• Korean Journal of Radiology
• /
• v.21 no.8
• /
• pp.987-997
• /
• 2020

Objective: Measurement of the liver and spleen volumes has clinical implications. Although computed tomography (CT) volumetry is considered to be the most reliable noninvasive method for liver and spleen volume measurement, it has limited application in clinical practice due to its time-consuming segmentation process. We aimed to develop and validate a deep learning algorithm (DLA) for fully automated liver and spleen segmentation using portal venous phase CT images in various liver conditions. Materials and Methods: A DLA for liver and spleen segmentation was trained using a development dataset of portal venous CT images from 813 patients. Performance of the DLA was evaluated in two separate test datasets: dataset-1 which included 150 CT examinations in patients with various liver conditions (i.e., healthy liver, fatty liver, chronic liver disease, cirrhosis, and post-hepatectomy) and dataset-2 which included 50 pairs of CT examinations performed at ours and other institutions. The performance of the DLA was evaluated using the dice similarity score (DSS) for segmentation and Bland-Altman 95% limits of agreement (LOA) for measurement of the volumetric indices, which was compared with that of ground truth manual segmentation. Results: In test dataset-1, the DLA achieved a mean DSS of 0.973 and 0.974 for liver and spleen segmentation, respectively, with no significant difference in DSS across different liver conditions (p = 0.60 and 0.26 for the liver and spleen, respectively). For the measurement of volumetric indices, the Bland-Altman 95% LOA was -0.17 ± 3.07% for liver volume and -0.56 ± 3.78% for spleen volume. In test dataset-2, DLA performance using CT images obtained at outside institutions and our institution was comparable for liver (DSS, 0.982 vs. 0.983; p = 0.28) and spleen (DSS, 0.969 vs. 0.968; p = 0.41) segmentation. Conclusion: The DLA enabled highly accurate segmentation and volume measurement of the liver and spleen using portal venous phase CT images of patients with various liver conditions.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.7
• /
• pp.271-278
• /
• 2021

As the current cyber attacks become more intelligent, the existing Intrusion Detection System is difficult for detecting intelligent attacks that deviate from the existing stored patterns. In an attempt to solve this, a model of a deep learning-based intrusion detection system that analyzes the pattern of intelligent attacks through data learning has emerged. Intrusion detection systems are divided into host-based and network-based depending on the installation location. Unlike network-based intrusion detection systems, host-based intrusion detection systems have the disadvantage of having to observe the inside and outside of the system as a whole. However, it has the advantage of being able to detect intrusions that cannot be detected by a network-based intrusion detection system. Therefore, in this study, we conducted a study on a host-based intrusion detection system. In order to evaluate and improve the performance of the host-based intrusion detection system model, we used the host-based Leipzig Intrusion Detection-Data Set (LID-DS) published in 2018. In the performance evaluation of the model using that data set, in order to confirm the similarity of each data and reconstructed to identify whether it is normal data or abnormal data, 1D vector data is converted to 3D image data. Also, the deep learning model has the drawback of having to re-learn every time a new cyber attack method is seen. In other words, it is not efficient because it takes a long time to learn a large amount of data. To solve this problem, this paper proposes the Siamese Convolutional Neural Network (Siamese-CNN) to use the Few-Shot Learning method that shows excellent performance by learning the little amount of data. Siamese-CNN determines whether the attacks are of the same type by the similarity score of each sample of cyber attacks converted into images. The accuracy was calculated using Few-Shot Learning technique, and the performance of Vanilla Convolutional Neural Network (Vanilla-CNN) and Siamese-CNN was compared to confirm the performance of Siamese-CNN. As a result of measuring Accuracy, Precision, Recall and F1-Score index, it was confirmed that the recall of the Siamese-CNN model proposed in this study was increased by about 6% from the Vanilla-CNN model.

