检索结果-内蒙古大学图书馆

A Study on Optimization and Augmentation Techniques for Improving the Performance of CADx System for Cat Skin Disease

引用

Transactions of the Korean Institute of Electrical Engineers 2025年第1期74卷 135-141页

作者： Won, Hyeong-Sik Park, Jae-Beom Cho, Hyun-Chong Dept. of Electronics Engineering Interdisciplinary Graduate Program for BIT Medical Convergence Dept. of Data Science Kangwon National University Korea Republic of Dept. of Interdisciplinary Graduate Program for BIT Medical Convergence Kangwon National University Korea Republic of

In modern society, the number of households raising pets is increasing. As pet ownership increases, the cost of treating companion cats is also rising, with a significant portion of these costs going toward the treatment of skin diseases. Skin diseases are among the common ailments in pets, and swift action is required if they occur in cats. However, early lesions lack distinctive characteristics, making accurate diagnosis difficult. Therefore, this study proposes a CADx(Computer-Aided Diagnosis) system that classifies images of cat skin conditions into inflammatory lesions, non-inflammatory lesions, and normal images using a dataset of cat skin diseases. We selected the ConvNeXt model, based on CNN, which can learn regional information and features. To learn various patterns of skin lesions, we applied Mixup, an image augmentation technique. When Mixup was applied, the model accuracy was 0.8679, showing a 3.35% improvement compared to the original dataset. Additionally, Lookahead Optimizer was applied to ensure stable learning of Mixup. When both Mixup and Lookahead Optimizer were applied, the model accuracy was 0.8741, showing a high improvement of 3.97% compared to the original dataset. Therefore, the Lookahead Optimizer helps with the stable training of mixup, which can improve the performance of the model. Copyright © The Korean Institute of Electrical Engineers.

关键词： Dermatology

来源：评论

学校读者我要写书评

暂无评论

Development of a Deep Learning-Based System for Cat Skin Disease Classification and Grad-CAM Visualization

引用

Transactions of the Korean Institute of Electrical Engineers 2025年第2期74卷 339-344页

作者： Won, Hyeong-Sik Cho, Hyun-Chong Dept. of Interdisciplinary Graduate Program for BIT Medical Convergence Kangwon National University Korea Republic of Dept. of Electronics Engineering and Interdisciplinary Graduate Program for BIT Medical Convergence Dept. of Data Science Kangwon National University Korea Republic of

Skin diseases in companion cats can worsen if not treated promptly, and this can increase the financial burden on pet owners. To prevent this, early and accurate diagnosis is essential. This study introduces a deep learning-based Computer-Aided Diagnosis (CADx) system designed to classify cat skin diseases into non-inflammatory and inflammatory lesions by comparing them with normal images. The system employs the EfficientNetV2 model and incorporates image augmentation techniques like AutoAugment and AugMix to enhance classification performance. The study results indicate that the developed model achieved an 84.68% accuracy for non-inflammatory lesions, reflecting a 10.06% improvement, and a 97.19% accuracy for inflammatory lesions, reflecting a 2.45% improvement. Furthermore, we applied Grad-CAM to visualize the regions of interest in the images, offering veterinarians critical insights into the location and characteristics of the lesions. This system has the potential to significantly improve the precision of diagnosing skin diseases in companion cats, thereby supporting better veterinary care. Copyright © The Korean Institute of Electrical Engineers.

关键词： Veterinary medicine

来源：评论

学校读者我要写书评

暂无评论

Development of a Deep Learning-Based Model for Canine Skin Disease Diagnosis Using RandAugment and Patterned-GridMask

引用

Transactions of the Korean Institute of Electrical Engineers 2025年第1期74卷 142-148页

作者： Kim, Min-Jun Park, Jae-Beom Cho, Hyun-Chong Dept. of Electronics Engineering Interdisciplinary Graduate Program for BIT Medical Convergence Dept. of Data Science Kangwon National University Korea Republic of Dept. of Interdisciplinary Graduate Program for BIT Medical Convergence Kangwon National University Korea Republic of

