• Smart Media Journal
• /
• v.12 no.11
• /
• pp.57-66
• /
• 2023

In this paper, we present a study on an AI-based system utilizing the CCTV system within city buses to predict the intentions of boarding and alighting passengers, with the aim of preventing accidents. The proposed system employs the YOLOv7 Pose model to detect passengers, while utilizing an LSTM model to predict intentions of tracked passengers. The system can be installed on the bus's CCTV terminals, allowing for real-time visual confirmation of passengers' intentions throughout driving. It also provides alerts to the driver, mitigating potential accidents during passenger transitions. Test results show accuracy rates of 0.81 for analyzing boarding intentions and 0.79 for predicting alighting intentions onboard. To ensure real-time performance, we verified that a minimum of 5 frames per second analysis is achievable in a GPU environment. his algorithm enhance the safety of passenger transitions during bus operations. In the future, with improved hardware specifications and abundant data collection, the system's expansion into various safety-related metrics is promising. This algorithm is anticipated to play a pivotal role in ensuring safety when autonomous driving becomes commercialized. Additionally, its applicability could extend to other modes of public transportation, such as subways and all forms of mass transit, contributing to the overall safety of public transportation systems.

• Information Systems Review
• /
• v.25 no.1
• /
• pp.93-110
• /
• 2023

Recently, attempts have been made to convert unstructured text into vectors and to analyze vast amounts of natural language for various purposes. In particular, the demand for analyzing texts in specialized domains is rapidly increasing. Therefore, studies are being conducted to analyze specialized and general-purpose documents simultaneously. To analyze specific terms with general terms, it is necessary to align the embedding space of the specific terms with the embedding space of the general terms. So far, attempts have been made to align the embedding of specific terms into the embedding space of general terms through a transformation matrix or mapping function. However, the linear transformation based on the transformation matrix showed a limitation in that it only works well in a local range. To overcome this limitation, various types of nonlinear vector alignment methods have been recently proposed. We propose a vector alignment model that matches the embedding space of specific terms to the embedding space of general terms through end-to-end learning that simultaneously learns the autoencoder and regression model. As a result of experiments with R&D documents in the "Healthcare" field, we confirmed the proposed methodology showed superior performance in terms of accuracy compared to the traditional model.

• Journal of the Korean Society of Radiology
• /
• v.82 no.1
• /
• pp.116-127
• /
• 2021

Purpose To evaluate the feasibility of texture analysis of gray-scale ultrasound (US) images for staging of hepatic fibrosis. Materials and Methods Altogether, 167 patients who had undergone routine US and laboratory tests for a fibrosis-4 (FIB-4) index were included. Texture parameters were measured using a dedicated in-house software. Regions of interest were placed in five different segments (3, 5, 6, 7, 8) for each patient. The FIB-4 index was used as the reference standard for hepatic fibrosis grade. Comparisons of the texture parameters between different fibrosis groups were performed with the Student's t-test or Mann-Whitney U-test. Diagnostic performance was evaluated by receiver operating curve analysis. Results The study population comprised of patients with no fibrosis (FIB-4 < 1.45, n = 50), mild fibrosis (1.45 ≤ FIB-4 ≤ 2.35, n = 37), moderate fibrosis (2.35 < FIB-4 ≤ 3.25, n = 27), and severe fibrosis (FIB-4 > 3.25, n = 53). Skewness in hepatic segment 5 showed a difference between patients with no fibrosis and mild fibrosis (0.2392 ± 0.3361, 0.4134 ± 0.3004, respectively, p = 0.0109). The area under the curve of skewness for discriminating patients with no fibrosis from those with mild fibrosis was 0.660 (95% confidence interval, 0.551-0.758), with an estimated accuracy, sensitivity, specificity of 64%, 87%, 48%, respectively. Conclusion A significant difference was observed regarding skewness in segment 5 between patients with no fibrosis and patients with mild fibrosis.

