• Investigative Magnetic Resonance Imaging
• /
• v.25 no.4
• /
• pp.266-280
• /
• 2021

Advances in radiomics and deep learning (DL) hold great potential to be at the forefront of precision medicine for the treatment of patients with brain metastases. Radiomics and DL can aid clinical decision-making by enabling accurate diagnosis, facilitating the identification of molecular markers, providing accurate prognoses, and monitoring treatment response. In this review, we summarize the clinical background, unmet needs, and current state of research of radiomics and DL for the treatment of brain metastases. The promises, pitfalls, and future roadmap of radiomics and DL in brain metastases are addressed as well.

• Annual Conference of KIPS
• /
• 2021.05a
• /
• pp.331-334
• /
• 2021

In the latest eras, there has been a significant amount of development and research is done on the usage of Deep Learning (DL) for speech emotion recognition (SER) based on Convolutional Neural Network (CNN). These techniques are usually focused on utilizing CNN for an application associated with emotion recognition. Moreover, numerous mechanisms are deliberated that is based on deep learning, meanwhile, it's important in the SER-based human-computer interaction (HCI) applications. Associating with other methods, the methods created by DL are presenting quite motivating results in many fields including automatic speech recognition. Hence, it appeals to a lot of studies and investigations. In this article, a review with evaluations is illustrated on the improvements that happened in the SER domain though likewise arguing the existing studies that are existence SER based on DL and CNN methods.

• Clinical Endoscopy
• /
• v.55 no.5
• /
• pp.594-604
• /
• 2022

Over the past decade, technological advances in deep learning have led to the introduction of artificial intelligence (AI) in medical imaging. The most commonly used structure in image recognition is the convolutional neural network, which mimics the action of the human visual cortex. The applications of AI in gastrointestinal endoscopy are diverse. Computer-aided diagnosis has achieved remarkable outcomes with recent improvements in machine-learning techniques and advances in computer performance. Despite some hurdles, the implementation of AI-assisted clinical practice is expected to aid endoscopists in real-time decision-making. In this summary, we reviewed state-of-the-art AI in the field of gastrointestinal endoscopy and offered a practical guide for building a learning image dataset for algorithm development.

• Current Optics and Photonics
• /
• v.8 no.5
• /
• pp.508-514
• /
• 2024

This paper presents a deep learning-based weight sensor, using optical speckle patterns of multimode fiber, designed for real-time intrusion detection. The weight sensor has been trained to identify 11 distinct speckle patterns, ranging in weight from 0.0 kg to 2.0 kg, with an interval of 200 g between each pattern. The estimation for untrained weights is based on the generalization capability of deep learning. This results in an average weight error of 243.8 g. Although this margin of error precludes accurate weight measurement, the system's ability to detect abrupt weight changes makes it a suitable choice for intrusion detection applications. The weight sensor is integrated with the Google Teachable Machine, and real-time intrusion notifications are facilitated by the ThingSpeak^TM cloud platform, an open-source Internet of Things (IoT) application developed by MathWorks.

• Journal of Internet Computing and Services
• /
• v.19 no.1
• /
• pp.27-35
• /
• 2018

This paper is a study to classify malicious applications in Android environment. And studying the threat and behavioral analysis of malicious Android applications. In addition, malicious apps classified by machine learning were performed as experiments. Android behavior analysis can use dynamic analysis tools. Through this tool, API Calls, Runtime Log, System Resource, and Network information for the application can be extracted. We redefined the properties extracted for machine learning and evaluated the results of machine learning classification by verifying between the overall features and the main features. The results show that key features have been improved by 1~4% over the full feature set. Especially, SVM classifier improved by 10%. From these results, we found that the application of the key features as a key feature was more effective in the performance of the classification algorithm than in the use of the overall features. It was also identified as important to select meaningful features from the data sets.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.8
• /
• pp.738-745
• /
• 2014

Recently, mobile users continuously increase, and mobile applications also increase As mobile applications increase, the mobile users used to store sensitive and private information such as Bank information, location information, ID, password on their mobile devices. Therefore, recent malicious application targeted to mobile device instead of PC environment is increasing. In particular, since the Android is an open platform and includes security vulnerabilities, attackers prefer this environment. This paper analyzes the performance of malware detection system applying linear SVM machine learning classifier to detect Android malware application. This paper also performs feature selection in order to improve detection performance.

