• Journal of Korea Multimedia Society
• /
• v.25 no.5
• /
• pp.661-669
• /
• 2022

The purpose of this study is to compare and analyze Deep Learning (DL) and Digital Image Processing (DIP) techniques using the results of the glottis segmentation of the two methods followed by the quantification of the asymmetric degree of the vocal cord mucosa. The data consists of 40 normal and abnormal images. The DL model is based on Deeplab V3 architecture, and the Canny edge detector algorithm and morphological operations are used for the DIP technique. According to the segmentation results, the average accuracy of the DL model and the DIP was 97.5% and 94.7% respectively. The quantification results showed high correlation coefficients for both the DL experiment (r=0.8512, p<0.0001) and the DIP experiment (r=0.7784, p<0.0001). In the conclusion, the DL model showed relatively higher segmentation accuracy than the DIP. In this paper, we propose the clinical applicability of this technique applying the segmentation and asymmetric quantification algorithm to the glottal area in the laryngoscopic images.

• Journal of Veterinary Clinics
• /
• v.40 no.4
• /
• pp.243-259
• /
• 2023

Machine learning and deep learning (ML/DL) algorithms have been successfully applied in medical practice. However, their application in veterinary medicine is relatively limited, possibly due to a lack in the quantity and quality of relevant research. Because the potential demands for ML/DL applications in veterinary clinics are significant, it is important to note the current gaps in the literature and explore the possible directions for advancement in this field. Thus, a scoping review was conducted as a situation analysis. We developed a search strategy following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. PubMed and Embase databases were used in the initial search. The identified items were screened based on predefined inclusion and exclusion criteria. Information regarding model development, quality of validation, and model performance was extracted from the included studies. The current review found 55 studies that passed the criteria. In terms of target animals, the number of studies on industrial animals was similar to that on companion animals. Quantitative scarcity of prediction studies (n = 11, including duplications) was revealed in both industrial and non-industrial animal studies compared to diagnostic studies (n = 45, including duplications). Qualitative limitations were also identified, especially regarding validation methodologies. Considering these gaps in the literature, future studies examining the prediction and validation processes, which employ a prospective and multi-center approach, are highly recommended. Veterinary practitioners should acknowledge the current limitations in this field and adopt a receptive and critical attitude towards these new technologies to avoid their abuse.

• ETRI Journal
• /
• v.43 no.3
• /
• pp.549-560
• /
• 2021

Cyberattacks are often difficult to identify with traditional signature-based detection, because attackers continually find ways to bypass the detection methods. Therefore, researchers have introduced artificial intelligence (AI) technology for cybersecurity analysis to detect malicious PowerShell scripts. In this paper, we propose a feature optimization technique for AI-based approaches to enhance the accuracy of malicious PowerShell script detection. We statically analyze the PowerShell script and preprocess it with a method based on the tokens and abstract syntax tree (AST) for feature selection. Here, tokens and AST represent the vocabulary and structure of the PowerShell script, respectively. Performance evaluations with optimized features yield detection rates of 98% in both machine learning (ML) and deep learning (DL) experiments. Among them, the ML model with the 3-gram of selected five tokens and the DL model with experiments based on the AST 3-gram deliver the best performance.

• Geomechanics and Engineering
• /
• v.37 no.6
• /
• pp.629-641
• /
• 2024

The presence of excavations or cavities beneath the foundations of a building can have a significant impact on their stability and cause extensive damage. Traditional methods for calculating the bearing capacity and subsidence of foundations over cavities can be complex and time-consuming, particularly when dealing with conditions that vary. In such situations, machine learning (ML) and deep learning (DL) techniques provide effective alternatives. This study concentrates on constructing a prediction model based on the performance of ML and DL algorithms that can be applied in real-world settings. The efficacy of eight algorithms, including Regression Analysis, k-Nearest Neighbor, Decision Tree, Random Forest, Multivariate Regression Spline, Artificial Neural Network, and Deep Neural Network, was evaluated. Using a Python-assisted automation technique integrated with the PLAXIS 2D platform, a dataset containing 272 cases with eight input parameters and one target variable was generated. In general, the DL model performed better than the ML models, and all models, except the regression models, attained outstanding results with an R² greater than 0.90. These models can also be used as surrogate models in reliability analysis to evaluate failure risks and probabilities.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.2
• /
• pp.91-108
• /
• 2022

