• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.1
• /
• pp.51-58
• /
• 2023

In this paper, we propose an organ segmentation technique for the automatic extraction of medical diagnostic indicators from X-ray images. In order to calculate diagnostic indicators of heart disease and spinal disease such as VHS(vertebral heart scale) and Cobb angle, it is necessary to accurately segment the thoracic spine, carina, and heart in a chest X-ray image. A deep neural network model in which the high-resolution representation of the image for each layer and the structure converted into a low-resolution feature map are connected in parallel was adopted. This structure enables the relative position information in the image to be effectively reflected in the segmentation process. It is shown that learning performance can be improved by combining the OCR module, in which pixel information and object information are mutually interacted in a multi-step process, and the channel attention module, which allows each channel of the network to be reflected as different weight values. In addition, a method of augmenting learning data is presented in order to provide robust performance against changes in the position, shape, and size of the subject in the X-ray image. The effectiveness of the proposed theory was evaluated through an experiment using 145 human chest X-ray images and 118 animal X-ray images.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.77-89
• /
• 2022

It has been proven through many previous researches that the pretrained language model with a large corpus helps improve performance in various natural language processing tasks. However, there is a limit to building a large-capacity corpus for training in a language environment where resources are scarce. Using the Cross-lingual Post-Training (XPT) method, we analyze the method's efficiency in Korean, which is a low resource language. XPT selectively reuses the English pretrained language model parameters, which is a high resource and uses an adaptation layer to learn the relationship between the two languages. This confirmed that only a small amount of the target language dataset in the relationship extraction shows better performance than the target pretrained language model. In addition, we analyze the characteristics of each model on the Korean language model and the Korean multilingual model disclosed by domestic and foreign researchers and companies.

• Explosives and Blasting
• /
• v.40 no.4
• /
• pp.15-26
• /
• 2022

With the recent increase in old and dangerous buildings, the demand for technology in the field of structure demolition is rapidly increasing. In particular, in the case of structures with severe deformation of damage, there is a risk of deterioration in stability and disaster due to changes in the load distribution characteristics in the structure, so rapid structure demolition technology that can be efficiently dismantled in a short period of time is drawing attention. However, structural deformation such as unauthorized extension or illegal remodeling occurs frequently in many old structures, which is not reflected in structural information such as building drawings, and acts as an obstacle in the demolition design process. In this study, as an effective way to overcome the discrepancy between the structural information of old structures and the actual structure, access to actual structures through 3D modeling was considered. 3D point cloud data inside and outside the building were obtained through LiDAR and drone photography for buildings scheduled to be blasting demolition, and precision matching between the two spatial data groups was performed using an open-source based spatial information construction system. The 3D structure model was completed by importing point cloud data matched with 3D modeling software to create structural drawings for each layer and forming each member along the structure slab, pillar, beam, and ceiling boundary. In addition, the modeling technique proposed in this study was verified by comparing it with the actual measurement value for selected structure member.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.519-524
• /
• 2023

In hashing-based image retrieval, the hash code of a manipulated image is different from the original image, making it difficult to search for the same image. This paper proposes and evaluates a self-supervised deephashing model that generates perceptual hash codes from feature information such as texture, shape, and color of images. The comparison models are autoencoder-based variational inference models, but the encoder is designed with a fully connected layer, convolutional neural network, and transformer modules. The proposed model is a variational inference model that includes a SimAM module of extracting geometric patterns and positional relationships within images. The SimAM module can learn latent vectors highlighting objects or local regions through an energy function using the activation values of neurons and surrounding neurons. The proposed method is a representation learning model that can generate low-dimensional latent vectors from high-dimensional input images, and the latent vectors are binarized into distinguishable hash code. From the experimental results on public datasets such as CIFAR-10, ImageNet, and NUS-WIDE, the proposed model is superior to the comparative model and analyzed to have equivalent performance to the supervised learning-based deephashing model. The proposed model can be used in application systems that require low-dimensional representation of images, such as image search or copyright image determination.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.13 no.3
• /
• pp.295-301
• /
• 2008

Bathymetric data collected using a multi-beam echo sounder during marine scientific survey is essential for geologic and oceanographic research works. Accurate measurment of sound velocity profile(SVP) in water-column is important for bathymetric data processing. SVP can vary at different locations during the survey undertaken for wide areas. In addition, an observational error can occur when different equipments(Sound Velocity Profiler, Conductivity Temperature Depth, eXpendable BathyThermograph) are used for measuring SVP at the same water column. In this study, we used an MB-system software to show changes in bathymetry caused by variation of SVP. The analyses showed that the sound velocity(SV) changes due to the depth and thickness of thermocline had more significant effects on the resulting bathymetric data than those of surface mixed layer. The observational errors between SVP measuring instruments did not cause much differneces in the processed bathymetric data. Bathymetric survey line is better to be established to the direction that the change of temperature can be minimize to reduce the variation of SVP during the data acquisition along the survey line.

