• Imaging Science in Dentistry
• /
• v.50 no.2
• /
• pp.169-174
• /
• 2020

Purpose: Periodontal disease causes tooth loss and is associated with cardiovascular diseases, diabetes, and rheumatoid arthritis. The present study proposes using a deep learning-based object detection method to identify periodontally compromised teeth on digital panoramic radiographs. A faster regional convolutional neural network (faster R-CNN) which is a state-of-the-art deep detection network, was adapted from the natural image domain using a small annotated clinical data- set. Materials and Methods: In total, 100 digital panoramic radiographs of periodontally compromised patients were retrospectively collected from our hospital's information system and augmented. The periodontally compromised teeth found in each image were annotated by experts in periodontology to obtain the ground truth. The Keras library, which is written in Python, was used to train and test the model on a single NVidia 1080Ti GPU. The faster R-CNN model used a pretrained ResNet architecture. Results: The average precision rate of 0.81 demonstrated that there was a significant region of overlap between the predicted regions and the ground truth. The average recall rate of 0.80 showed that the periodontally compromised teeth regions generated by the detection method excluded healthiest teeth areas. In addition, the model achieved a sensitivity of 0.84, a specificity of 0.88 and an F-measure of 0.81. Conclusion: The faster R-CNN trained on a limited amount of labeled imaging data performed satisfactorily in detecting periodontally compromised teeth. The application of a faster R-CNN to assist in the detection of periodontally compromised teeth may reduce diagnostic effort by saving assessment time and allowing automated screening documentation.

• Earthquakes and Structures
• /
• v.12 no.1
• /
• pp.67-77
• /
• 2017

In this paper, it is aimed to present a comparative study about the structural behavior of tall buildings consisting of different type of materials such as concrete, steel or timber using finite element analyses and experimental measurements on shaking table. For this purpose, two 1/60 scaled 28 and 30-stories wooden building models with $40{\times}40cm$ and $35{\times}35cm$ ground/floor area and 1.45 m-1.55 m total height are built in laboratory condition. Considering the frequency range, mode shapes, maximum displacements and relative story drifts for structural models as well as acceleration, displacement and weight limits for shaking table, to obtain the typical building response as soon as possible, balsa is selected as a material property, and additional masses are bonded to some floors. Finite element models of the building models are constituted in SAP2000 program. According to the main purposes of earthquake resistant design, three different earthquake records are used to simulate the weak, medium and strong ground motions. The displacement and acceleration time-histories are obtained for all earthquake records at the top of building models. To validate the numerical results, shaking table tests are performed. The selected earthquake records are applied to first mode (lateral) direction, and the responses are recorded by sensitive accelerometers. Comparisons between the numerical and experimental results show that shaking table tests are enough to identify the structural response of wooden buildings. Considering 20%, 10% and 5% damping rations, differences are obtained within the range 4.03-26.16%, 3.91-65.51% and 6.31-66.49% for acceleration, velocity and displacements in Model-1, respectively. Also, these differences are obtained as 0.49-31.15%, 6.03-6.66% and 16.97-66.41% for Model-2, respectively. It is thought that these differences are caused by anisotropic structural characteristic of the material due to changes in directions parallel and perpendicular to fibers, and should be minimized using the model updating procedure.

• Journal of the Korean Geotechnical Society
• /
• v.32 no.12
• /
• pp.45-57
• /
• 2016

The Inclined Earth Retaining Structure (IER) is the structure using an integrated system of both front supports and inclined back supports to increase the stability for excavation. The IER is a structurally stable temporary excavation method using the back supports restraining the lateral displacement of the front supports as stabilizing piles. The back supports connected to the front supports significantly reduce the earth pressure acting on both the front wall and the front supports by distributing it to the back supports in order to increase the structural stability. In this study, mechanical behaviors of IER according to the head connection type using fixed- or hinge-connection were found by performing numerical analysis and laboratory model tests in the sandy ground. The maximum lateral displacement of fixed-connection was 88% of that of hinge-connection in the numerical analysis. The lateral displacement of fixed-connection was 7% of that of hinge-connection in the laboratory model test results. Furthermore, the earth pressure of the fixed-connection was 67% of that of the hinge-connection in the shear-strain analysis results of the model ground.

• Journal of the Korean Geotechnical Society
• /
• v.19 no.4
• /
• pp.223-233
• /
• 2003

Sand compaction pile (SCP) is one of the ground improvement techniques which are being used for not only accelerating consolidation but also increasing bearing capacity of loose sands or soft clay grounds. In this study, laboratory model tests and 3-D finite element analyses were performed to investigate the interaction between sand compaction piles and surrounding soft soils. Based on the results obtained, as the area replacement ratio increases, the stress concentration ratio increases at the pile point, the settlement decreases, and the relative displacement between column and soil also decreases. It is also found that numerical study is illustrated by good comparison with model test results, and the numerical analysis revealed slip effects which could not be specifically identified in the model tests.