• Journal of the Korean GEO-environmental Society
• /
• v.24 no.2
• /
• pp.5-12
• /
• 2023

In this study, the amount of heaving soil and the heave characteristics of the heaving soil generated at the actual site were quantitatively analyzed through DCM laboratory test construction. By reproducing a series of construction processes of the DCM method in a large-scale soil tank close to the actual site, the amount of heaving soil was predicted and the elevation characteristics such as elevation, diffusion range, diffusion angle and amount of elevation of the heaving soil were evaluated. As a result of the laboratory test construction, the actual elevation in terms of similarity within the DCM improvement section is 0~8.18m, and an average of 3.50m is observed. The actual diffusion range of the heaving soil converted to the similarity ratio is distributed from 28.0 to 38.0m on the left and right sides of the improvement section. The total amount of heaving soil calculated by the SUFFER program based on the results of the laboratory test construction is 19,901m³. Compared with the injected slurry amount of 16,992m³, the amount of heave compared to the injected amount is analyzed as 85.4%. The diffusion angle of DCM heaving soil, which analyzed the results of DCM laboratory test construction with the SUFFER program, is measured to be 30.0~38.0° at a depth of 50.0m, and is evaluated as an average of 34.0°. On the other hand, based on the DCM laboratory test construction and the analysis results using the program performed in this study, the amount of heaving soil at the DCM depths of 40.0m and 60.0m is predicted.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.52 no.4
• /
• pp.372-385
• /
• 2016

This present study investigated characteristics of distribution and community structure of macrobenthic invertebrates through the survey of commercial Danish seine fisheries from 2011 to 2013. In this study, a total of 28 species were sampled with a mean density of $32,568ind./km^2$ and mean biomass of $1,649.5kg/km^2$. The dominant species, comprising over 1.0% of the total number of individuals, were Chionoecetes opilio ($11,203ind./km^2$, 34.4%), Pandalus eous ($9,247ind./km^2$, 28.4%), Ophiuridae spp. ($5,750ind./km^2$, 17.7%), Argis lar ($2,631ind./km^2$, 8.1%), Neocrangon communis ($994ind./km^2$, 3.1%), Berryteuthis magister ($612ind./km^2$, 1.9%), Sepiola birostrata ($499ind./km^2$, 1.5%) and Strongylocentrotidae sp. ($424ind./km^2$, 1.3%). The dominant species, in terms of biomass, comprising over 1.0% of the total biomass, were C. opilio ($1,167.2kg/km^2$, 70.8%), B. magister ($130.3kg/km^2$, 7.9%), P. eous ($102.4kg/km^2$, 6.2%), Ophiuridae spp. ($84.6kg/km^2$, 5.1%), Enteroctopus dofleini ($45.5kg/km^2$, 2.8%), A. lar ($35.7kg/km^2$, 2.2%), Strongylocentrotidae sp. ($25.0kg/km^2$, 1.5%) and S. birostrata ($22.1kg/km^2$, 1.3%). Among them, S. birostrata, E. dofleini, Strongylocentrotidae sp. and Ophiuridae spp. were higher abundance and biomass in the shallow water (<200 meters in depth), whereas C. opilio, P. eous, A. lar, N. communis and B. magister were higher in the deep water (301 ~ 500 meters in depth). As the results of cluster analysis and non-metric multidimensional scaling (nMDS) analysis based on the Bray-Curtis similarity of fourth root transformed data for number of species and individuals, the macrobenthic invertebrates community by Danish seine survey was divided into two groups of station in the shallow water (<200 meters in depth, Group A) and the deep water (201 ~ 500 meters in depth, Group B). The major individual-dominant species was S. birostrata, Ophiuridae spp. and immature C. opilio in group A. But Group B was P. eous, A. lar, B. magister and mature C. opilio.

• KSBB Journal
• /
• v.28 no.1
• /
• pp.60-64
• /
• 2013

Two strains were isolated from the traditional Deep Blue Sediment Dye of Polygoum tinctoria, Niram, and temporarily named Niram A and Niram B, respectively. The phylogenetic analysis revealed that strain Niram A and B were closely related to the members of the genus Comamonas and Microbacterium, respectively. Strain Niram A exhibited the highest 16S rRNA gene sequence similarity to C. aquatica LMG $2370^T$ (98.06%). Strain Niram B showed 100% homology with M. oxydans DSM 20578T and M. maritypicum DSM $12512^T$. The growth of the strain Niram A and B was not inhibited in Niram medium containing high calcium concentration without free sugar as carbon source. The reducing Niram is greenish. Therefore, the reducing ability on the Niram of the strains Niram A and B were determined with the color difference of the $a^*$ values of Niram fermented-fluids. The $a^*$ value indicates the level of redness (positive value) or greenness (negative value). The green color is increasing towards the negative value. In all samples fermented for 10 days, the $a^*$ values among samples were no significant difference. However, samples fermented for 15 days have an appreciable change. After fermentation for 15 days, the control Niram sample had $-3.96{\pm}0.02$ of the $a^*$ value. On the other hand, the Niram samples fermented with the strain Niram A and B showed $-4.20{\pm}0.02$ of the $a^*$ value and $-7.86{\pm}0.03$ of the $a^*$ value, respectively. In the reducing ability on the Niram, the strain Niram B was significantly better than the strain Niram A.

