• Korean Journal of Radiology
• /
• v.23 no.7
• /
• pp.720-731
• /
• 2022

Objective: We aimed to develop and test a deep learning algorithm (DLA) for fully automated measurement of the volume and signal intensity (SI) of the liver and spleen using gadoxetic acid-enhanced hepatobiliary phase (HBP)-magnetic resonance imaging (MRI) and to evaluate the clinical utility of DLA-assisted assessment of functional liver capacity. Materials and Methods: The DLA was developed using HBP-MRI data from 1014 patients. Using an independent test dataset (110 internal and 90 external MRI data), the segmentation performance of the DLA was measured using the Dice similarity score (DSS), and the agreement between the DLA and the ground truth for the volume and SI measurements was assessed with a Bland-Altman 95% limit of agreement (LOA). In 276 separate patients (male:female, 191:85; mean age ± standard deviation, 40 ± 15 years) who underwent hepatic resection, we evaluated the correlations between various DLA-based MRI indices, including liver volume normalized by body surface area (LV_BSA), liver-to-spleen SI ratio (LSSR), MRI parameter-adjusted LSSR (aLSSR), LSSR × LV_BSA, and aLSSR × LV_BSA, and the indocyanine green retention rate at 15 minutes (ICG-R15), and determined the diagnostic performance of the DLA-based MRI indices to detect ICG-R15 ≥ 20%. Results: In the test dataset, the mean DSS was 0.977 for liver segmentation and 0.946 for spleen segmentation. The Bland-Altman 95% LOAs were 0.08% ± 3.70% for the liver volume, 0.20% ± 7.89% for the spleen volume, -0.02% ± 1.28% for the liver SI, and -0.01% ± 1.70% for the spleen SI. Among DLA-based MRI indices, aLSSR × LV_BSA showed the strongest correlation with ICG-R15 (r = -0.54, p < 0.001), with area under receiver operating characteristic curve of 0.932 (95% confidence interval, 0.895-0.959) to diagnose ICG-R15 ≥ 20%. Conclusion: Our DLA can accurately measure the volume and SI of the liver and spleen and may be useful for assessing functional liver capacity using gadoxetic acid-enhanced HBP-MRI.

• Korean Journal of Radiology
• /
• v.23 no.3
• /
• pp.343-354
• /
• 2022

Objective: To develop and evaluate a deep learning-based artificial intelligence (AI) model for detecting skull fractures on plain radiographs in children. Materials and Methods: This retrospective multi-center study consisted of a development dataset acquired from two hospitals (n = 149 and 264) and an external test set (n = 95) from a third hospital. Datasets included children with head trauma who underwent both skull radiography and cranial computed tomography (CT). The development dataset was split into training, tuning, and internal test sets in a ratio of 7:1:2. The reference standard for skull fracture was cranial CT. Two radiology residents, a pediatric radiologist, and two emergency physicians participated in a two-session observer study on an external test set with and without AI assistance. We obtained the area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity along with their 95% confidence intervals (CIs). Results: The AI model showed an AUROC of 0.922 (95% CI, 0.842-0.969) in the internal test set and 0.870 (95% CI, 0.785-0.930) in the external test set. The model had a sensitivity of 81.1% (95% CI, 64.8%-92.0%) and specificity of 91.3% (95% CI, 79.2%-97.6%) for the internal test set and 78.9% (95% CI, 54.4%-93.9%) and 88.2% (95% CI, 78.7%-94.4%), respectively, for the external test set. With the model's assistance, significant AUROC improvement was observed in radiology residents (pooled results) and emergency physicians (pooled results) with the difference from reading without AI assistance of 0.094 (95% CI, 0.020-0.168; p = 0.012) and 0.069 (95% CI, 0.002-0.136; p = 0.043), respectively, but not in the pediatric radiologist with the difference of 0.008 (95% CI, -0.074-0.090; p = 0.850). Conclusion: A deep learning-based AI model improved the performance of inexperienced radiologists and emergency physicians in diagnosing pediatric skull fractures on plain radiographs.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.10
• /
• pp.43-57
• /
• 2022

In this paper, we propose a method of analyzing research papers published by researchers belonging to university departments in the field of disaster & safety for the scientometric analysis of the research status in the field of disaster safety. In order to conduct analysis research, the dataset constructed in previous studies was newly improved and utilized. In detail, for research papers of authors belonging to the disaster prevention and safety engineering type department of domestic universities, institution identification, cited journal identification of references, department type classification, disaster safety type classification, researcher major information, KSIC(Korean Standard Industrial Classification) mapping information was reflected in the experimental data. The proposed method has a difference from previous studies in the field of disaster & safety and data set based on related keyword searches. As a result of the analysis, the type and regional distribution of organizations belonging to the department of disaster prevention and safety engineering, the composition of co-authored department types, the researchers' majors, the status of disaster safety types and standard industry classification, the status of citations in academic journals, and major keywords were identified in detail. In addition, various co-occurrence networks were created and visualized for each analysis unit to identify key connections. The research results will be used to identify and recommend major organizations and information by disaster type for the establishment of an intelligent crisis warning system. In order to provide comprehensive and constant analysis information in the future, it is necessary to expand the analysis scope and automate the identification and classification process for data set construction.