• Korean Journal of Radiology
• /
• v.24 no.6
• /
• pp.553-563
• /
• 2023

Objective: Functional magnetic resonance imaging (fMRI) and diffusion tensor imaging-derived tractography (DTI-t) contribute to the localization of language areas, but their accuracy remains controversial. This study aimed to investigate the diagnostic performance of preoperative fMRI and DTI-t obtained with a simultaneous multi-slice technique using intraoperative direct cortical stimulation (DCS) or corticocortical evoked potential (CCEP) as reference standards. Materials and Methods: This prospective study included 26 patients (23-74 years; male:female, 13:13) with tumors in the vicinity of Broca's area who underwent preoperative fMRI and DTI-t. A site-by-site comparison between preoperative (fMRI and DTI-t) and intraoperative language mapping (DCS or CCEP) was performed for 226 cortical sites to calculate the sensitivity and specificity of fMRI and DTI-t for mapping Broca's areas. For sites with positive signals on fMRI or DTI-t, the true-positive rate (TPR) was calculated based on the concordance and discordance between fMRI and DTI-t. Results: Among 226 cortical sites, DCS was performed in 100 sites and CCEP was performed in 166 sites. The specificities of fMRI and DTI-t ranged from 72.4% (63/87) to 96.8% (122/126), respectively. The sensitivities of fMRI (except for verb generation) and DTI-t were 69.2% (9/13) to 92.3% (12/13) with DCS as the reference standard, and 40.0% (16/40) or lower with CCEP as the reference standard. For sites with preoperative fMRI or DTI-t positivity (n = 82), the TPR was high when fMRI and DTI-t were concordant (81.2% and 100% using DCS and CCEP, respectively, as the reference standards) and low when fMRI and DTI-t were discordant (≤ 24.2%). Conclusion: fMRI and DTI-t are sensitive and specific for mapping Broca's area compared with DCS and specific but insensitive compared with CCEP. A site with a positive signal on both fMRI and DTI-t represents a high probability of being an essential language area.

• Korean Journal of Radiology
• /
• v.24 no.5
• /
• pp.454-464
• /
• 2023

Objective: We aimed to investigate current expectations and clinical adoption of artificial intelligence (AI) software among neuroradiologists in Korea. Materials and Methods: In April 2022, a 30-item online survey was conducted by neuroradiologists from the Korean Society of Neuroradiology (KSNR) to assess current user experiences, perceptions, attitudes, and future expectations regarding AI for neuro-applications. Respondents with experience in AI software were further investigated in terms of the number and type of software used, period of use, clinical usefulness, and future scope. Results were compared between respondents with and without experience with AI software through multivariable logistic regression and mediation analyses. Results: The survey was completed by 73 respondents, accounting for 21.9% (73/334) of the KSNR members; 72.6% (53/73) were familiar with AI and 58.9% (43/73) had used AI software, with approximately 86% (37/43) using 1-3 AI software programs and 51.2% (22/43) having up to one year of experience with AI software. Among AI software types, brain volumetry software was the most common (62.8% [27/43]). Although 52.1% (38/73) assumed that AI is currently useful in practice, 86.3% (63/73) expected it to be useful for clinical practice within 10 years. The main expected benefits were reducing the time spent on repetitive tasks (91.8% [67/73]) and improving reading accuracy and reducing errors (72.6% [53/73]). Those who experienced AI software were more familiar with AI (adjusted odds ratio, 7.1 [95% confidence interval, 1.81-27.81]; P = 0.005). More than half of the respondents with AI software experience (55.8% [24/43]) agreed that AI should be included in training curriculums, while almost all (95.3% [41/43]) believed that radiologists should coordinate to improve its performance. Conclusion: A majority of respondents experienced AI software and showed a proactive attitude toward adopting AI in clinical practice, suggesting that AI should be incorporated into training and active participation in AI development should be encouraged.

• Korean Journal of Radiology
• /
• v.24 no.11
• /
• pp.1151-1163
• /
• 2023

Objective: To develop a deep-learning-based bone age prediction model optimized for Korean children and adolescents and evaluate its feasibility by comparing it with a Greulich-Pyle-based deep-learning model. Materials and Methods: A convolutional neural network was trained to predict age according to the bone development shown on a hand radiograph (bone age) using 21036 hand radiographs of Korean children and adolescents without known bone development-affecting diseases/conditions obtained between 1998 and 2019 (median age [interquartile range {IQR}], 9 [7-12] years; male:female, 11794:9242) and their chronological ages as labels (Korean model). We constructed 2 separate external datasets consisting of Korean children and adolescents with healthy bone development (Institution 1: n = 343; median age [IQR], 10 [4-15] years; male: female, 183:160; Institution 2: n = 321; median age [IQR], 9 [5-14] years; male: female, 164:157) to test the model performance. The mean absolute error (MAE), root mean square error (RMSE), and proportions of bone age predictions within 6, 12, 18, and 24 months of the reference age (chronological age) were compared between the Korean model and a commercial model (VUNO Med-BoneAge version 1.1; VUNO) trained with Greulich-Pyle-based age as the label (GP-based model). Results: Compared with the GP-based model, the Korean model showed a lower RMSE (11.2 vs. 13.8 months; P = 0.004) and MAE (8.2 vs. 10.5 months; P = 0.002), a higher proportion of bone age predictions within 18 months of chronological age (88.3% vs. 82.2%; P = 0.031) for Institution 1, and a lower MAE (9.5 vs. 11.0 months; P = 0.022) and higher proportion of bone age predictions within 6 months (44.5% vs. 36.4%; P = 0.044) for Institution 2. Conclusion: The Korean model trained using the chronological ages of Korean children and adolescents without known bone development-affecting diseases/conditions as labels performed better in bone age assessment than the GP-based model in the Korean pediatric population. Further validation is required to confirm its accuracy.

