• Journal of the Korean Society of Radiology
• /
• v.8 no.6
• /
• pp.357-364
• /
• 2014

The filtered back projection in the image reconstruction algorithms for the clinic computed tomography system has been widely used. Noise of the reconstructed image was examined under the input noise for parallel and fan beam geometries. The reconstruction images of $512{\times}512$ size were carried out under 360 and 720 projection by the Visual C++ for parallel beam and fan beam, respectively, and those agreed with the original Shepp-Logan head phantom very much. Noise was generated because of intrinsic restriction (finite number of projections) for the image reconstruction algorithm, filtered back projection, when no input noise was applied. Because the result noise was rapidly increased under 0.5% input noise ratio, technologies for reducing noise in CT system and image processing is important.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.80.4-80
• /
• 2002

Performance of the previously proposed 2D-image reconstruction method for soft tissues in x-ray computed tomography is evaluated thoroughly through numerical experiments with 4 assumed absorption rates of different symmetries under practical conditions, and the following special features are made clear: It is quite precise, especially at points where the object taking larger values; about two orders less magnitude errors than the conventional most precise method when no noise existing, without any 1D- or 2D-interpolation. In spite of its high sensitivity to the noises, it is even more precise by about 8dB than the latter, to relative pojection data noise power of 5%.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.2
• /
• pp.118-126
• /
• 2008

Statistical image reconstruction methods have played an important role in emission computed tomography (ECT) since they accurately model the statistical noise associated with gamma-ray projection data. Although the use of statistical methods in clinical practice in early days was of a difficult problem due to high per-iteration costs and large numbers of iterations, with the development of fast algorithms and dramatically improved speed of computers, it is now inevitably becoming more practical. Some statistical methods are indeed commonly available from nuclear medicine equipment suppliers. In this paper, we first describe a mathematical background for statistical reconstruction methods, which includes assumptions underlying the Poisson statistical model, maximum likelihood and maximum a posteriori approaches, and prior models in the context of a Bayesian framework. We then review a recent progress in developing fast iterative algorithms.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.4
• /
• pp.294-301
• /
• 2016

In this paper, we present methods to efficiently parallelize iterative 3D image reconstruction by exploiting trigeneous devices (three different types of device) at the same time: a CPU, an integrated GPU, and a discrete GPU. We first present a technique that exploits single instruction multiple data (SIMD) architectures in GPUs. Then, we propose a performance estimation model, based on which we can easily find the optimal data partitioning on trigeneous devices. We found that the performance significantly varies by up to 6.23 times, depending on how SIMD units in GPUs are accessed. Then, by using trigeneous devices and the proposed estimation models, we achieve optimal partitioning and throughput, which corresponds to a 9.4% further improvement, compared to discrete GPU-only execution.

• The Journal of the Korea Contents Association
• /
• v.11 no.12
• /
• pp.814-821
• /
• 2011

The purpose of this study was to evaluate of resolution parameter and reconstitution filter in the 256 multi detector computed tomography(MDCT) by using the head phantom. We used 256 MDCT, and head phantom of philips system. We evaluated to image quality by using Extended Brilliance Workspace. The protocol were axial scan method with 120 kVp, 0.5 sec of rotation time, 5 mm of slice thickness and increment, 250 mm of field of view(FOV), $512{\times}512$ of matrix size, 1.0 of pitch, $128{\times}0.625$ mm of collimations. The resolution parameter was applied for 'Standard', 'High' and 'Ultrahigh'. The reconstitution filters were changed to seven type of 'A', 'B', 'C', 'D', 'UA', 'UB', 'UC'. The assesment factors of image quality were the uniformity, the noise, the linearity and 50% and 10% of the modulation transfer function(MTF). Finally The good image quality in 'High' resolution parameter showed at the uniformity, the linearity and 50% and 10% of MTF. The 'UA', 'UB' reconstitution filter showed at the good image quality of the uniformity and the noise and 'C' reconstitution filter showed at the same result of the linearity and 50% and 10% of MTF.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.4
• /
• pp.1714-1720
• /
• 2012

The purpose of the study was to evaluation of the radiation dose reduction and the possibility of the maintainability of the adequate image quality using various automatic exposure control (AEC) systems in multi-detector computed tomography (MDCT). We used three AEC systems for the study: General Electric Healthcare (Auto-mA 3D), Philips Medical systems (DoseRight) and Siemens Medical Solutions (Care Dose 4D). The general scanning protocol was created for the each examination with the same scanning parameters as many as possible. In the various AEC systems, the evaluation of reduced-dose was evaluated by comparing to fixed mAs with using human phantom. The image quality of the phantom was evaluated with measuring the image noise (standard deviation) by insert regions of interests. Finally, when we applied to AEC for three manufacturers, the radiation dose reduction decreased each 35.3% in the Auto-mA 3D, 58.2% in the DoseRight, and 48.6% in the Care Dose 4D. And, there was not statistical significant difference among the image quality in the Strong/Weak of the Care Dose 4D(P=.269). This applies to variety of the AEC systems which will be very useful to reduce the dose and to maintain the high quality.

