• Journal of Biomedical Engineering Research
• /
• v.42 no.4
• /
• pp.150-158
• /
• 2021

Intracranial hemorrhage (ICH) refers to acute bleeding inside the intracranial vault. Not only does this devastating disease record a very high mortality rate, but it can also cause serious chronic impairment of sensory, motor, and cognitive functions. Therefore, a prompt and professional diagnosis of the disease is highly critical. Noninvasive brain imaging data are essential for clinicians to efficiently diagnose the locus of brain lesion, volume of bleeding, and subsequent cortical damage, and to take clinical interventions. In particular, computed tomography (CT) images are used most often for the diagnosis of ICH. In order to diagnose ICH through CT images, not only medical specialists with a sufficient number of diagnosis experiences are required, but even when this condition is met, there are many cases where bleeding cannot be successfully detected due to factors such as low signal ratio and artifacts of the image itself. In addition, discrepancies between interpretations or even misinterpretations might exist causing critical clinical consequences. To resolve these clinical problems, we developed a diagnostic model predicting intracranial bleeding and its subtypes (intraparenchymal, intraventricular, subarachnoid, subdural, and epidural) by applying deep learning algorithms to CT images. We also constructed a visualization tool highlighting important regions in a CT image for predicting ICH. Specifically, 1) 27,758 CT brain images from RSNA were pre-processed to minimize the computational load. 2) Three different CNN-based models (ResNet, EfficientNet-B2, and EfficientNet-B7) were trained based on a training image data set. 3) Diagnosis performance of each of the three models was evaluated based on an independent test image data set: As a result of the model comparison, EfficientNet-B7's performance (classification accuracy = 91%) was a way greater than the other models. 4) Finally, based on the result of EfficientNet-B7, we visualized the lesions of internal bleeding using the Grad-CAM. Our research suggests that artificial intelligence-based diagnostic systems can help diagnose and treat brain diseases resolving various problems in clinical situations.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.115-122
• /
• 2016

This paper proposes a system to detect driver's cognitive state by internal and external information of vehicle. The proposed system can measure driver's eye gaze. This is done by concept of information delivery and mutual information measure. For this study, we set up two web-cameras at vehicles to obtain visual information of the driver and front of the vehicle. We propose Gestalt principle based selective attention model to define information quantity of road scene. The saliency map based on gestalt principle is prominently represented by stimulus such as traffic signals. The proposed system assumes driver's cognitive resource allocation on the front scene by gaze analysis and head pose direction information. Then we use several feature algorithms for detecting driver's characteristics in real time. Modified census transform (MCT) based Adaboost is used to detect driver's face and its component whereas POSIT algorithms are used for eye detection and 3D head pose estimation. Experimental results show that the proposed system works well in real environment and confirm its usability.

• Imaging Science in Dentistry
• /
• v.50 no.2
• /
• pp.81-92
• /
• 2020

Intelligent systems(i.e., artificial intelligence), particularly deep learning, are machines able to mimic the cognitive functions of humans to perform tasks of problem-solving and learning. This field deals with computational models that can think and act intelligently, like the human brain, and construct algorithms that can learn from data to make predictions. Artificial intelligence is becoming important in radiology due to its ability to detect abnormalities in radiographic images that are unnoticed by the naked human eye. These systems have reduced radiologists' workload by rapidly recording and presenting data, and thereby monitoring the treatment response with a reduced risk of cognitive bias. Intelligent systems have an important role to play and could be used by dentists as an adjunct to other imaging modalities in making appropriate diagnoses and treatment plans. In the field of maxillofacial radiology, these systems have shown promise for the interpretation of complex images, accurate localization of landmarks, characterization of bone architecture, estimation of oral cancer risk, and the assessment of metastatic lymph nodes, periapical pathologies, and maxillary sinus pathologies. This review discusses the clinical applications and scope of intelligent systems such as machine learning, artificial intelligence, and deep learning programs in maxillofacial imaging.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.6 s.38
• /
• pp.9-16
• /
• 2005

Recently, since the number of internet users is increasing rapidly and, by using the Public hacking tools, general network users can intrude computer systems easily, the hacking problem is setting more serious. In order to prevent the intrusion. it is needed to detect the sign in advance of intrusion in a Positive Prevention by detecting the various forms of hackers intrusion trials to know the vulnerability of systems. The existing network-based anomaly detection algorithms that cope with port-scanning and the network vulnerability scans have some weakness in intrusion detection. they can not detect slow scans and coordinated scans. therefore, the new concept of algorithm is needed to detect effectively the various. In this Paper, we propose a detection algorithm for session patterns and FCM.