In contemporary society, pets are increasingly regarded as integral family members, contributing significantly to human quality of life. The growing prevalence of dog ownership has concurrently escalated the economic burden associated with veterinary care, particularly in managing common conditions like skin diseases. This study introduces an advanced deep learning-based diagnostic system for canine skin diseases, designed for practical application in home environments. We employed RandAugment to enhance data augmentation, thereby increasing the diversity of the training dataset. Furthermore, the implementation of Patterned-GridMask significantly improved the model's generalization capabilities. The use of the AdamW optimization algorithm was instrumental in mitigating overfitting, thus enhancing the model's overall learning efficiency. The proposed Transformer-based model, ViT/B-16, achieved an accuracy of 78.65% with the original dataset. With the integration of RandAugment and Patterned-GridMask techniques, the model's accuracy improved to 84.33%, underscoring its potential effectiveness for practical veterinary applications. Copyright © The Korean Institute of Electrical Engineers.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

VR-Based Implicit Authentication Using Electromyogram and Inertial Measurement Unit 23

VR-Based Implicit Authentication Using Electromyogram and In...

引用

23rd International Conference on Cyberworlds, CW 2024

作者： Yu, Da-Young Lee, Hyung-Tak Kim, Youngsam Roh, Jong-Hyuk Kim, Soohyung Ku, Jeonghun Hwang, Han-Jeong Korea University Interdisciplinary Graduate Program for Artificial Intelligence Smart Convergence Technology Dept. Electronics and Information Engineering Sejong Korea Republic of Korea University Dept. Electronics and Information Engineering Sejong Korea Republic of Daejeon Korea Republic of Keimyung University College of Medicine Department of Biomedical Engineering Daegu Korea Republic of

ISBN: (纸本)9798331527174

Biometrics based on electromyogram (EMG) and inertial measurement unit (IMU) signals has received significant attention due to its resistance to forgery and tampering, as well as its relatively higher convenience and security compared to other biometric modalities. However, most EMG and IMU signal-based authentication methods require explicit authentication using unnatural gestures, which is not userfriendly. To address this, we collected EMG and IMU data during virtual reality (VR) rhythm game play, which does not require explicit gestures, and performed implicit authentication. Using a linear discriminant analysis (LDA) classifier, we achieved classification accuracies of 91.71% for human identification and 87.91% for human authentication. The result demonstrated the feasibility of a seamless and implicit authentication in VR environments. © 2024 IEEE.

关键词： Electromyography

来源：评论

学校读者我要写书评

暂无评论

Regionalization for Flood Frequency Analysis: Sensitivity to Choice of Clustering Algorithm and Distance Metric

Regionalization for Flood Frequency Analysis: Sensitivity to...

引用

World Environmental and Water Resources Congress 2023: Adaptive Planning and Design in an Age of Risk and Uncertainty

作者： Mondal, Arpita Roy, Reema Kalai, Chingka Dept. of Civil Engineering Interdisciplinary Program in Climate Studies Indian Institute of Technology Bombay Mumbai Powai India Dept. of Civil Engineering Indian Institute of Technology Bombay Mumbai Powai India

ISBN: (数字)9780784484852

ISBN: (纸本)9780784484852

Regional flood frequency analysis (RFFA) is a widely recognized technique for estimating reliable flood quantiles at ungauged sites or to augment the data available at gauged sites. The primary step in RFFA is regionalization, that is, the process of identifying hydrologically similar regions for spatial pooling. Artificial intelligence techniques such as distance-based clustering algorithms are most commonly adopted for the delineation of homogeneous regions, using the Euclidean distance metric by default. This study aims to understand the performance of regionalization techniques under different choices of distance measures, and how such sensitivity varies across clustering techniques used. Leave-one-out cross-validation (LOOCV) analysis is performed to compare estimated flood quantiles with at-site values. The distance measures considered are squared Euclidean, cityblock, cosine, correlation, Minkowski, and Mahalanobis distances, while two clustering algorithms are chosen - the conventional fuzzy C-means (FCM) approach and an advanced machine learning algorithm called the random forest classifier (RFC). The Nash-Sutcliffe efficiency (NSE) is used as a performance evaluation statistic to evaluate the accuracy of the estimated flood quantiles at ungauged sites. For illustration of regionalization and prediction in ungauged basins, naturalized flows are considered at 243 catchments in the State of Indiana, United States. It is found that in general the algorithms are sensitive to the choice of distance measure, with the Mahalanobis and cosine distances performing poorly both for region formation and subsequently for flood quantile estimation at ungauged sites. However, choice of a more suitable clustering algorithm is found to overcome such limitations with the RFC showing improved performance (NSE > 0.73 across all distance metrics), even with the Mahalanobis distance. Therefore, this study concludes that the choice of clustering algorithm is the more important par

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Dental Imaging Using a Higher Signal-to-Noise Ratio Optical Coherence Tomography System

Dental Imaging Using a Higher Signal-to-Noise Ratio Optical ...