• Korean Journal of Radiology
• /
• v.23 no.12
• /
• pp.1251-1259
• /
• 2022

Objective: T1 mapping provides valuable information regarding cardiomyopathies. Manual drawing is time consuming and prone to subjective errors. Therefore, this study aimed to test a DL algorithm for the automated measurement of native T1 and extracellular volume (ECV) fractions in cardiac magnetic resonance (CMR) imaging with a temporally separated dataset. Materials and Methods: CMR images obtained for 95 participants (mean age ± standard deviation, 54.5 ± 15.2 years), including 36 left ventricular hypertrophy (12 hypertrophic cardiomyopathy, 12 Fabry disease, and 12 amyloidosis), 32 dilated cardiomyopathy, and 27 healthy volunteers, were included. A commercial deep learning (DL) algorithm based on 2D U-net (Myomics-T1 software, version 1.0.0) was used for the automated analysis of T1 maps. Four radiologists, as study readers, performed manual analysis. The reference standard was the consensus result of the manual analysis by two additional expert readers. The segmentation performance of the DL algorithm and the correlation and agreement between the automated measurement and the reference standard were assessed. Interobserver agreement among the four radiologists was analyzed. Results: DL successfully segmented the myocardium in 99.3% of slices in the native T1 map and 89.8% of slices in the post-T1 map with Dice similarity coefficients of 0.86 ± 0.05 and 0.74 ± 0.17, respectively. Native T1 and ECV showed strong correlation and agreement between DL and the reference: for T1, r = 0.967 (95% confidence interval [CI], 0.951-0.978) and bias of 9.5 msec (95% limits of agreement [LOA], -23.6-42.6 msec); for ECV, r = 0.987 (95% CI, 0.980-0.991) and bias of 0.7% (95% LOA, -2.8%-4.2%) on per-subject basis. Agreements between DL and each of the four radiologists were excellent (intraclass correlation coefficient [ICC] of 0.98-0.99 for both native T1 and ECV), comparable to the pairwise agreement between the radiologists (ICC of 0.97-1.00 and 0.99-1.00 for native T1 and ECV, respectively). Conclusion: The DL algorithm allowed automated T1 and ECV measurements comparable to those of radiologists.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.346-348
• /
• 2004

In this rarer, we introduce a new Fuzzy Polynomial Neural Networks (FPNNs)-like structure whose neuron is based on the Fuzzy Set-based Fuzzy Inference System (FS-FIS) and is different from that of FPNNs based on the Fuzzy relation-based Fuzzy Inference System (FR-FIS) and discuss the ability of the new FPNNs-like structurenamed Fuzzy Set-based Polynomial Neural Networks (FSPNN). The premise parts of their fuzzy rules are not identical, while the consequent parts of the both Networks (such as FPNN and FSPNN) are identical. This difference results from the angle of a viewpoint of partition of input space of system. In other word, from a point of view of FS-FIS, the input variables are mutually independent under input space of system, while from a viewpoint of FR-FIS they are related each other. In considering the structures of FPNN-like networks such as FPNN and FSPNN, they are almost similar. Therefore they have the same shortcomings as well as the same virtues on structural side. The proposed design procedure for networks' architecture involves the selection of appropriate nodes with specific local characteristics such as the number of input variables, the order of the polynomial that is constant, linear, quadratic, or modified quadratic functions being viewed as the consequent part of fuzzy rules, and a collection of the specific subset of input variables. On the parameter optimization phase, we adopt Information Granulation (IG) based on HCM clustering algorithm and a standard least square method-based learning. Through the consecutive process of such structural and parametric optimization, an optimized and flexible fuzzy neural network is generated in a dynamic fashion. To evaluate the performance of the genetically optimized FSPNN (gFSPNN), the model is experimented with using gas furnace process dataset.

• Journal of Environmental Science International
• /
• v.18 no.5
• /
• pp.489-499
• /
• 2009

In an effort to examine the Regional Atmospheric Modeling System (RAMS ver. 4.3) to the initial meteorological input data, detailed observational data of NOAA satellite SST (Sea Surface Temperature) was employed. The NOAA satellite SST which is currently provided daily as a seven-day mean value with resolution of 0.1 $^{\circ}$ grid spacing was used instead of the climatologically derived monthly mean SST using in RAMS. In addition, the RAMS SST data must be changed new one because it was constructed in 1993. For more realistic initial meteorological fields, the NOAA satellite SST was incorporated into the RAMS-preprocess package named ISentropic Analysis package (ISAN). When the NOAA SST data was imposed to the initial condition of prognostic RAMS model, the resultant performance of near surface atmospheric fields was discussed and compared with that of default option of SST. We got the good results that the new SST data was made in a standard RAMS format and showed the detailed variation of SST. As the modeling grid became smaller, the SST differences of the NOAA SST run and the RAMS SST43 (default) run in diurnal variation were very minor but this research can apply to further study for the realistic SST situation and the development in predicting regional atmospheric field which imply the regional circulation due to differential surface heating between sea and land or climatological phenomenon.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.5
• /
• pp.685-691
• /
• 2008