• Journal of the Korean Chemical Society
• /
• v.54 no.3
• /
• pp.329-337
• /
• 2010

In this study, we analyzed the end-of-chapter questions in 8 types of chemistry II textbooks for science teachers according to revised Bloom's taxonomy of educational objectives not only to raise interests of questions in textbooks but also acquire a basic material for using questions in textbooks effectively. The results of classification following Bloom's cognitive category showed that 'Understanding(44.7%)' level was the most, then 'Application(29.9%)', Knowledge(15.6%) and 'Analysis (9.5%)' in order, which is distinct difference from the result of classification of the end-of-chapter questions in college general chemistry books which was 'Application', 'Analysis' and 'Understanding' in order. Especially, questions of 'Evaluation' level were not found at all in any textbook investigated and 'Synthesis(0.3%)' level was very few. On the other hand, the percentage of questions in 'Understanding' and 'Executing Quantitative' which required specific algorithms was 70% of total with most of the questions in 'Application' were 'Executing Quantitative'.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.3A
• /
• pp.291-303
• /
• 2006

In this paper, we analyze functional requirements of the IEEE 802.22 WRAN, and propose a downlink 프레임 structure satisfying the requirements. The proposed downlink 프레임 structure maximizes e transmission efficiency by adopting the cognative radio to assign the sub-channel by reflecting the channel environment of WRAN. We also calculate the signalling overhead for both downlink and uplink, and analyze the performances of time synchronization, frequency synchronization and cell identification based on the 프리앰블 in downlink and suggest the channel estimation method tough 프리앰블 or pilot. As a final result, e stationary beamforming (SBF) algorithm with dynamic channel allocation(DCA) is proposed. The proposed OFDMA downlink 프레임 structure with channel adaptive sub-channel allocation for cognitive radio applications is verified to meet the requirements of IEEE 802.22 WRAN, by computer simulations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.2
• /
• pp.520-541
• /
• 2023

In CRNs, SS is of utmost significance. Every CR user generates a sensing report during the training phase beneath various circumstances, and depending on a collective process, either communicates or remains silent. In the training stage, the fusion centre combines the local judgments made by CR users by a majority vote, and then returns a final conclusion to every CR user. Enough data regarding the environment, including the activity of PU and every CR's response to that activity, is acquired and sensing classes are created during the training stage. Every CR user compares their most recent sensing report to the previous sensing classes during the classification stage, and distance vectors are generated. The posterior probability of every sensing class is derived on the basis of quantitative data, and the sensing report is then classified as either signifying the presence or absence of PU. The ISVM technique is utilized to compute the quantitative variables necessary to compute the posterior probability. Here, the iterations of SVM are tuned by novel GO-PSA by combining GOA and PSO. Novel GO-PSA is developed since it overcomes the problem of computational complexity, returns minimum error, and also saves time when compared with various state-of-the-art algorithms. The dependability of every CR user is taken into consideration as these local choices are then integrated at the fusion centre utilizing an innovative decision combination technique. Depending on the collective choice, the CR users will then communicate or remain silent.

• Journal of Information Science Theory and Practice
• /
• v.7 no.2
• /
• pp.40-53
• /
• 2019

Although fake news has been present in human history at any time, nowadays, with social media, deceptive information has a stronger effect on society than before. This article answers two research questions, namely (1) Is the dissemination of fake news supported by machines through the automatic construction of filter bubbles, and (2) Are echo chambers of fake news manmade, and if yes, what are the information behavior patterns of those individuals reacting to fake news? We discuss the role of filter bubbles by analyzing social media's ranking and results' presentation algorithms. To understand the roles of individuals in the process of making and cultivating echo chambers, we empirically study the effects of fake news on the information behavior of the audience, while working with a case study, applying quantitative and qualitative content analysis of online comments and replies (on a blog and on Reddit). Indeed, we found hints on filter bubbles; however, they are fed by the users' information behavior and only amplify users' behavioral patterns. Reading fake news and eventually drafting a comment or a reply may be the result of users' selective exposure to information leading to a confirmation bias; i.e. users prefer news (including fake news) fitting their pre-existing opinions. However, it is not possible to explain all information behavior patterns following fake news with the theory of selective exposure, but with a variety of further individual cognitive structures, such as non-argumentative or off-topic behavior, denial, moral outrage, meta-comments, insults, satire, and creation of a new rumor.

• Smart Structures and Systems
• /
• v.20 no.1
• /
• pp.99-114
• /
• 2017

In cognitive science, it is illustrated how the collective opinions of a group of individuals answers to questions involving quantity estimation. One example of this approach is introduced in this article as Star Graph (SG) algorithm. This graph describes the details of communication among individuals to share their information and make a new decision. A new labyrinthine network of neighbors is defined in the decision-making process of the algorithm. In order to prevent getting trapped in local optima, the neighboring networks are regenerated in each iteration of the algorithm. In this algorithm, the normal distribution is utilized for a group of agents with the best results (guidance group) to replace the existing infeasible solutions. Here, some new functions are introduced to provide a high convergence for the method. These functions not only increase the local and global search capabilities but also require less computational effort. Various benchmark functions and engineering problems are examined and the results are compared with those of some other algorithms to show the capability and performance of the presented method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.3
• /
• pp.921-940
• /
• 2015

In this paper, we investigate the multi-resource allocation problem, a unique feature of which is that the multiple resources can compensate each other while achieving the desired system performance. In particular, power and time allocations are jointly optimized with the target of energy efficiency under the resource-limited constraints. Different from previous studies on the power-time tradeoff, we consider a multi-server case where the concurrent serving users are quantitatively restricted. Therefore user selection is investigated accompanying the resource allocation, making the power-time tradeoff occur not only between the users in the same server but also in different servers. The complex multivariate optimization problem can be modeled as a variant of 2-Dimension Bin Packing Problem (V2D-BPP), which is a joint non-linear and integer programming problem. Though we use state decomposition model to transform it into a convex optimization problem, the variables are still coupled. Therefore, we propose an Iterative Dual Optimization (IDO) algorithm to obtain its optimal solution. Simulations show that the joint multi-resource allocation algorithm outperforms two existing non-joint algorithms from the perspective of energy efficiency.