• Korean Journal of Radiology
• /
• v.23 no.4
• /
• pp.466-478
• /
• 2022

Objective: ¹⁸F-fluorodeoxyglucose (FDG) PET/CT is often used for detecting malignancy in patients with newly diagnosed hemophagocytic lymphohistiocytosis (HLH), with acceptable sensitivity but relatively low specificity. The aim of this study was to improve the diagnostic ability of ¹⁸F-FDG PET/CT in identifying malignancy in patients with HLH by combining ¹⁸F-FDG PET/CT and clinical parameters. Materials and Methods: Ninety-seven patients (age ≥ 14 years) with secondary HLH were retrospectively reviewed and divided into the derivation (n = 71) and validation (n = 26) cohorts according to admission time. In the derivation cohort, 22 patients had malignancy-associated HLH (M-HLH) and 49 patients had non-malignancy-associated HLH (NM-HLH). Data on pretreatment ¹⁸F-FDG PET/CT and laboratory results were collected. The variables were analyzed using the Mann-Whitney U test or Pearson's chi-square test, and a nomogram for predicting M-HLH was constructed using multivariable binary logistic regression. The predictors were also ranked using decision-tree analysis. The nomogram and decision tree were validated in the validation cohort (10 patients with M-HLH and 16 patients with NM-HLH). Results: The ratio of the maximal standardized uptake value (SUVmax) of the lymph nodes to that of the mediastinum, the ratio of the SUVmax of bone lesions or bone marrow to that of the mediastinum, and age were selected for constructing the model. The nomogram showed good performance in predicting M-HLH in the validation cohort, with an area under the receiver operating characteristic curve of 0.875 (95% confidence interval, 0.686-0.971). At an appropriate cutoff value, the sensitivity and specificity for identifying M-HLH were 90% (9/10) and 68.8% (11/16), respectively. The decision tree integrating the same variables showed 70% (7/10) sensitivity and 93.8% (15/16) specificity for identifying M-HLH. In comparison, visual analysis of ¹⁸F-FDG PET/CT images demonstrated 100% (10/10) sensitivity and 12.5% (2/16) specificity. Conclusion: ¹⁸F-FDG PET/CT may be a practical technique for identifying M-HLH. The model constructed using ¹⁸F-FDG PET/CT features and age was able to detect malignancy with better accuracy than visual analysis of ¹⁸F-FDG PET/CT images.

• Korean Journal of Radiology
• /
• v.23 no.12
• /
• pp.1281-1289
• /
• 2022

Objective: Radiomic modeling using multiple regions of interest in MRI of the brain to diagnose juvenile myoclonic epilepsy (JME) has not yet been investigated. This study aimed to develop and validate radiomics prediction models to distinguish patients with JME from healthy controls (HCs), and to evaluate the feasibility of a radiomics approach using MRI for diagnosing JME. Materials and Methods: A total of 97 JME patients (25.6 ± 8.5 years; female, 45.5%) and 32 HCs (28.9 ± 11.4 years; female, 50.0%) were randomly split (7:3 ratio) into a training (n = 90) and a test set (n = 39) group. Radiomic features were extracted from 22 regions of interest in the brain using the T1-weighted MRI based on clinical evidence. Predictive models were trained using seven modeling methods, including a light gradient boosting machine, support vector classifier, random forest, logistic regression, extreme gradient boosting, gradient boosting machine, and decision tree, with radiomics features in the training set. The performance of the models was validated and compared to the test set. The model with the highest area under the receiver operating curve (AUROC) was chosen, and important features in the model were identified. Results: The seven tested radiomics models, including light gradient boosting machine, support vector classifier, random forest, logistic regression, extreme gradient boosting, gradient boosting machine, and decision tree, showed AUROC values of 0.817, 0.807, 0.783, 0.779, 0.767, 0.762, and 0.672, respectively. The light gradient boosting machine with the highest AUROC, albeit without statistically significant differences from the other models in pairwise comparisons, had accuracy, precision, recall, and F1 scores of 0.795, 0.818, 0.931, and 0.871, respectively. Radiomic features, including the putamen and ventral diencephalon, were ranked as the most important for suggesting JME. Conclusion: Radiomic models using MRI were able to differentiate JME from HCs.

