• Journal of the Korean Society of Radiology
• /
• v.16 no.3
• /
• pp.295-302
• /
• 2022

Sparse-view computed tomography (CT) imaging technique is able to reduce radiation dose, ensure the uniformity of image characteristics among projections and suppress noise. However, the reconstructed images obtained by the sparse-view CT imaging technique suffer from severe artifacts, resulting in the distortion of image quality and internal structures. In this study, we proposed a convolutional neural network (CNN) with wavelet transformation and residual learning for reducing artifacts in sparse-view CT image, and the performance of the trained model was quantitatively analyzed. The CNN consisted of wavelet transformation, convolutional and inverse wavelet transformation layers, and input and output images were configured as sparse-view CT images and residual images, respectively. For training the CNN, the loss function was calculated by using mean squared error (MSE), and the Adam function was used as an optimizer. Result images were obtained by subtracting the residual images, which were predicted by the trained model, from sparse-view CT images. The quantitative accuracy of the result images were measured in terms of peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). The results showed that the trained model is able to improve the spatial resolution of the result images as well as reduce artifacts in sparse-view CT images effectively. Also, the trained model increased the PSNR and SSIM by 8.18% and 19.71% in comparison to the imaging model trained without wavelet transformation and residual learning, respectively. Therefore, the imaging model proposed in this study can restore the image quality of sparse-view CT image by reducing artifacts, improving spatial resolution and quantitative accuracy.

• The Journal of Korean Orthopaedic Ultrasound Society
• /
• v.3 no.2
• /
• pp.91-96
• /
• 2010

The artifacts in sonography is not always harmful. Sometimes it is beneficial for the evaluation of the water contents of the soft tissue and estimating the degree of inflammation and character of the regenerated tissue indirectly using artifacts such as acoustic shadowing and the enhanced transmission. It can also be useful to evaluate the possibility of aspiration of the calcifies masses by knowing of the contents of the water among them. Unlike the MRI it is useful to get real time informations with low cost in diagnosis and treatment of the soft tissue disease usinf artifacts in sonography.

• Informatization Policy
• /
• v.30 no.1
• /
• pp.67-88
• /
• 2023

The expansion of the scope of product liability is necessary because the industrial environment has changed following the enactment of the Product Liability Act. Unlike human-coded algorithms, artificial intelligence is black-boxed according to machine learning, and even developers cannot explain the results. In particular, since the cause of the problem by artificial intelligence is unknown, the responsibility is unclear, and compensation for victims is not easy. This is because software or artificial intelligence is a non-object, and its productivity is not recognized under the Product Liability Act, which is limited to movable property. As a desperate measure, productivity may be recognized if it is stored or embedded in the medium. However, it is not reasonable to apply differently depending on the medium. The EU revise the product liability guidelines that recognize product liability when artificial intelligence is included. Although compensation for victims is the value pursued by the Product Liability Act, the essence has been overlooked by focusing on productivity. Even if an accident occurs using an artificial intelligence-adopted service, however, it is desirable to present standards according to practical risks instead of unconditionally holding product responsibility.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.17 no.2
• /
• pp.90-94
• /
• 2013

Purpose: In the PET/CT images, various artifacts cause degradation of the quantitative assessment. Most hotspot generated by radiopharmaceutical injection errors cause an artifact and degrade the quality of the images as well as the accuracy of the quantitative evaluation. The purpose of this study is to assess effectiveness of the elimination of the hotspot at the injection sites using shifting the center of DFOV (Display Field of View, DFOV) method and evaluate the quantitative evaluation of result. Materials and Methods: GE Discovery STE 16 (GE Healthcare, Milwaukee, USA) and 1994 NEMA phantom were used for imaging acquisition. Phantom was filled with 0.005 MBq/mL of $^{18}F-FDG$. A hotspot was artificially placed on the outside of the phantom. The ratio of hotspot area activity to background area activity was regulated as 200:1. After image acquisition with routine protocol, all of the images were reconstructed using the shifting the center of DFOV method that wasn't overlapped with hotspot. Those images obtained before and after applying the shifting reconstruction method were compared. ROIs (Region Of Interests) were set in the hotspot areas, meanSUVs and standard deviations were calculated. Percentage differences were calculated with those meanSUVs and standard deviations. The evaluation on the effects of the shifting reconstruction method was done by comparison of the meanSUVs and the standard deviations, which were calculated for background areas unaffected by hotspot. Results: In the areas of unaffected by hotspot, meanSUVs before and after applying the shifting of center of DFOV method were $0.67{\pm}0.06g/mL$ and $0.65{\pm}0.06g/mL$, respectively. In the artifact areas affected by hotspot, meanSUVs before and after applying the shifting of center of DFOV method were $0.32{\pm}0.08g/mL$ and $0.56{\pm}0.12g/mL$, respectively. The percentage differences of the area adjacent to the hotspot and the area distant from the hotspot were 65.3% and 97.4%, respectively. Conclusion: In the PET/CT images, meanSUV was improved by 32.1% when the effect of artifact was removed with application of the shifting the center of DFOV methode. In other areas unaffected by artifacts, meanSUVs were not significantly different after applying DFOV center shift method. As shown in the result, adverse effects of hotspot made by swelling in the injection site can be reduced by applying DFOV center shift method. Therefore, DFOV center shift method can be applied for the more precise quantitative evaluation, and contribute to the increase of the diagnostic value of the images.