In most cases, optical images have been used as training data of DL (Deep Learning) models for object detection, recognition, identification, classification, semantic segmentation, and instance segmentation. However, properties of 3D objects in the real-world could not be fully explored with 2D images. One of the major sources of the 3D geospatial information is DSM (Digital Surface Model). In this matter, characteristic information derived from DSM would be effective to analyze 3D terrain features. Especially, man-made objects such as buildings having geometrically unique shape could be described by geometric elements that are obtained from 3D geospatial data. The background and motivation of this paper were drawn from concept of the intrinsic image that is involved in high-level visual information processing. This paper aims to extract buildings after classifying terrain features by training DL model with DSM-derived information including slope, aspect, and SRI (Shaded Relief Image). The experiments were carried out using DSM and label dataset provided by ISPRS (International Society for Photogrammetry and Remote Sensing) for CNN-based SegNet model. In particular, experiments focus on combining multi-source information to improve training performance and synergistic effect of the DL model. The results demonstrate that buildings were effectively classified and extracted by the proposed approach.

• Journal of Internet Technology
• /
• v.22 no.4
• /
• pp.903-911
• /
• 2021

As cyber-attacks on Cyber-Physical System (CPS) become more diverse and sophisticated, it is important to quickly detect malicious behaviors occurring in CPS. Since CPS can collect sensor data in near real time throughout the process, there have been many attempts to detect anomaly behavior through normal behavior learning from the perspective of data-driven security. However, since the CPS datasets are big data and most of the data are normal data, it has always been a great challenge to analyze the data and implement the anomaly detection model. In this paper, we propose and evaluate the Clustered Deep One-Class Classification (CD-OCC) model that combines the clustering algorithm and deep learning (DL) model using only a normal dataset for anomaly detection. We use auto-encoder to reduce the dimensions of the dataset and the K-means clustering algorithm to classify the normal data into the optimal cluster size. The DL model trains to predict clusters of normal data, and we can obtain logit values as outputs. The derived logit values are datasets that can better represent normal data in terms of knowledge distillation and are used as inputs to the OCC model. As a result of the experiment, the F1 score of the proposed model shows 0.93 and 0.83 in the SWaT and HAI dataset, respectively, and shows a significant performance improvement over other recent detectors such as Com-AE and SVM-RBF.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.1100-1118
• /
• 2021

In a wide range of ML applications, the training data contains privacy-sensitive information that should be kept secure. Training the ML systems by privacy-sensitive data makes the ML model inherent to the data. As the structure of the model has been fine-tuned by training data, the model can be abused for accessing the data by the estimation in a reverse process called model inversion attack (MIA). Although, MIA has been applied to shallow neural network models of recognizers in literature and its threat in privacy violation has been approved, in the case of a deep learning (DL) model, its efficiency was under question. It was due to the complexity of a DL model structure, big number of DL model parameters, the huge size of training data, big number of registered users to a DL model and thereof big number of class labels. This research work first analyses the possibility of MIA on a deep learning model of a recognition system, namely a face recognizer. Second, despite the conventional MIA under the white box scenario of having partial access to the users' non-sensitive information in addition to the model structure, the MIA is implemented on a deep face recognition system by just having the model structure and parameters but not any user information. In this aspect, it is under a semi-white box scenario or in other words a gray-box scenario. The experimental results in targeting five registered users of a CNN-based face recognition system approve the possibility of regeneration of users' face images even for a deep model by MIA under a gray box scenario. Although, for some images the evaluation recognition score is low and the generated images are not easily recognizable, but for some other images the score is high and facial features of the targeted identities are observable. The objective and subjective evaluations demonstrate that privacy cyber-attack by MIA on a deep recognition system not only is feasible but also is a serious threat with increasing alert state in the future as there is considerable potential for integration more advanced ML techniques to MIA.