• Restorative Dentistry and Endodontics
• /
• v.33 no.1
• /
• pp.9-19
• /
• 2008

The purpose of this study was: (1) to compare nanoleakage patterns of a conventional 3-step etch and rinse adhesive system and two experimental hydrophobic adhesive systems and (2) to investigate the change of the nanoleakage patterns after load cycling. Two kinds of hydrophobic experimental adhesives, ethanol containing adhesive (EA) and methanol containing adhesive (MA), were prepared. Thirty extracted human molars were embedded in resin blocks and occlusal thirds of the crowns were removed. The polished dentin surfaces were etched with a 35 % phosphoric acid etching gel and rinsed with water. Scotchbond Multi-Purpose (MP), EA and MA were used for bonding procedure. Z-250 composite resin was built-up on the adhesive-treated surfaces. Five teeth of each dentin adhesive group were subjected to mechanical load cycling. The teeth were sectioned into 2 mm thick slabs and then stained with 50 % ammoniacal silver nitrate. Ten specimens for each group were examined under scanning electron microscope in backscattering electron mode. All photographs were analyzed using image analysis software. Three regions of each specimen were used for evaluation of the silver uptake within the hybrid layer. The area of silver deposition was calculated and expressed in gray value. Data were statistically analyzed by two-way ANOVA and post-hoc testing of multiple comparisons was done with the Scheffe's test. Silver particles were observed in all the groups. However, silver particles were more sparsely distributed in the EA group and the MA group than in the MP group (p < .0001). There were no changes in nanoleakage patterns after load cycling.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.26 no.1
• /
• pp.1-12
• /
• 2010

The purpose of this study was to compare the strain on the alveolar ridge in the centric, eccentric and protrusive position according to the occlusal scheme (bilateral balanced occlusion with 33 degree anatomical teeth, group B; monoplane occlusion with non-anatomical teeth, group M; lingualized occlusion with 33 degree anatomical teeth and non-anatomical teeth, group L; of complete dentures. Experimental dentures were set bilateral balanced occlusion, lingualized occlusion and monoplane occlusion. They are analysed through T-Scan II(Tekscan, Boston, U.S.A) and 1.5mm thick layer was removed from the denture-supporting surface of resin model and then replaced with silicone to simulate resilient edentulous ridge mucosa. A $4{\times}6$ linear strain gauge is attached to the $1^{st}$ premolar and $1^{st}$ molar area. The strain values are recorded according to the occlusal scheme in the centric, eccentric and protrusive position after uniformly applying 50 N and 150 N force through a Universal Testing Machine(instron$^{(R)}$ 5567, Bluehill 2.0 software ,U.S.A.) with the models mounted in the articulator. When performing centric and protrusive occlusion, the three groups of occlusal scheme were compared in the anterior region and in the posterior region. The strains of each group were also compared in the working side and in the non-working side during eccentric excursion. It was observed that the strain in the bilateral balanced occlusion showed a higher value than the lingualized occlusion and monoplane occlusion in every position except the non-working side. However, during the eccentric movement the strain value in the non-working side showed the lowest value in the bilaterally balanced occlusion. The strain change amount from the working side or centric occlusion to non-working side and also the strain variation rate within the non-working side showed the highest value in bilateral balanced occlusion.