• Proceedings of KOSOMES biannual meeting
• /
• 2005.11a
• /
• pp.133-137
• /
• 2005

To investigated the variation of marine environments due to set up of artificial structure, we carried out field observations. High temperature and salinity waters were distributed clearly in the southeastern part of study area during summer season. The variation of current structure was also occurred around study area where artificial structure set up. In 2005 after set up of artificial structure, the nutrient concentration increased greater than that in 2002 before set up artificial structures. To illustrate the characteristics of marine environment due to set up of artificial structure, quantitative analyses on the effect of artificial structure are important.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2010.10a
• /
• pp.108-109
• /
• 2010

본 연구에서는 비선형예측기법으로서 그 우수성을 인정받고 있는 인공신경망모형을 사용하여 컨테이너 물동량 예측을 수행하였다. 그러나 인공신경망모형을 사용해 시계열의 예측결과를 ARIMA모형과 같이 널리 알려진 다른 전통적인 수요예측기법들과 비교 평가한 과거 연구들을 보게 되면 각기 주장하는 바와 그 결론이 상반됨을 알 수 있다. 그래서 인공신경망의 예측성과를 높이기 위한 기존의 선행연구들의 다양한 시도들을 바탕으로 국내 항만의 컨테이너물동량을 예측하고, 그를 통해 여러 모형간의 비교 검증작업을 수행하였다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.18 no.1
• /
• pp.71-78
• /
• 2005

Rescaled Simulated Annealing(RSA) has been devised for improving the disadvantage of Simulated Annealing(SA) which requires tremendous amount of computation time. RSA and SA have been for optimization of truss and satellite structures and for comparison of results from two algorithms. Ten bar truss structure which has continuous design variables are optimized.. As a practical application, a satellite structure is optimized by the two algorithms. Weights of satellite upper platform and propulsion module are minimized. MSC/NASTRAN is used for the static and dynamic analysis. The optimization results of the RSA are compared with results of the classical SA. The numbers of optimization iterations could be effectively reduced by the RSA.

• Applied Microscopy
• /
• v.35 no.1
• /
• pp.1-13
• /
• 2005

Fine photographs are essential in the electron microscopy. Artifacts can be introduced during all steps of electron microscopy; specimen processing, observation and printing. Every caution is necessary to avoid the artifact formation. In this review, the authors discussed the causes of various artifacts and suggested the solution to help the correct tissue handling and electron microscopic observations.

• Clinics in Shoulder and Elbow
• /
• v.13 no.1
• /
• pp.153-160
• /
• 2010

Purpose: The goal of prosthetic replacement of the shoulder is the restoration of the normal anatomy of the joint. Materials and Methods: The physician should review the variations in normal anatomy because it does vary widely and the placement of the prosthetic needs to be modified to accommodate the variations. Results and Conclusion: Several factors including anatomic, prosthetic and surgical ones can lead to the best clinical results, and these are described.

• Journal of Korea Water Resources Association
• /
• v.41 no.5
• /
• pp.473-482
• /
• 2008

The purpose of this study is to quantitatively estimate the erosional properties for cohesive sediment from Saemankeum artificial lake. A series of erosion tests were conducted with Chonbuk annular flume, which is the first one constructed in this country and verified with validities. Each erosion tests were conducted under a uniform bed condition but a different bed density respectively, and its critical shear stress for erosion(${\tau}_{ce}$) as well as the erosion rate coefficient (${\varepsilon}_M$) were determined quantitatively. Since the erosional properties of the cohesive sediments vary largely depending in the physico-chemical properties, such properties of Saemankeum sediments were also estimated and their effects on the erosional properties were analyzed. For Saemankeum sediments, it can be seen that ${\tau}_{ce}$ increases from $0.26N/m^2$ to $0.52N/m^2$ and ${\varepsilon}_M$ decreases exponentially from $14.28mg/cm^2\;hr$ to $6.02mg/cm^2\;hr$, as the bed density varies from $1.17g/cm^3$ to $1.34g/cm^3$. The erosional parameters of Saemankeum sediments are found to be remarkably different in quantity as compared with those for cohesive sediments from other sites. Particularly, ${\tau}_{ce}$ for Saemankeum sediments is known to be larger than that of Kunsan sediments but similar with that of Shihwa sediments, while ${\varepsilon}_M$ for Saemankeum sediments is shown to be smaller than that for Kunsan sediments.