• Journal of the Korean Geotechnical Society
• /
• v.37 no.11
• /
• pp.107-118
• /
• 2021

Scan-to-BIM can be precisely mod eled by measuring structures with Light Detection And Ranging (LiDAR) and build ing a 3D BIM (Building Information Modeling) model based on it, but has a limitation in that it consumes a lot of manpower, time, and cost. To overcome these limitations, studies are being conducted to perform semantic segmentation of 3D point cloud data applying deep learning algorithms, but studies on how segmentation result changes depending on learning data are insufficient. In this study, a parametric study was conducted to determine how the size and track type of railroad tunnels constituting learning data affect the semantic segmentation of railroad tunnels through deep learning. As a result of the parametric study, the similar size of the tunnels used for learning and testing, the higher segmentation accuracy, and the better results when learning through a double-track tunnel than a single-line tunnel. In addition, when the training data is composed of two or more tunnels, overall accuracy (OA) and mean intersection over union (MIoU) increased by 10% to 50%, it has been confirmed that various configurations of learning data can contribute to efficient learning.

• Journal of Biomedical Engineering Research
• /
• v.38 no.5
• /
• pp.237-241
• /
• 2017

Early detection of the pulmonary nodule is important for diagnosis and treatment of lung cancer. Recently, CT has been used as a screening tool for lung nodule detection. And, it has been reported that computer aided detection(CAD) systems can improve the accuracy of the radiologist in detection nodules on CT scan. The previous study has been proposed a method using Convolutional Neural Network(CNN) in Lung CAD system. But the proposed model has a limitation in accuracy due to its sparse layer structure. Therefore, we propose a Deep Convolutional Neural Network to overcome this limitation. The model proposed in this work is consist of 14 layers including 8 convolutional layers and 4 fully connected layers. The CNN model is trained and tested with 61,404 regions-of-interest (ROIs) patches of lung image including 39,760 nodules and 21,644 non-nodules extracted from the Lung Image Database Consortium(LIDC) dataset. We could obtain the classification accuracy of 91.79% with the CNN model presented in this work. To prevent overfitting, we trained the model with Augmented Dataset and regularization term in the cost function. With L1, L2 regularization at Training process, we obtained 92.39%, 92.52% of accuracy respectively. And we obtained 93.52% with data augmentation. In conclusion, we could obtain the accuracy of 93.75% with L2 Regularization and Data Augmentation.

• The Journal of Advanced Prosthodontics
• /
• v.10 no.4
• /
• pp.265-270
• /
• 2018

PURPOSE. The purpose of this in vitro study was to evaluate the accuracy of a new implant impression technique using bite impression coping and a dual arch tray. MATERIALS AND METHODS. Two implant fixtures were placed on maxillary left second premolar and first molar area in dentoform model. The model with two fixtures was used as the reference. The impression was divided into 2 groups, n=10 each. In group 1, heavy/light body silicone impression was made with pick up impression copings and open tray. In group 2, putty/light body silicone impression was made with bite impression copings and dual arch tray. The reference model and the master casts with implant scan bodies were scanned by a laboratory scanner. Surface tessellation language (STL) datasets from test groups was superimposed with STL dataset of reference model using inspection software. The three-dimensional deviation between the reference model and impression models was calculated and illustrated as a color-map. Data was analyzed by independent samples T-test of variance at ${\alpha}=.05$. RESULTS. The mean 3D implant deviations of pick up impression group (group 1) and dual arch impression group (group 2) were 0.029 mm and 0.034 mm, respectively. The difference in 3D deviations between groups 1 and 2 was not statistically significant (P=.075). CONCLUSION. Within limitations of this study, the accuracy of implant impression using a bite impression coping and dual arch tray is comparable to that of conventional pick-up impression.

• 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.

• The Bulletin of The Korean Astronomical Society
• /
• v.37 no.2
• /
• pp.211.1-211.1
• /
• 2012

The on-orbit test simulation for predicting the instrument directional responsivity was conducted by the Monte Carlo based integrated ray tracing (IRT) computation technique and analytic flux-to-signal conversion algorithms. For the on-orbit test simulation, the Sun model consists of the Lambertian scattering sphere and emitting spheroid rays, the Amon-Ra instrument is a two-channel including a broadband scanning radiometer (energy channel) and an imager with ${\pm}2^{\circ}$ FOV (visible channel). The solar radiation produced by the Sun model is directed to the instrument viewing port and traced through the dual channel optical train. The instrument model is rotated on its rotation axis and this gives a slow scan of the Sun model over the full field of view. The direction of the incident lights are fed with scanned images obtained from the visible channel instrument. The instrument responsivity was computed by the ratio of the incident radiation input to the instrument output. In the radiometric simulation, especially, measured BRDF of the 3D CPC was used for scattering effects on radiometry. With diamond turned 3D CPC inner surface, the anisotropic surface scattering model from the measured data was applied to ray tracing computation. The technical details of the on-orbit test simulation are presented together with field-of-view calibration plan.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.31 no.3 s.162
• /
• pp.463-474
• /
• 2007