引用

2023 IEEE International Conference on Electrical, Computer and Energy Technologies, ICECET 2023

作者： Azeez, Akeem. A. Mezgebo, Biniyam Franca, Rodrigo Sherif, Sherif S. University of Manitoba Biomedical Engineering Graduate Program Winnipeg Canada University of Manitoba Dept. of Electrical and Computer Engineering Winnipeg Canada University of Manitoba Dept. of Restorative Dentistry Winnipeg Canada

ISBN: (纸本)9798350327816

This paper presents an optical coherence tomography (OCT) system in conjunction with a novel image reconstruction technique employed for in vitro imaging of human teeth. The primary goal is to enhance the signal-to-noise ratio (SNR) in the obtained images. The study entails a comparative analysis between the conventional Fast Fourier Transform (FFT) OCT image reconstruction method and a newly introduced scaled nonuniform discrete Fourier transform (NDFT) approach. The findings reveal that the NDFT method consistently delivers superior results in terms of peak signal-to-noise ratio (PSNR) and overall image quality, even when dealing with redundant and nonuniform frequency domain samples. In light of these results, this paper concludes that integrating NDFT into OCT procedures has the potential to significantly enhance the quality of image reconstructions, thereby fostering its broader application in the field of dental imaging. © 2023 IEEE.

关键词： dental imaging frame theory image reconstruction Optical Coherence Tomography (OCT) signal processing

来源：评论

学校读者我要写书评

暂无评论

ReMP: Reusable Motion Prior for Multi-domain 3D Human Pose Estimation and Motion Inbetweening

ReMP: Reusable Motion Prior for Multi-domain 3D Human Pose E...

引用

2025 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2025

作者： Jang, Hojun Kim, Young Min Seoul National University Dept. of Electrical and Computer Engineering Korea Republic of Interdisciplinary Program in Artificial Intelligence and Inmc Seoul National University Korea Republic of

ISBN: (纸本)9798331510831

We present Reusable Motion prior (ReMP), an effective motion prior that can accurately track the temporal evolution of motion in various downstream tasks. Inspired by the success of foundation models, we argue that a robust spatio-temporal motion prior can encapsulate underlying 3D dynamics applicable to various sensor modalities. We learn the rich motion prior from a sequence of complete parametric models of posed human body shape. Our prior can easily estimate poses in missing frames or noisy measurements despite significant occlusion by employing a temporal attention mechanism. More interestingly, our prior can guide the system with incomplete and challenging input measurements to quickly extract critical information to estimate the sequence of poses, significantly improving the training efficiency for mesh sequence recovery. ReMP consistently outperforms the baseline method on diverse and practical 3D motion data, including dept. point clouds, LiDAR scans, and IMU sensor data. Project page is available in https://***/ReMP. © 2025 IEEE.

关键词： Reusability

来源：评论

学校读者我要写书评

暂无评论

Classification of Human Tactile Perception Using Vibration Data 23

Classification of Human Tactile Perception Using Vibration D...

引用

23rd International Conference on Cyberworlds, CW 2024

作者： Lee, Hyung-Tak Bak, Keungyonh Chun, Sungwoo Hwang, Han-Jeong Korea University Dept. Electronics and Information Engineering Sejong Korea Republic of Korea University Interdisciplinary Graduate Program for Artificial Intelligence Smart Convergence Technology Sejong Korea Republic of

