• 대한방사선기술학회지:방사선기술과학
• /
• 제38권4호
• /
• pp.383-387
• /
• 2015

본 연구에서는 소아 CT검사에서 환자의 크기를 고려한 보정을 위해 검상돌기부위, 장골능부위, 대퇴골두가 처음으로 보이는 부위의 스캔영상과 전체 스캔영역의 중간 부위에서 산정된 크기보정선량을 비교하였다. 각 위치에서 측정한 size specific dose estimates(SSDE)와 $CTDI_{vol}$ 값의 평균오차는 스캔영역 중간에서 산정한 보정선량 값에서 107.63%로 가장 큰 차이를 보였으며 검상돌기 영역에서 산정한 평균오차는 92.91%로 가장 작은 차이를 보였다. 또한 스캔영역의 중간 부위에서 산정한 SSDE와 골반의 장골능 부위에서 산정한 SSDE의 오차는 최대 7.48%의 오차를 나타냈고, 검상돌기 부위에서 산정한 SSDE와는 17.81%, 대퇴골두 부위와는 14.04%의 차이를 나타냈다. 이와 같이 SSDE는 산정부위에 따라 상당한 오차를 나타내는 것을 확인할 수 있다. 따라서 임상에서 SSDE의 산정을 위해서는 환자의 신체적 형태를 파악하고 각 부위에서 SSDE를 산정하여 최대값을 사용하는 것이 방사선 방어의 입장에서 옳을 것으로 생각된다.

• 대한의용생체공학회:의공학회지
• /
• 제11권1호
• /
• pp.71-74
• /
• 1990

In the medical modalities, three-dimensional objects must be reconstructed from the consecutive slices. but the slime separation is usually much greater than the pixel size within an individual slices. In this paper, an interpolation scheme for filling the spare between the shapes in two successive slices is developed. It minimizes the computation involvement in segmentation of 3-D reconst ructlon process as well as more accurately approximates the object than the linear interpolation method.

• Korean Journal of Radiology
• /
• 제21권4호
• /
• pp.450-461
• /
• 2020

Objective: We performed a meta-analysis to evaluate the agreement of cardiac computed tomography (CT) with cardiac magnetic resonance imaging (CMRI) in the assessment of right ventricle (RV) volume and functional parameters. Materials and Methods: PubMed, EMBASE, and Cochrane library were systematically searched for studies that compared CT with CMRI as the reference standard for measurement of the following RV parameters: end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), or ejection fraction (EF). Meta-analytic methods were utilized to determine the pooled weighted bias, limits of agreement (LOA), and correlation coefficient (r) between CT and CMRI. Heterogeneity was also assessed. Subgroup analyses were performed based on the probable factors affecting measurement of RV volume: CT contrast protocol, number of CT slices, CT reconstruction interval, CT volumetry, and segmentation methods. Results: A total of 766 patients from 20 studies were included. Pooled bias and LOA were 3.1 mL (-5.7 to 11.8 mL), 3.6 mL (-4.0 to 11.2 mL), -0.4 mL (5.7 to 5.0 mL), and -1.8% (-5.7 to 2.2%) for EDV, ESV, SV, and EF, respectively. Pooled correlation coefficients were very strong for the RV parameters (r = 0.87-0.93). Heterogeneity was observed in the studies (I² > 50%, p < 0.1). In the subgroup analysis, an RV-dedicated contrast protocol, ≥ 64 CT slices, CT volumetry with the Simpson's method, and inclusion of the papillary muscle and trabeculation had a lower pooled bias and narrower LOA. Conclusion: Cardiac CT accurately measures RV volume and function, with an acceptable range of bias and LOA and strong correlation with CMRI findings. The RV-dedicated CT contrast protocol, ≥ 64 CT slices, and use of the same CT volumetry method as CMRI can improve agreement with CMRI.

• 대한전자공학회논문지SP
• /
• 제41권4호
• /
• pp.51-61
• /
• 2004

치과 분야에서 치아가 개별적으로 조작될 수 있는 3차원 치아 모델은 치과 치료와 시술 시뮬레이션을 위해 중요한 요소이다. CT영상으로부터 이러한 치아 모델을 재구성하기 위해서는 각 치아의 경계를 정확하게 분할할 수 있어야 한다. 그러나 기존의 3차원 재구성 시스템에서 사용되는 임계화 방법은 치아들과 치아와 비슷한 밝기의 치조골이 서로 인접해서 나타나는 CT 영상열에서 효율적이지 못하다. 본 논문에서는 CT영상에서 부드러운 치아 경계를 추출하기 위해 B-spline 곡선 적합을 이용한 치아 분할 방법을 제안한다. 성공적인 적합을 위해서 이전 슬라이스의 분할정보와 적응 최적 임계화 방법을 기반으로 한 초기경계 생성방법을 제안한다. 그리고 적합과정에서 이웃한 유사한 물체에 적합되는 것을 줄일 수 있는 유전자 알고리즘을 이용한 B-spline 적합방법을 제안한다. 평가결과 제안된 알고리즘은 개별치아의 경계를 성공적으로 검출하였으며 이를 이용하여 시술과 치료 과정의 시뮬레이션을 위한 치아의 3차원 모델을 정확하고 부드럽게 생성할 수 있음을 보였다.