• Korean Journal of Radiology
• /
• v.22 no.1
• /
• pp.131-138
• /
• 2021

Objective: Iterative reconstruction degrades image quality. Thus, further advances in image reconstruction are necessary to overcome some limitations of this technique in low-dose computed tomography (LDCT) scan of the chest. Deep-learning image reconstruction (DLIR) is a new method used to reduce dose while maintaining image quality. The purposes of this study was to evaluate image quality and noise of LDCT scan images reconstructed with DLIR and compare with those of images reconstructed with the adaptive statistical iterative reconstruction-Veo at a level of 30% (ASiR-V 30%). Materials and Methods: This retrospective study included 58 patients who underwent LDCT scan for lung cancer screening. Datasets were reconstructed with ASiR-V 30% and DLIR at medium and high levels (DLIR-M and DLIR-H, respectively). The objective image signal and noise, which represented mean attenuation value and standard deviation in Hounsfield units for the lungs, mediastinum, liver, and background air, and subjective image contrast, image noise, and conspicuity of structures were evaluated. The differences between CT scan images subjected to ASiR-V 30%, DLIR-M, and DLIR-H were evaluated. Results: Based on the objective analysis, the image signals did not significantly differ among ASiR-V 30%, DLIR-M, and DLIR-H (p = 0.949, 0.737, 0.366, and 0.358 in the lungs, mediastinum, liver, and background air, respectively). However, the noise was significantly lower in DLIR-M and DLIR-H than in ASiR-V 30% (all p < 0.001). DLIR had higher signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) than ASiR-V 30% (p = 0.027, < 0.001, and < 0.001 in the SNR of the lungs, mediastinum, and liver, respectively; all p < 0.001 in the CNR). According to the subjective analysis, DLIR had higher image contrast and lower image noise than ASiR-V 30% (all p < 0.001). DLIR was superior to ASiR-V 30% in identifying the pulmonary arteries and veins, trachea and bronchi, lymph nodes, and pleura and pericardium (all p < 0.001). Conclusion: DLIR significantly reduced the image noise in chest LDCT scan images compared with ASiR-V 30% while maintaining superior image quality.

• Imaging Science in Dentistry
• /
• v.33 no.3
• /
• pp.137-141
• /
• 2003

Purpose: To evaluate the precision of measurements taken of dental implants in bucco-lingually sectioned views of the maxilla by linear tomograms of the panorama and to assess the visibility of the inferior wall of the maxillary sinus. Materials and Methods : Eighty sites prepared with implants of gutta percha cone in the sockets of the upper premolars and molars of 10 dry skulls were radiographically examined using linear tomograms of panorama, and scanned coronally and axially by computed tomography. The differences in mm between the measurements in bucco-lingually sectioned images of maxillary alveolar bone and the true length and width of the implanted gutta percha cones were compared as mean values (mean) and standard deviations (SD) for each radiographic technique. Linear tomography of panorama was compared with computed tomography for visualization of the relationship between the inferior wall of maxillary sinus and the end of each implant. Results: The deviations between the actual implant length and the measured values taken from the linear tomograms (0.44±0.39 mm) was significantly less than the measured values from the multiplanar reconstructed images of the axially scanned computed tomogram (1.21 ± 0.90 mm). There was statistically significant difference (p < 0.05) between two techniques in the differences between the measurements and true implant length. The relationship of the inferior border of maxillary sinus with end of implant was worse identified with the linear tomogram of panorama (68%) than the multiplanar reconstructed image of axially scanned computed tomogram (99%). Conclusion: We could not find any differences in the accuracy of length measurement between the linear tomogram of panorama and computed tomogram, but computed tomogram allowed for a better visualization of the inferior wall of the maxillary sinus than the linear tomogram.

• KIEE International Transactions on Electrophysics and Applications
• /
• v.4C no.3
• /
• pp.123-128
• /
• 2004

In electrical impedance tomography (EIT), the internal resistivity distribution of the unknown object is computed using the boundary voltage data induced by different current patterns using various reconstruction algorithms. This paper presents a new image reconstruction algorithm based on the genetic algorithm (GA) via a two-step approach for the solution of the EIT inverse problem, in particular for the reconstruction of "static" images. The computer simulation for the 32 channels synthetic data shows that the spatial resolution of reconstructed images in the proposed scheme is improved compared to that of the modified Newton-Raphson algorithm at the expense of an increased computational burden.rden.

• Journal of Biomedical Engineering Research
• /
• v.27 no.4
• /
• pp.189-196
• /
• 2006

Micro computed tomography (micro-CT) has been used for in vivo animal study owing to its noninvasive and high spatial resolution capability. However, the sizes of existing detectors for micro-CT systems are too small to obtain whole-body images of a small animal object with $\sim$10 micron resolution and a part of its bones or other organs should be extracted. So, we have introduced the zoom-in micro-tomography technique which can obtain high-resolution images of a local region of an live animal object without extracting samples. In order to verify our zoom-in technique, we performed in vivo animal bone study. We prepared some SD (Sprague-Dawley) rats for making osteoporosis models. They were divided into control and ovariectomized groups. Again, the ovariectomized group is divided into two groups fed with normal food and with calcium-free food. And we took 3D tomographic images of their femurs with 20 micron resolution using our zoom-in tomography technique and observed the bone changes for 12 weeks. We selected ROI (region of interest) of a femur image and applied 2D FDT (fuzzy distance transform) to measure the trabecular bone thickness. The measured results showed obvious bone changes and big differences between control and ovariectomized groups. However, we found that the reliability of the measurement depended on the selection of ROI in a bone image for thickness calculation. So, we extended the method to 3D FDT technique. We selected 3D VOI (volume of interest) in the obtained 3D tomographic images and applied 3D FDT algorithm. The results showed that the 3D technique could give more accurate and reliable measurement.