• Smart Structures and Systems
• /
• v.24 no.5
• /
• pp.567-585
• /
• 2019

In the past two decades, structural health monitoring (SHM) systems have been widely installed on various civil infrastructures for the tracking of the state of their structural health and the detection of structural damage or abnormality, through long-term monitoring of environmental conditions as well as structural loadings and responses. In an SHM system, there are plenty of sensors to acquire a huge number of monitoring data, which can factually reflect the in-service condition of the target structure. In order to bridge the gap between SHM and structural maintenance and management (SMM), it is necessary to employ advanced data processing methods to convert the original multi-source heterogeneous field monitoring data into different types of specific physical indicators in order to make effective decisions regarding inspection, maintenance and management. Conventional approaches to data analysis are confronted with challenges from environmental noise, the volume of measurement data, the complexity of computation, etc., and they severely constrain the pervasive application of SHM technology. In recent years, with the rapid progress of computing hardware and image acquisition equipment, the deep learning-based data processing approach offers a new channel for excavating the massive data from an SHM system, towards autonomous, accurate and robust processing of the monitoring data. Many researchers from the SHM community have made efforts to explore the applications of deep learning-based approaches for structural damage detection and structural condition assessment. This paper gives a review on the deep learning-based SHM of civil infrastructures with the main content, including a brief summary of the history of the development of deep learning, the applications of deep learning-based data processing approaches in the SHM of many kinds of civil infrastructures, and the key challenges and future trends of the strategy of deep learning-based SHM.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.22 no.1
• /
• pp.49-59
• /
• 2019

Generative adversarial networks(GANs) have received great attention in the machine learning field for their capacity to model high-dimensional and complex data distribution implicitly and generate new data samples from the model distribution. This paper investigates the model training methodology, architecture, and various applications of generative adversarial networks. Experimental evaluation is also conducted for generating synthetic image dataset for defense using two types of GANs. The first one is for military image generation utilizing the deep convolutional generative adversarial networks(DCGAN). The other is for visible-to-infrared image translation utilizing the cycle-consistent generative adversarial networks(CycleGAN). Each model can yield a great diversity of high-fidelity synthetic images compared to training ones. This result opens up the possibility of using inexpensive synthetic images for training neural networks while avoiding the enormous expense of collecting large amounts of hand-annotated real dataset.

• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.479-486
• /
• 2024

This study reviews the recently conducted case studies to explore the innovative integration of Artificial Intelligence (AI) and Machine Learning (ML) in the domain of building facility management and predictive maintenance. It systematically examines recent developments and applications of advanced computational methods, emphasizing their role in enhancing asset management accuracy, energy efficiency, and occupant comfort. The study investigates the implementation of various AI and ML techniques, such as regression methods, Artificial Neural Networks (ANNs), and deep learning models, demonstrating their utility in asset management. It also discusses the synergistic use of ML with domain-specific technologies such as Geographic Building Information Modeling (BIM), Information Systems (GIS), and Digital Twin (DT) technologies. Through a critical analysis of current trends and methodologies, the paper highlights the importance of algorithm selection based on data attributes and operational challenges in deploying sophisticated AI models. The findings underscore the transformative potential of AI and ML in facility management, offering insights into future research directions and the development of more effective, data-driven management strategies.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.14 no.5
• /
• pp.277-286
• /
• 2019

In this paper, we propose an efficient hyperplane generation technique to classify human activity from combination of events and sequence information obtained from multiple-event sensors. By generating hyperplane efficiently, our machine learning algorithm classify with less memory and run time than the LSVM (Linear Support Vector Machine) for embedded system. Because the fact that light weight and high speed algorithm is one of the most critical issue in the IoT, the study can be applied to smart home to predict human activity and provide related services. Our approach is based on reducing numbers of hyperplanes and utilizing robust string comparing algorithm. The proposed method results in reduction of memory consumption compared to the conventional ML (Machine Learning) algorithms; 252 times to LSVM and 34,033 times to LSTM (Long Short-Term Memory), although accuracy is decreased slightly. Thus our method showed outstanding performance on accuracy per hyperplane; 240 times to LSVM and 30,520 times to LSTM. The binarized image is then divided into groups, where each groups are converted to binary number, in order to reduce the number of comparison done in runtime process. The binary numbers are then converted to string. The test data is evaluated by converting to string and measuring similarity between hyperplanes using Levenshtein algorithm, which is a robust dynamic string comparing algorithm. This technique reduces runtime and enables the proposed algorithm to become 27% faster than LSVM, and 90% faster than LSTM.