• Proceedings of the KACD Conference
• /
• 2008.05a
• /
• pp.246-257
• /
• 2008

The purpose of this study was to investigate the influence of various occlusal loading sites and directions on the stress distribution of the cervical composite resin restorations of maxillary second premolar, using 3 dimensional (3D) finite element (FE) analysis. Extracted maxillary second premolar was scanned serially with Micro-CT (SkyScan1072; SkyScan, Aartselaar, Belgium). The 3D images were processed by 3D-DOCTOR (Able Software Co., Lexington, MA, USA). HyperMesh (Altair Engineering. Inc., Troy, USA) and ANSYS (Swanson Analysis Systems. Inc., Houston, USA) was used to mesh and analyze 3D FE model. Notch shaped cavity was filled with hybrid (Z100, 3M Dental Products, St. Paul, MN, USA) or flowable resin (Tetric Flow, Viva dent Ets., FL-9494-Schaan, Liechtenstein) and each restoration was simulated with adhesive layer thickness ($40{\mu}m$). A static load of 200 N was applied on the three points of the buccal incline of the palatal cusp and oriented in $20^{\circ}$ increments, from vertical (long axis of the tooth) to oblique $40^{\circ}$ direction towards the buccal. The maximum principal stresses in the occlusal and cervical cavosurface margin and vertical section of buccal surfaces of notch-shaped class V cavity were analyzed using ANSYS. As the angle of loading direction increased, tensile stress increased. Loading site had little effect on it. Under same loading condition. Tetric Flow showed relatively lower stress than Z100 overall, except both point angles. Loading direction and the elastic modulus of restorative material seem to be important factor on the cervical restoration.

• Restorative Dentistry and Endodontics
• /
• v.33 no.3
• /
• pp.246-257
• /
• 2008

The purpose of this study was to investigate the influence of various occlusal loading sites and directions on the stress distribution of the cervical composite resin restorations of maxillary second premolar, using 3 dimensional (3D) finite element (FE) analysis. Extracted maxillary second premolar was scanned serially with Micro-CT (SkyScan1072; SkyScan, Aartselaar, Belgium). The 3D images were processed by 3D-DOCTOR (Able Software Co., Lexington, MA, USA). HyperMesh (Altair Engineering, Inc., Troy, USA) and ANSYS (Swanson Analysis Systems, Inc., Houston, USA) was used to mesh and analyze 3D FE model. Notch shaped cavity was filled with hybrid (Z100, 3M Dental Products, St. Paul, MN, USA) or flowable resin (Tetric Flow, Vivadent Ets., FL-9494-Schaan, Liechtenstein) and each restoration was simulated with adhesive layer thickness ($40{\mu}m$). A static load of 200 N was applied on the three points of the buccal incline of the palatal cusp and oriented in $20^{\circ}$ increments, from vertical (long axis of the tooth) to oblique $40^{\circ}$ direction towards the buccal. The maximum principal stresses in the occlusal and cervical cavosurface margin and vertical section of buccal surfaces of notch-shaped class V cavity were analyzed using ANSYS. As the angle of loading direction increased, tensile stress increased. Loading site had little effect on it. Under same loading condition, Tetric Flow showed relatively lower stress than Z100 overall, except both point angles. Loading direction and the elastic modulus of restorative material seem to be important factor on the cervical restoration.

• The Journal of Korean Academy of Prosthodontics
• /
• v.57 no.2
• /
• pp.110-117
• /
• 2019

Purpose: This study was to evaluate the wear resistance of 3D printed, milled, and conventionally cured provisional resin materials. Materials and methods: Four types of resin materials made with different methods were examined: Stereolithography apparatus (SLA) 3D printed resin (S3P), digital light processing (DLP) 3D printed resin (D3P), milled resin (MIL), conventionally self-cured resin (CON). In the 3D printed resin specimens, the build orientation and layer thickness were set to $0^{\circ}$ and $100{\mu}m$, respectively. The specimens were tested in a 2-axis chewing simulator with the steatite as the antagonist under thermocycling condition (5 kg, 30,000 cycles, 0.8 Hz, $5^{\circ}C/55^{\circ}C$). Wear losses of the specimens were calculated using CAD software and scanning electron microscope (SEM) was used to investigate wear surface of the specimens. Statistical significance was determined using One-way ANOVA and Dunnett T3 analysis (${\alpha}=.05$). Results: Wear losses of the S3P, D3P, and MIL groups significantly smaller than those of the CON group (P < .05). There was no significant difference among S3P, D3P, and MIL group (P > .05). In the SEM observations, in the S3P and D3P groups, vertical cracks were observed in the sliding direction of the antagonist. In the MIL group, there was an overall uniform wear surface, whereas in the CON group, a distinct wear track and numerous bubbles were observed. Conclusion: Within the limits of this study, provisional resin materials made with 3D printing show adequate wear resistance for applications in dentistry.