ISBN: (纸本)9798331527174

To enhance the sense of reality in the VR environment, it is essential to incorporate not only visual and hearing but also tactile feedback. Previous studies related to tactile sensation have only focused on object classification to identify the object in contact with the tactile sensor. When people encounter a new object, they can use tactile perception information, such as roughness, to understand the object's attributes and functions. Therefore, for tactile research to be applicable in real-life scenarios, tactile perception classification is essential not only for classifying an object but also for recognizing its attributes and functions. This study aims to investigate the possibility of predicting roughness, one of the important human tactile perceptions, using deep learning models. We created 63 tactile samples, measured the vibration data of each sample using an acceleration sensor, and investigated how humans perceive the roughness of the tactile samples with five different levels. Object classification was performed using five deep learning models, among which the Shallow ConvNet achieved the highest performance, confirming its suitability as the most appropriate model for analyzing vibration data. Using the Shallow ConvNet, we attained a high accuracy of 98.6 ±5.6% for the identification of five levels of roughness, demonstrating that different levels of tactile perception information can be discriminated with high performance. © 2024 IEEE.

关键词： Deep learning Human tactile perception Roughness Vibration data Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Influence of RSSI Due To Cavitation-Induced Scattering In The Ballistic ESWT Device 24

Influence of RSSI Due To Cavitation-Induced Scattering In Th...

引用

24th International Congress on Acoustics, ICA 2022

作者： Kwon, Ohbin Choi, Min Joo Interdisciplinary Postgraduate Program in Biomedical Eng Korea Republic of Dept. of Medicine Jeju National University Jeju Korea Republic of

A laser Doppler method(LDM) is suffering from severe noises involved with cavitation in the ballistic extracorporeal shock wave therapy(bESWT) device. It is difficult to distinguish between a shielded signal due to cavitation or noise due to cavitation collapse. Although not accurate, you can determine the reliability of a signal from the signal level of the Receive Signal Strength Indicator (RSSI). The scattering due to cavitation of the shock wave transmitter surface of the bESWT, the RSSI signal, and the impulse oscillation relationship measured by LDM was observed. it uses a synchronized high-speed camera. The bubbles grew with time, and some of them were fused into larger bubbles and formed gradually an air layer on the surface of the shock wave transmitter. The interrogating laser toward the center of the vibrating circular surface was likely scattered and the laser returning to the sensor is largely reduced in power(i.e. RSSI). The red light captured by a high-speed camera exhibits strong scattering, consistent with the strong, dense noise of the impulse pulses. It was confirmed through RSSI that cavitation lowered the light intensity. The Red light is faster than RSSI for reaction rate. It has the potential to be proposed as a new evaluation method. © ICA *** rights reserved

关键词： cavitation extracorporeal shock wave therapy scattering light beam

来源：评论

学校读者我要写书评

暂无评论

Development of EEG-Based Optimal Biomarkers for Decoding Fatigue during Motor Imagery Tasks 13

Development of EEG-Based Optimal Biomarkers for Decoding Fat...

引用

13th International Winter Conference on Brain-Computer Interface, BCI 2025

作者： Lee, Jun-Seok Lee, Seung-Won Kim, Sang-Rok Hwang, Han-Jeong Dept. Electronics and Information Engineering Korea University Sejong30019 Korea Republic of Interdisciplinary Graduate Program for Artificial Intelligence Smart Convergence Technology Korea University Sejong30019 Korea Republic of

ISBN: (纸本)9798331521929

This study aimed to identify electroencephalography (EEG)-based neurophysiological biomarkers for monitoring mental fatigue during motor imagery (MI) tasks. We used EEG data recorded from 29 healthy adults during MI tasks. For the preprocessing of the EEG data, we sequentially performed down-sampling into 200 Hz, independent component analysis (ICA), band-pass filtering between 0.5 and 50 Hz, and common average reference (CAR). The preprocessed EEG data were analyzed using a fast Fourier transform (FFT) to calculate spectral power in the δ (0.5–3 Hz), θ (4–7 Hz), α (8–13 Hz), and β (13–30 Hz) frequency bands. Subsequently, seventeen fatigue-related metrics were calculated using individual spectral powers and their combinations. These fatigue-related metrics were averaged for the frontal lobe (F3 and F4), central lobe (FCC3h and FCC4h), parietal lobe (P3 and P4), and occipital lobe (POO1 and POO2). The statistical significance of the average fatigue metrics was evaluated through Pearson correlation analysis with the sequence of trials. As a result, we observed strong positive correlations (r > 0.9, p © 2025 IEEE.

关键词： Fast Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：