• International journal of advanced smart convergence
• /
• v.11 no.1
• /
• pp.19-27
• /
• 2022

Across the world, 'housing' comprises a significant portion of wealth and assets. For this reason, fluctuations in real estate prices are highly sensitive issues to individual households. In Korea, housing prices have steadily increased over the years, and thus many Koreans view the real estate market as an effective channel for their investments. However, if one purchases a real estate property for the purpose of investing, then there are several risks involved when prices begin to fluctuate. The purpose of this study is to design a real estate price 'return rate' prediction model to help mitigate the risks involved with real estate investments and promote reasonable real estate purchases. Various approaches are explored to develop a model capable of predicting real estate prices based on an understanding of the immovability of the real estate market. This study employs the LSTM method, which is based on artificial intelligence and deep learning, to predict real estate prices and validate the model. LSTM networks are based on recurrent neural networks (RNN) but add cell states (which act as a type of conveyer belt) to the hidden states. LSTM networks are able to obtain cell states and hidden states in a recursive manner. Data on the actual trading prices of apartments in autonomous districts between January 2006 and December 2019 are collected from the Actual Trading Price Disclosure System of the Ministry of Land, Infrastructure and Transport (MOLIT). Additionally, basic data on apartments and commercial buildings are collected from the Public Data Portal and Seoul Metropolitan Government's data portal. The collected actual trading price data are scaled to monthly average trading amounts, and each data entry is pre-processed according to address to produce 168 data entries. An LSTM model for return rate prediction is prepared based on a time series dataset where the training period is set as April 2015~August 2017 (29 months), the validation period is set as September 2017~September 2018 (13 months), and the test period is set as December 2018~December 2019 (13 months). The results of the return rate prediction study are as follows. First, the model achieved a prediction similarity level of almost 76%. After collecting time series data and preparing the final prediction model, it was confirmed that 76% of models could be achieved. All in all, the results demonstrate the reliability of the LSTM-based model for return rate prediction.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.10
• /
• pp.29-41
• /
• 2022

Recently, with the development of deep learning technology, a variety of huge models with excellent performance have been devised by pre-training massive amounts of text data. However, in order for such a model to be applied to real-life services, the inference speed must be fast and the amount of computation must be low, so the technology for model compression is attracting attention. Knowledge distillation, a representative model compression, is attracting attention as it can be used in a variety of ways as a method of transferring the knowledge already learned by the teacher model to a relatively small-sized student model. However, knowledge distillation has a limitation in that it is difficult to solve problems with low similarity to previously learned data because only knowledge necessary for solving a given problem is learned in a teacher model and knowledge distillation to a student model is performed from the same point of view. Therefore, we propose a heterogeneous knowledge distillation method in which the teacher model learns a higher-level concept rather than the knowledge required for the task that the student model needs to solve, and the teacher model distills this knowledge to the student model. In addition, through classification experiments on about 18,000 documents, we confirmed that the heterogeneous knowledge distillation method showed superior performance in all aspects of learning efficiency and accuracy compared to the traditional knowledge distillation.