• Nuclear Engineering and Technology
• /
• v.56 no.3
• /
• pp.803-811
• /
• 2024

Nuclear accidents such as Fukushima Daiichi have highlighted the potential of passive safety systems to replace or complement active safety systems as part of the overall prevention and/or mitigation strategies. In addition, passive systems are key features of Small Modular Reactors (SMRs), for which they are becoming almost unavoidable and are part of the basic design of many reactors available in today's nuclear market. Nevertheless, their potential to significantly increase the safety of nuclear power plants still needs to be strengthened, in particular the ability of computer codes to determine their performance and reliability in industrial applications and support the safety demonstration. The PASTELS project (September 2020-February 2024), funded by the European Commission "Euratom H2020" programme, is devoted to the study of passive systems relying on natural circulation. The project focuses on two types, namely the SAfety COndenser (SACO) for the evacuation of the core residual power and the Containment Wall Condenser (CWC) for the reduction of heat and pressure in the containment vessel in case of accident. A specific design for each of these systems is being investigated in the project. Firstly, a straight vertical pool type of SACO has been implemented on the Framatome's PKL loop at Erlangen. It represents a tube bundle type heat exchanger that transfers heat from the secondary circuit to the water pool in which it is immersed by condensing the vapour generated in the steam generator. Secondly, the project relies on the CWC installed on the PASI test loop at LUT University in Finland. This facility reproduces the thermal-hydraulic behaviour of a Passive Containment Cooling System (PCCS) mainly composed of a CWC, a heat exchanger in the containment vessel connected to a water tank at atmospheric pressure outside the vessel which represents the ultimate heat sink. Several activities are carried out within the framework of the project. Different tests are conducted on these integral test facilities to produce new and relevant experimental data allowing to better characterize the physical behaviours and the performances of these systems for various thermo-hydraulic conditions. These test programmes are simulated by different codes acting at different scales, mainly system and CFD codes. New "system/CFD" coupling approaches are also considered to evaluate their potential to benefit both from the accuracy of CFD in regions where local 3D effects are dominant and system codes whose computational speed, robustness and general level of physical validation are particularly appreciated in industrial studies. In parallel, the project includes the study of single and two-phase natural circulation loops through a bibliographical study and the simulations of the PERSEO and HERO-2 experimental facilities. After a synthetic presentation of the project and its objectives, this article provides the reader with findings related to the physical analysis of the test results obtained on the PKL and PASI installations as well an overall evaluation of the capability of the different numerical tools to simulate passive systems.

• Information Systems Review
• /
• v.20 no.4
• /
• pp.79-96
• /
• 2018

This study started by focusing on the internalization of the technology appraisal model into the credit rating model to increase the discriminative power of the credit rating model not only for SMEs but also for all companies, reflecting the items related to the financial stability of the enterprises among the technology appraisal items. Therefore, it is aimed to verify whether the technology appraisal model can be applied to identify high-stability SMEs in advance. We classified companies into industries (manufacturing vs. non-manufacturing) and the age of company (initial vs. non-initial), and defined as a high-stability company that has achieved an average debt ratio less than 1/2 of the group for three years. The C5.0 was applied to verify the discriminant power of the model. As a result of the analysis, there is a difference in importance according to the type of industry and the age of company at the sub-item level, but in the mid-item level the R&D capability was a key variable for discriminating high-stability SMEs. In the early stage of establishment, the funding capacity (diversification of funding methods, capital structure and capital cost which taking into account profitability) is an important variable in financial stability. However, we concluded that technology development infrastructure, which enables continuous performance as the age of company increase, becomes an important variable affecting financial stability. The classification accuracy of the model according to the age of company and industry is 71~91%, and it is confirmed that it is possible to identify high-stability SMEs by using technology appraisal items.