• Journal of Biomedical Engineering Research
• /
• v.44 no.4
• /
• pp.255-263
• /
• 2023

Clinical ultrasound (US) is a widely used imaging modality with various clinical applications. However, capturing a large field of view often requires specialized transducers which have limitations for specific clinical scenarios. Panoramic imaging offers an alternative approach by sequentially aligning image sections acquired from freehand sweeps using a standard transducer. To reconstruct a 3D volume from these 2D sections, an external device can be employed to track the transducer's motion accurately. However, the presence of optical or electrical interferences in a clinical setting often leads to incorrect measurements from such sensors. In this paper, we propose a deep learning (DL) framework that enables the prediction of scan trajectories using only US data, eliminating the need for an external tracking device. Our approach incorporates diverse data types, including correlation volume, optical flow, B-mode images, and rawer data (IQ data). We develop a DL network capable of effectively handling these data types and introduce an attention technique to emphasize crucial local areas for precise trajectory prediction. Through extensive experimentation, we demonstrate the superiority of our proposed method over other DL-based approaches in terms of long trajectory prediction performance. Our findings highlight the potential of employing DL techniques for trajectory estimation in clinical ultrasound, offering a promising alternative for panoramic imaging.

• Journal of Periodontal and Implant Science
• /
• v.54 no.5
• /
• pp.349-358
• /
• 2024

Purpose: Bone quality is one of the most important clinical factors for the primary stability and successful osseointegration of dental implants. This preliminary pilot study aimed to evaluate the clinical applicability of deep learning (DL) for assessing bone quality using panoramic (PA) radiographs compared with an implant surgeon's subjective tactile sense and cone-beam computed tomography (CBCT) values. Methods: In total, PA images of 2,270 edentulous sites for implant placement were selected, and the corresponding CBCT relative gray value measurements and bone quality classification were performed using 3-dimensional dental image analysis software. Based on the pre-trained and fine-tuned ResNet-50 architecture, the bone quality classification of PA images was classified into 4 levels, from D1 to D4, and Spearman correlation analyses were performed with the implant surgeon's tactile sense and CBCT values. Results: The classification accuracy of DL was evaluated using a test dataset comprising 454 cropped PA images, and it achieved an area under the receiving characteristic curve of 0.762 (95% confidence interval [CI], 0.714-0.810). Spearman correlation analysis of bone quality showed significant positive correlations with the CBCT classification (r=0.702; 95% CI, 0.651-0.747; P<0.001) and the surgeon's tactile sense (r=0.658; 95% CI, 0.600-0.708, P<0.001) versus the DL classification. Conclusions: DL classification using PA images showed a significant and consistent correlation with CBCT classification and the surgeon's tactile sense in classifying the bone quality at the implant placement site. Further research based on high-quality quantitative datasets is essential to increase the reliability and validity of this method for actual clinical applications.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.6
• /
• pp.469-481
• /
• 2018

DL (Deep Learning) is getting popular in various fields to implement artificial intelligence that resembles human learning and cognition. DL based on complicate structure of the ANN (Artificial Neural Network) requires computing power and computation cost. Variety of DL models with improved performance have been developed with powerful computer specification. The main purpose of this paper is to detect buildings from aerial images and evaluate performance of Mask R-CNN (Region-based Convolutional Neural Network) developed by FAIR (Facebook AI Research) team recently. Mask R-CNN is a R-CNN that is evaluated to be one of the best ANN models in terms of performance for semantic segmentation with pixel-level accuracy. The performance of the DL models is determined by training ability as well as architecture of the ANN. In this paper, we characteristics of the Mask R-CNN with various types of the images and evaluate possibility of the generalization which is the ultimate goal of the DL. As for future study, it is expected that reliability and generalization of DL will be improved by using a variety of spatial information data for training of the DL models.