• Journal of Korean Neurosurgical Society
• /
• 제63권3호
• /
• pp.386-396
• /
• 2020

Objective : To generate synthetic spine magnetic resonance (MR) images from spine computed tomography (CT) using generative adversarial networks (GANs), as well as to determine the similarities between synthesized and real MR images. Methods : GANs were trained to transform spine CT image slices into spine magnetic resonance T2 weighted (MRT2) axial image slices by combining adversarial loss and voxel-wise loss. Experiments were performed using 280 pairs of lumbar spine CT scans and MRT2 images. The MRT2 images were then synthesized from 15 other spine CT scans. To evaluate whether the synthetic MR images were realistic, two radiologists, two spine surgeons, and two residents blindly classified the real and synthetic MRT2 images. Two experienced radiologists then evaluated the similarities between subdivisions of the real and synthetic MRT2 images. Quantitative analysis of the synthetic MRT2 images was performed using the mean absolute error (MAE) and peak signal-to-noise ratio (PSNR). Results : The mean overall similarity of the synthetic MRT2 images evaluated by radiologists was 80.2%. In the blind classification of the real MRT2 images, the failure rate ranged from 0% to 40%. The MAE value of each image ranged from 13.75 to 34.24 pixels (mean, 21.19 pixels), and the PSNR of each image ranged from 61.96 to 68.16 dB (mean, 64.92 dB). Conclusion : This was the first study to apply GANs to synthesize spine MR images from CT images. Despite the small dataset of 280 pairs, the synthetic MR images were relatively well implemented. Synthesis of medical images using GANs is a new paradigm of artificial intelligence application in medical imaging. We expect that synthesis of MR images from spine CT images using GANs will improve the diagnostic usefulness of CT. To better inform the clinical applications of this technique, further studies are needed involving a large dataset, a variety of pathologies, and other MR sequence of the lumbar spine.

• 한국환경복원기술학회지
• /
• 제24권1호
• /
• pp.53-68
• /
• 2021

The study aims to propose a locating method of green space for reducing Particulate Matter (PM) in ambient air in conjunction with its source traces and vulnerable groups. In order to carry out the aims and purposes, a literature review was conducted to derive indicators of vulnerable area to PM. Based on the developed indicators, the vulnerable areas and green spaces creation strategies for each cluster were developed for the case of Seongdong-gu, Seoul. As a result, six indicators for vulnerability analysis were came out including the vulnerable groups (children's facilities, old people's facilities), emission sources (air pollutant emission workplaces, roads), and environmental indicators (particulate matter concentration, NDVI). According to the six selected indicators, the target area was divided into 39 hexagons and analyzed to result the most vulnerable areas to particulate matter. As a result of comprehensive vulnerability analysis, the Seongsu-dong area was found to be the most vulnerable to particulate matter, and 5 clusters were derived through k-means cluster analysis. Cluster 1 was analyzed as areas that most vulnerable to particulate matter as a result of the comprehensive analysis, therefore urgent need to create green spaces to reduce particulate matter. Cluster 2 was areas that mostly belonged to the Han River. Cluster 3 corresponds to the largest number of hexagons, and since many vulnerable groups are distributed, it was analyzed as a cluster that required the creation of a green spaces to reduce particulate matter, focusing on facilities for vulnerable groups. Three hexagons are included in cluster 4, and the cluster has many roads and lacks vegetation in common. Cluster 5 has a lot of green spaces and is generally distributed with fewer vulnerable groups and emission sources; however, it has a high level of particulate matter concentration. In a situation where various green spaces creation projects for reducing particulate are being implemented, it is necessary to consider the vulnerable groups and emission sources and to present green space creation strategies for each space characteristic in order to increase the effectiveness of such projects. Therefore, this study is regarded as meaningful in suggesting a method for selecting a green area for reducing PM.

• 한국방재안전학회논문집
• /
• 제15권4호
• /
• pp.21-30
• /
• 2022

우리나라는 1994년에 발생한 성수대교 붕괴사고를 계기로 ｢시설물의 안전 및 유지관리에 관한 특별법｣이 제정되면서 시설물의 안전점검 및 유지관리 분야에 많은 노력을 기울이고 있다. 이러한 시설물 중 건축물 다음으로 가장 많은 비중을 차지하는 교량은 현재 총33,165개소가 공용 중이며, 2, 3종 교량의 경우 28,150개소로 전체 비중의 약 85%를 차지하고 있다. 하지만 교량의 대부분을 차지하고 있는 2, 3종 교량은 상대적으로 중요도가 높은 1종 교량에 비해 점검진단 등에 관한 유지관리 현황이 미흡한 실정이다. 이러한 이유는 교량의 점검진단 수행 시 점검진단 수행자의 주관적인 판단 및 기록, 체계적으로 수립되지 못한 점검진단 보고서 양식 등에서 나타난다. 따라서 본 연구에서는 기존 점검진단 보고서의 문제점을 수정 및 보완하여 체계적이고 지속적으로 활용 가능한 점검진단 보고서의 개선(안)을 도출하고자 한다.

• 한국구조물진단유지관리공학회 논문집
• /
• 제27권3호
• /
• pp.1-11
• /
• 2023

1990년대 성수대교 붕괴 이후 시설물의 유지관리에 대한 전국민적 관심이 집중되었다. 이에 정부는 1995년 법의 제정을 시작으로 최근까지 다양한 정책을 수립해왔다. 일반적으로 안전점검을 수행하여 시설물의 안전등급을 평가하고 결함에 대한 보수·보강을 실시하여 시설물을 유지·관리하고 있다. 그러나 유지관리 예산은 한정적이기 때문에 모든 결함에 대한 보수·보강 사업을 실시할 수 없는 실정이다. 이에 보수·보강 조치의 우선순위를 선정하는 것이 필요하다. 이를 위해 부재의 중요도, 결함의 심각성, 경제성을 고려한 우선순위지수(PI, Priority Index)를 제시하고 있다. 이 지수는 전문가의 주관적인 판단에 의해 영향도를 50~100%범위 안에서 조정할 수 있으며, 일부 특정 부재는 동일한 가중치가 설정되어 있다. 또한 결정적으로 보수보강을 통한 효과를 고려하지 못하고, 대부분 경제성에 의해 우선순위가 결정되는 한계점이 발생한다. 따라서, 본 연구에서는 몇 가지 사례분석을 통해 현재 우선순위지수에 대한 문제점을 고찰하고 이를 개선할 수 있는 우선순위 산정식(PI)을 제시하였다.