• Korean Journal of Remote Sensing
• /
• v.24 no.5
• /
• pp.483-496
• /
• 2008

Qualified ground control points (GCPs) are required to construct a digital elevation model (DEM) from a pushbroom stereo pair. An inverse geolocation algorithm for extracting GCPs from ERS SAR data and the SRTM DEM was recently developed. However, not all GCPs established by this method are accurate enough for direct application to the geometric correction of pushbroom images such as SPOT, IRS, etc, and thus a method for selecting and removing inaccurate points from the sets of GCPs is needed. In this study, we propose a method for evaluating GCP accuracy and winnowing sets of GCPs through orientation modeling of pushbroom image and validate performance of this method using SPOT stereo pair of Daejon City. It has been found that the statistical distribution of GCP positional errors is approximately Gaussian without bias, and that the residual errors estimated by orientation modeling have a linear relationship with the positional errors. Inaccurate GCPs have large positional errors and can be iteratively eliminated by thresholding the residual errors. Forty-one GCPs were initially extracted for the test, with mean the positional error values of 25.6m, 2.5m and -6.1m in the X-, Y- and Z-directions, respectively, and standard deviations of 62.4m, 37.6m and 15.0m. Twenty-one GCPs were eliminated by the proposed method, resulting in the standard deviations of the positional errors of the 20 final GCPs being reduced to 13.9m, 8.5m and 7.5m in the X-, Y- and Z-directions, respectively. Orientation modeling of the SPOT stereo pair was performed using the 20 GCPs, and the model was checked against 15 map-based points. The root mean square errors (RMSEs) of the model were 10.4m, 7.1m and 12.1m in X-, Y- and Z-directions, respectively. A SPOT DEM with a 20m ground resolution was successfully constructed using a automatic matching procedure.

• Journal of the Korean Geotechnical Society
• /
• v.35 no.2
• /
• pp.37-51
• /
• 2019

This study focuses on the numerical simulations of the cone penetration tests in a sand ground. The mechanical responses of sand were described using the modified Mohr Coulomb plasticity model based on the critical state soil mechanics. In the plasticity model, the dilatancy angle was not a constant, but a function of the distance to the critical state line from the current state of void ratio and mean effective stress. To simulate cone penetration tests numerically, this study relied on Lagrangian finite element method under the axisymmetric condition. To enable penetration of the cone penetrometer without tearing elements along the symmetric axis, the penetration guide concept was adopted in this study. The results of numerical simulations on the calibration chamber cone penetration tests had good agreement with the experimental results.

• Journal of the Korean Geosynthetics Society
• /
• v.4 no.2
• /
• pp.11-18
• /
• 2005

A pilot scale filed model test and 2-D numerical analysis was conducted to evaluate the effectiveness of constructing a geogrid-reinforced and pile-supported embankment system over soft ground to reduce differential settlement, and the results are presented hearin. Three-by-three pile groups with varying the space between pile were driven into a layer of soft marine clay and a layer of geogrid was used as reinforcement over each pile group. 2-D numerical analysis has been conducted by using the FLAC-2D(Fast Lagrangian Analysis of Continua) program for same condition of field model test. The settlement, vertical stress, and strain of geogrid due to the construction of embankment were measured at various locations. Based on the field model test and numerical analysis results, pile reinforcement generated the soil arching at the midspan of pile cap and the geogrid reinforcement helps reduce the differential settlement of the soft ground by tensile strength of geogrid. Also for $D/b{\geq}6.0$, the effectiveness of geogrid reinforcement in reducing settlement is negligible.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.397-418
• /
• 2019

It is very important to understand and analyze the interactive behaviour between ground and adjacent structures due to tunneling. With many technological advancement in modern society, numerous methods for analyzing the interactive behaviour are used in a wide range of civil engineering fields. Close range photogrammetry is mainly being used in the field of geotechnical engineering and research on measuring methods associated with GeoPIV has been currently increased. Originally, the close range photogrammetry using target points and aluminum rods for VMS (Vision Measurement System) program has been used. However, applying this has a problem that external errors can be occurred because the target points are artificially installed by hand, and if the grid between points is being wider or narrower, deficient data can be obtained. Therefore, in this study, MATLAB-based No-target program that can analyze displacement without using target was developed. Additionally, this study focused on comparison and verification with existing program through numerical analysis and laboratory model test. Three cases of Greenfield condition, Strip foundation, and Pile foundation were analyzed. From results of VMS program and No-target program, the error rate and reliability of the total displacement and the vertical displacement were analyzed. It was also compared and verified through the finite element numerical program, PLAXIS.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.17 no.3
• /
• pp.189-204
• /
• 2015

This paper presents the results of reduced scale model tests on the effect of fault zone characteristics on the tunnel deformation behavior. A series of model tests were carried out on deep tunnels considering different fault zone orientations and offset distance. The tunnelling process was simulated in the model tests using compressed air technique. During the tests, the tunnel and ground deformation were mainly monitored while reducing the pressure inside the tunnel and the relationship between the pressure level and the tunnel deformation were established. The results indicate that for a given offset distance the tunnel behavior is influenced the most when the fault zone dips vertically while smallest influence occurs when the fault zone dips 45 degrees.

• Smart Media Journal
• /
• v.9 no.2
• /
• pp.22-32
• /
• 2020

The hippocampal volume atrophy is known to be linked with neuro-degenerative disorders and it is also one of the most important early biomarkers for Alzheimer's disease detection. The measurements of hippocampal pure volumes from Magnetic Resonance Imaging (MRI) is a crucial task and state-of-the-art methods require a large amount of time. In addition, the structural brain development is investigated using MRI data, where brain morphometry (e.g. cortical thickness, volume, surface area etc.) study is one of the significant parts of the analysis. In this study, we have proposed a patch-based ensemble model of 3-D convolutional neural network (CNN) to measure the hippocampal pure volume from MRI data. The 3-D patches were extracted from the volumetric MRI scans to train the proposed 3-D CNN models. The trained models are used to construct the ensemble 3-D CNN model and the aggregated model predicts the pure volume in one-step in the test phase. Our approach takes only 5 seconds to estimate the volumes from an MRI scan. The average errors for the proposed ensemble 3-D CNN model are 11.7±8.8 (error%±STD) and 12.5±12.8 (error%±STD) for the left and right hippocampi of 65 test MRI scans, respectively. The quantitative study on the predicted volumes over the ground truth volumes shows that the proposed approach can be used as a proxy.