The selection of classification boundaries in fuzzy rule- based classification systems is an important and difficult problem. So various methods based on learning processes such as neural network, genetic algorithm, and so on have been proposed for it. In a previous study, we pointed out the limitation of the methods and discussed a method for fuzzy partitioning in the overlapped region on feature space in order to overcome the time-consuming when the additional parameters for tuning fuzzy membership functions are necessary. In this paper, we propose a method to determine three types of classification boundaries(i.e., non-overlapping, overlapping, and a boundary point) on the basis of statistical information of the given dataset without learning by extending the method described in the study. Finally, we show the effectiveness of the proposed method through experimental results applied to pattern classification problems using the modified IRIS and standard IRIS datasets.

• Journal of Korean Society on Water Environment
• /
• v.25 no.5
• /
• pp.735-741
• /
• 2009

In order to support the basic information for planning and performing the environment management such as Total Maximum Daily Loads (TMDLs), it is highly recommended to understand the spatial distribution of water quality and runoff data in the unit watersheds. Therefore, in the present study, we applied Self-Organizing Map (SOM) to detect the characteristics of spatial distribution of Biological Oxygen Demand (BOD) concentration and runoff data which have been measured in the Yeongsan, Seomjin, and Tamjin River basins. For the purpose, the input dataset for SOM was constructed with the mean, standard deviation, skewness, and kurtosis values of the respective data measured from the stations of 22-subbasins in the rivers. The results showed that the $4{\times}4$ array structure of SOM was selected by the trial and error method and the best performance was revealed when it classified the stations into three clusters according to the basic statistics. The cluster-1 and 2 were classified primarily by the skewness and kurtosis of runoff data and the cluster-3 including the basic statistics of YB_B, YB_C, and YB_D stations was clearly decomposed by the mean value of BOD concentration showing the worst condition of water quality among the three clusters. Consequently, the methodology based on the SOM proposed in the present study can be considered that it is highly applicable to detect the spatial distribution of BOD concentration and runoff data and it can be used effectively for the further utilization using different water quality items as a data analysis tool.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.433-436
• /
• 2004

In this paper, we introduce a new Fuzzy Polynomial Neural Networks (FPNNS)-like structure whose neuron is based on the Fuzzy Set-based Fuzzy Inference System (FS-FIS) and is different from that of FPNNS based on the Fuzzy relation-based Fuzzy Inference System (FR-FIS) and discuss the ability of the new FPNNS-like structure named Fuzzy Set-based Polynomial Neural Networks (FSPNN). The premise parts of their fuzzy rules are not identical, while the consequent parts of the both Networks (such as FPNN and FSPNN) are identical. This difference results from the angle of a viewpoint of partition of input space of system. In other word, from a point of view of FS-FIS, the input variables are mutually independent under input space of system, while from a viewpoint of FR-FIS they are related each other. The proposed design procedure for networks architecture involves the selection of appropriate nodes with specific local characteristics such as the number of input variables, the order of the polynomial that is constant, linear, quadratic, or modified quadratic functions being viewed as the consequent part of fuzzy rules, and a collection of the specific subset of input variables. On the parameter optimization phase, we adopt Information Granulation (IC) based on HCM clustering algorithm and a standard least square method-based learning. Through the consecutive process of such structural and parametric optimization, an optimized and flexible fuzzy neural network is generated in a dynamic fashion. To evaluate the performance of the genetically optimized FSPNN (gFSPNN), the model is experimented with using the time series dataset of gas furnace process.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.6
• /
• pp.139-144
• /
• 2019

For recent decades, Non-volatile Memory (NVM) technologies have been drawing a high attention both in industry and academia due to its high density and short latency comparable to that of DRAM. However, NVM devices has write endurance problem and thus the current data structures that have been built around DRAM-specific features including unlimited program cycles is inadequate for NVM, reducing the device lifetime significantly. In this paper, we revisit a red-black tree extensively adopted for data indexing across a wide range of applications, and make it to better fit for NVM. Specifically, we observe that the conventional red-black tree wears out the specific location of memory because of its rebalancing operation to ensure fast access time over a whole dataset. However, this rebalancing operation frequently updates the long-lived nodes, which leads to the skewed wear out across the NVM cells. To resolve this problem, we present a new swapping wear-leveling red-black tree that periodically moves data in the worn-out node into the young node. The performance study with real-world traces demonstrates the proposed red-black tree reduces the standard deviation of the write count across nodes by up to 12.5%.