D-C which were showed on postero-superior area commonly. In coronal images, latero-superior and superior area is most common except for type E which were present on mesio-superior and superior area most frequently. The osseous abnormalities of the condyles are more common in TMD diagnosis group except for type D-C, that is type N and type D-C are more common in the other groups. In this study, abnormalities of the condyles are classified into 6 types and it has a common occurrence area each. And TMD diagnosis group shows a tendency to have higher rate for osseous abnormalities except for type D-C.