• Korean Journal of Radiology
• /
• v.22 no.6
• /
• pp.983-993
• /
• 2021

Objective: To investigate the image quality of ultralow-dose CT (ULDCT) of the chest reconstructed using a cycle-consistent generative adversarial network (CycleGAN)-based deep learning method in the evaluation of pulmonary tuberculosis. Materials and Methods: Between June 2019 and November 2019, 103 patients (mean age, 40.8 ± 13.6 years; 61 men and 42 women) with pulmonary tuberculosis were prospectively enrolled to undergo standard-dose CT (120 kVp with automated exposure control), followed immediately by ULDCT (80 kVp and 10 mAs). The images of the two successive scans were used to train the CycleGAN framework for image-to-image translation. The denoising efficacy of the CycleGAN algorithm was compared with that of hybrid and model-based iterative reconstruction. Repeated-measures analysis of variance and Wilcoxon signed-rank test were performed to compare the objective measurements and the subjective image quality scores, respectively. Results: With the optimized CycleGAN denoising model, using the ULDCT images as input, the peak signal-to-noise ratio and structural similarity index improved by 2.0 dB and 0.21, respectively. The CycleGAN-generated denoised ULDCT images typically provided satisfactory image quality for optimal visibility of anatomic structures and pathological findings, with a lower level of image noise (mean ± standard deviation [SD], 19.5 ± 3.0 Hounsfield unit [HU]) than that of the hybrid (66.3 ± 10.5 HU, p < 0.001) and a similar noise level to model-based iterative reconstruction (19.6 ± 2.6 HU, p > 0.908). The CycleGAN-generated images showed the highest contrast-to-noise ratios for the pulmonary lesions, followed by the model-based and hybrid iterative reconstruction. The mean effective radiation dose of ULDCT was 0.12 mSv with a mean 93.9% reduction compared to standard-dose CT. Conclusion: The optimized CycleGAN technique may allow the synthesis of diagnostically acceptable images from ULDCT of the chest for the evaluation of pulmonary tuberculosis.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.4
• /
• pp.1721-1737
• /
• 2020

Human apparent age estimation has become a popular research topic and attracted great attention in recent years due to its wide applications, such as personal security and law enforcement. To achieve the goal of age estimation, a large number of methods have been pro-posed, where the models derived through the cumulative attribute coding achieve promised performance by preserving the neighbor-similarity of ages. However, these methods afore-mentioned ignore the geometric structure of extracted facial features. Indeed, the geometric structure of data greatly affects the accuracy of prediction. To this end, we propose an age estimation algorithm through joint feature selection and manifold learning paradigms, so-called Feature-selected and Geometry-preserved Least Square Regression (FGLSR). Based on this, our proposed method, compared with the others, not only preserves the geometry structures within facial representations, but also selects the discriminative features. Moreover, a deep learning extension based FGLSR is proposed later, namely Feature selected and Geometry preserved Neural Network (FGNN). Finally, related experiments are conducted on Morph2 and FG-Net datasets for FGLSR and on Morph2 datasets for FGNN. Experimental results testify our method achieve the best performances.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2003.10a
• /
• pp.277-287
• /
• 2003

Acetolactate synthase (ALS, EC 4.1.3.18; also referred to as acetohydroxy acid synthase) catalyzes the first common step in the biosynthesis of valine, leucine, and isoleucine in microorganisms and plants. Recently X-ray structure of yeast ALS was available. Pair-wise alignment of yeast and tobacco ALS sequences revealed 63% sequence similarity. Using Deep View and automatic modeling on Swiss model server, we have generated reliable models of tobacco ALS based on yeast ALS template with a calculated pair-wise RMSD of 0.86 Angstrom. Functional roles of four residues located on the subunit interface (H142, El43, M350, and R376) were examined by site-directed mutagenesis. Seven mutants were generated and purified, of which three mutants (H142T, M350V, and R376F) were found to be inactivated under various assay conditions. The H142k mutant showed moderately altered kinetic properties. The E143A mutant increased 10-fold in K$_m$ value while other parameters remained unchanged. The M350C mutant was strongly resistant to three tested herbicides, while the R376k mutant can bind with herbicide carder at similar affinity to that of wild type enzyme, as determined by tryptophan quenching study. Except M350V mutant, all other mutants were ate to bind with cofactor FAD. Taken together, it is likely that residues H142 and E143 are located at the active site, while residues M350 and R376 are possibly located at the overlapping region of active site and herbicide binding site of the enzyme. Our data also allows us to hypothesize that the interaction between side chains of residues M350 and R376 are probably essential for the correct conformation of the active site. It remains to be elucidated that, whether the herbicide, upon binding with enzyme, inactivates the enzyme by causing change in the active site allosterically, which is unfavorable for catalytic activity.