• 대한방사선치료학회지
• /
• 제12권1호
• /
• pp.20-25
• /
• 2000

The accuracy of the target localization was evaluated by conventional and spiral CT in stereotactic radiosurgerv. Conventional and spiral CT images were obtained with geometrical phantom, which was designed to produce exact three-dimensional coordinates of several objects within 0.1mm error range. Geometrical phantom was attached by BRW headframe, intermediate head ring, and CT localizer. Twentv-seven slices of conventional CT image were scanned at 3 mm slice thickness. Spiral CT images were scanned at 3 mm slice thickness from the pitch value 1 to 3, and twenty-seven slices of image were obtained per each the pitch value. These CT images were transferred to a treatment planning system(X-knife, Radionics) by ethernet, Three-dimensional coordinates of these images measured from the treatment planning system were compared to known values of geometrical phantom. The mean localization error of the target localization of conventional CT was 1.4mm. In case of spiral CT, the error of the target localization was within 1.6mm from the pitch value 1 to 1.3, but was more than 30mm above the pitch value 1.5. In conclusion, as the localization error of spiral CT was increased in high pitch value compared to conventional CT, the application of spiral CT will be with caution in stereotactic radiosurgery.

• 한국멀티미디어학회논문지
• /
• 제24권3호
• /
• pp.396-405
• /
• 2021

In surgery to remove pancreatic cancer, it is important to figure out the shape of a patient's pancreas. However, previous studies have a limit to detect a pancreas automatically in abdominal CT images, because the pancreas varies in shape, size and location by patient. Therefore, in this paper, we propose a method of learning various shapes of pancreas according to the patients and adjacent slices using Faster R-CNN based on Inception V2, and automatically detecting the pancreas from abdominal CT images. Model training and testing were performed using the NIH Pancreas-CT Dataset, and intensity normalization was applied to all data to improve pancreatic detection accuracy. Additionally, according to the shape of the pancreas, the test dataset was classified into top, middle, and bottom slices to evaluate the model's performance on each data. The results show that the top data's mAP@.50IoU achieved 91.7% and the bottom data's mAP@.50IoU achieved 95.4%, and the highest performance was the middle data's mAP@.50IoU, 98.5%. Thus, we have confirmed that the model can accurately detect the pancreas in CT images.

• 대한의용생체공학회:의공학회지
• /
• 제26권3호
• /
• pp.171-176
• /
• 2005

This paper proposes an automatic liver segmentation method using improved partial histogram threshold (PHT) algorithms. This method removes neighboring abdominal organs regardless of random pixel variation of contrast enhanced CT images. Adaptive multi-modal threshold is first performed to extract a region of interest (ROI). A left PHT (LPHT) algorithm is processed to remove the pancreas, spleen, and left kidney. Then a right PHT (RPHT) algorithm is performed for eliminating the right kidney from the ROI. Finally, binary morphological filtering is processed for removing of unnecessary objects and smoothing of the ROI boundary. Ten CT slices of six patients (60 slices) were selected to evaluate the proposed method. As evaluation measures, an average normalized area and area error rate were used. From the experimental results, the proposed automatic liver segmentation method has strong similarity performance as the MSM by medical Doctor.

• 치과방사선
• /
• 제24권2호
• /
• pp.361-368
• /
• 1994

Radiographic planning is needed for implant placement in order to determine implant length, jaw bone volume, anatomical stucture and so on. Radiographic examination includes conventional radiography, conventional tomography and CT scan. The most accurate mesurement can be obtained from CT scan. For the cross-sectional view of mandible, CT scan reconstruction is generally needed. But the cross-sectional view of mandible can be reformed by personal computer. This study was performed to examine the clinical usefulness of reformed image using personal computer in comparison with CT scan reconstructed image. CT axial slices of 4 mandibles of 4 volunteers were used. Digital imaging system was composed of Macintosh Ⅱ ci computer, high resolution Sony XC-77 CCD camera, Quick Capture frame grabber board and 'NIH Image' program. Seven reconstructed cross-sectional images within CT machine(CT group) were obtained. And seven reformed cross-sectional images(PC group) after digitization of CT axial slices into the personal computer were obtained. PC group was compared with CT group in the objective and subjective aspects. The results were as follow: 1. Measurement of mandibular height & width in both group showed insignificant difference(P>0.05). 2. Subjective assessment of the mandibular canal in both group showed insignificant difference(P>0.05). 3. Image reformation using personal computer could provide panoramic view, which could not be obtained in CT scan reconstruction.

• Tuberculosis and Respiratory Diseases
• /
• 제55권3호
• /
• pp.239-249
• /
• 2003

Multi-detector row CT (MDCT) provides faster speed, longer coverage in conjunction with thin slices, improved spatial resolution, and ability to produce high quality muliplanar and three-dimensional (3D) images. MDCT has revolutionized the non-invasive evaluation of the central airways. Simultaneous display of axial, multiplanar, and 3D images raises precision and accuracy of the radiologic diagnosis of central airway disease. This article introduces central airway imaging with MDCT emphasizing on the emerging role of multiplanar and 3D reconstruction.

• 대한전자공학회논문지SP
• /
• 제41권3호
• /
• pp.163-174
• /
• 2004

치과 분야에서는 치아교정이나 수술 시뮬레이션을 위해서 각 치아를 개별적으로 조작할 수 있는 3차원 치아모델이 필요하다. 치아 CT 영상으로부터 이러한 치아모델의 재구성을 위해서는 각 치아를 이웃한 치아나 치조골로부터 정확하게 분리할 수 있어야 한다. 본 연구에서는 치아 영역을 효과적으로 분리하기 위한 영상정규화 방안과 최적임계화방안을 제안한다. 제안된 방법은 연속적인 CT 영상 슬라이스들에서 치아영역의 형태와 밝기는 점진적으로 변한다는 사실을 근거로 이전 슬라이스에서 추정된 임계치를 이용하여 현 슬라이스의 임시치아경계를 결정하고 이것을 바탕으로 보다 정확한 임계치를 계산한다.