With rapid expansion in e-retailing of apparel business, personalized fitting model service shows the possibility as the differentiated marketing strategy in cyber shopping. According as necessity of personalized fitting model construction rises, it is tried personalized fitting model creation in several fields such as computer engineering, mechanical engineering, information engineering. But, because existent study was concentrated only on human body modeling, it does not reflect average morphological characteristics of human body properly. In this study, we wish to examine if morphing is fit for expressing characteristic of average human body shape and suggest desirable morphing. We used 3-D scan data of 254 Korean middle aged men collected by Size Korea 2004. The result of this study are as follows: Lower body types were categorized by height hip girth and lower drop(hip girth-navel girth) which were main factors of lower body shape. Then each factor was divided into 3 groups respectively, 30% in the middle, over 30%, under 30%. In 27 groups, the group which belonged to 30% in the middle of height, 30% in the middle of hip girth, 30% in the middle of lower drop was selected as a representative group. We tested geometrical figure by differ volume, tilt, position of point. And we created a representative type of men's lower bodies by morphing the representative group and analyzed it's horizontal, vertical sections. A representative type which was created by morphing reflected a real body and changed realistically at the part of hip, crotch, calf muscle and so on. A cross sections of a representative type were similar to average cross sections of the representative group in size and shape. So it was proved that morphing was successful.

• Journal of radiological science and technology
• /
• v.43 no.6
• /
• pp.503-509
• /
• 2020

This study aims to develop a patient-specific radiation exposure dose prediction model based on anthropometric data that can be easily measurable during CT examination, and to be used as basic data for DRL setting and radiation dose management system in the future. In addition, among the machine learning algorithms, the most suitable model for predicting exposure doses is presented. The data used in this study were chest CT scan data, and a data set was constructed based on the data including the patient's anthropometric data. In the pre-processing and sample selection of the data, out of the total number of samples of 250 samples, only chest CT scans were performed without using a contrast agent, and 110 samples including height and weight variables were extracted. Of the 110 samples extracted, 66% was used as a training set, and the remaining 44% were used as a test set for verification. The exposure dose was predicted through random forest, linear regression analysis, and SVM algorithm using Orange version 3.26.0, an open software as a machine learning algorithm. Results Algorithm model prediction accuracy was R^2 0.840 for random forest, R^2 0.969 for linear regression analysis, and R^2 0.189 for SVM. As a result of verifying the prediction rate of the algorithm model, the random forest is the highest with R^2 0.986 of the random forest, R^2 0.973 of the linear regression analysis, and R^2 of 0.204 of the SVM, indicating that the model has the best predictive power.

• Composites Research
• /
• v.36 no.6
• /
• pp.395-401
• /
• 2023

In this study, finite element modeling of unidirectional composite materials of the computed tomography (CT) was conducted using a supervised learning-based segmentation technique. Firstly, Micro-CT scan was performed to obtain the raw volume of unidirectional composite materials, providing microstructure information. From the CT volume images, actual microstructure of the cross-section of unidirectional composite materials was extracted by the labeling process. Then, a U-net deep learning model was trained with a small number of raw images as inputs and their labeled images as outputs to generate a segmentation model. Subsequently, most of remaining images were input to the trained U-net deep learning model to segment all raw volume for identifying complex microstructure, which was used for the generation of finite element model. Finally, the fiber volume fraction of the finite element model was compared with that of experimentally measured volume to validate the appropriateness of the proposed method.

• The Journal of Advanced Prosthodontics
• /
• v.8 no.3
• /
• pp.219-223
• /
• 2016

PURPOSE. This study aimed to create a digitized database of fabricated obturators to be kept for patients' potential emergency needs. MATERIALS AND METHODS. A chairside intraoral scanner was used to scan the surfaces of an acrylic resin obturator. The scanned data was recorded and saved as a single standard tessellation language file using a three-dimensional modeling software. A simulated obturator model was manufactured using fused deposition modeling technique in a three-dimensional printer. RESULTS. The entire obturator was successfully scanned regardless of its structural complexity, modeled as three-dimensional data, and stored in the digital system of our clinic at a relatively small size (19.6 MB). A simulated obturator model was then accurately manufactured from these data. CONCLUSION. This study provides a proof-of-concept for the use of digital technology to create a digitized database of obturators for edentulous maxillectomy patients.

• Journal of Biomedical Engineering Research
• /
• v.29 no.6
• /
• pp.431-435
• /
• 2008

Quantitative information of a three dimensional(3D) kinematics of joint is very useful in knee joint surgery, understanding how knee kinematics related to joint injury, impairment, surgical treatment, and rehabilitation. In this paper, an automated 2D/3D image matching technique was developed to estimate the 3D in vivo knee kinematics using dual X-ray images. First, a 3D geometric model of the knee was reconstructed from CT scan data. The 3D in vivo position and orientation of femoral and tibial components of the knee joint could be estimated by minimizing the pixel by pixel difference between the projection images from the developed 3D model and the given X-ray images. The accuracy of the developed technique was validated by an experiment with a cubic phantom. The present 2D/3D image matching technique for the estimation of in vivo joint kinematics could be useful for pre-operative planning as well as post-operative evaluation of knee surgery.