检索结果-内蒙古大学图书馆

Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing

作者： Donggeun Ko Dongjun Lee Namjun Park Wonkyeong Shim Jaekwang Kim AiM Future Seoul Republic of Korea Maum AI Gyeonggi-do Republic of Korea Department of Computer Science and Engineering/Convergence Program for Social Innovation Sungkyunkwan University Suwon Republic of Korea Department of Computer Science and Engineering Sungkyunkwan University Suwon Republic of Korea School of Convergence / Convergence Program for Social Innovation Sungkyunkwan University Suwon Republic of Korea

ISBN: (纸本)9798400706295

Neural networks struggle with image classification when biases are learned and misleads correlations, affecting their generalization and performance. Previous methods require attribute labels (e.g. background, color) or utilizes Generative Adversarial Networks (GANs) to mitigate biases. We introduce DiffuBias, a novel pipeline for text-to-image generation that generates bias-conflict samples, without any training. By utilizing pretrained diffusion and image captioning models, DiffuBias generates, bias-conflict samples using the top-K losses from a biased classifier (fB) to debias the classifier. This method not only debiases effectively but also boosts classifier generalization capabilities. Our comprehensive experimental evaluations demonstrate that DiffuBias achieves state-of-the-art performance on benchmark datasets.

关键词： debiasing

来源：评论

学校读者我要写书评

暂无评论

PacECG-Net: A Multi-modal Approach Integrating LLMs and ECG for LVSD Classification in Pacemaker Patients 25

PacECG-Net: A Multi-modal Approach Integrating LLMs and ECG ...

引用

Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing

作者： Wonkyeong Shim Namjun Park Donggeun Ko San Kim Hye Bin Gwag Young Jun Park Seung-Jung Park Jaekwang Kim Department of Computer Science and Engineering Sungkyunkwan University Suwon Republic of Korea Aim Future AI Seoul Republic of Korea Sungkyunkwan University Suwon Republic of Korea Samsung Changwon Hospital Changwon Republic of Korea Wonju Severance Christian Hospital Wonju Republic of Korea Samsung Medical Center Seoul Republic of Korea School of Convergence / Convergence Program for Social Innovation Sungkyunkwan University Seoul Republic of Korea

ISBN: (纸本)9798400706295

This study presents an AI-based model using ECG signals to predict left ventricular systolic dysfunction (LVSD) in pacemaker patients. A 1D convolutional neural network (CNN) combined with large language models processed both sequential ECG data and nonsequential clinical metadata. The model achieved an AUROC of 0.97 on both general and pacemaker-specific datasets, demonstrating its high accuracy. This approach offers a fast, cost-effective alternative to traditional echocardiography, improving LVSD detection in patients with pacemakers.

关键词： pacemaker

来源：评论

学校读者我要写书评

暂无评论

AVTENet: A Human-Cognition-Inspired Audio-Visual Transformer-Based Ensemble Network for Video Deepfake Detection

引用

IEEE Transactions on Cognitive and Developmental Systems 2025年

作者： Hashmi, Ammarah Shahzad, Sahibzada Adil Lin, Chia-Wen Tsao, Yu Wang, Hsin-Min Taiwan International Graduate Program Academia Sinica Social Networks and Human-Centered Computing Program Taipei11529 Taiwan National Tsing Hua University Institute of Information Systems and Applications Hsinchu30013 Taiwan National Chengchi University Department of Computer Science Taipei11605 Taiwan National Tsing Hua University Department of Electrical Engineering Institute of Communications Engineering Hsinchu300044 Taiwan Academia Sinica Research Center for Information Technology Innovation Taipei11529 Taiwan

The recent proliferation of hyper-realistic deepfake videos has drawn attention to the threat of audio and visual forgeries. Most previous studies on detecting artificial intelligence-generated fake videos only utilize visual modality or audio modality. While some methods exploit audio and visual modalities to detect forged videos, they have not been comprehensively evaluated on multimodal datasets of deepfake videos involving acoustic and visual manipulations, and are mostly based on convolutional neural networks with low detection accuracy. Considering that human cognition instinctively integrates multisensory information including audio and visual cues to perceive and interpret content and the success of transformer in various fields, this study introduces the audio-visual transformer-based ensemble network (AVTENet). This innovative framework tackles the complexities of deepfake technology by integrating both acoustic and visual manipulations to enhance the accuracy of video forgery detection. Specifically, the proposed model integrates several purely transformer-based variants that capture video, audio, and audio-visual salient cues to reach a consensus in prediction. For evaluation, we use the recently released benchmark multimodal audio-video FakeAVCeleb dataset. For a detailed analysis, we evaluate AVTENet, its variants, and several existing methods on multiple test sets of the FakeAVCeleb dataset. Experimental results show that the proposed model outperforms all existing methods and achieves state-of-the-art performance on Testset-I and Testset-II of the FakeAVCeleb dataset. We also compare AVTENet against humans in detecting video forgery. The results show that AVTENet significantly outperforms humans. © 2016 IEEE.

关键词： audio-visual AVTENet Deepfake detection

来源：评论

学校读者我要写书评

暂无评论

CAWA: Correlation-Based Adaptive Weight Adjustment via Effective Fields for Debiasing 13

CAWA: Correlation-Based Adaptive Weight Adjustment via Effec...

引用

Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024

作者： Noh, Kyoungrae Kim, San Kim, Jaekwang Sungkyunkwan University Computer Science and Engineering Suwon Korea Republic of Sungkyunkwan University Convergence Program for Social Innovation Suwon Korea Republic of

ISBN: (纸本)9798350373332

In real-world image classification tasks, neural networks often encounter significant challenges due to unexpected or deceptive correlations and biases inherent in the dataset. These biases can emerge from disproportionate data distributions, causing models to generalize poorly to new, unseen data. Such data distribution issues are particularly problematic compared to more balanced datasets because they lead the model to rely on bias attributes rather than intrinsic attributes. Ideally, the model should classify based on intrinsic attributes, but due to the influence of bias attributes, frequent misclassification occurs. Such biases compromise the fairness and accuracy of the model, especially in critical scenarios such as medical diagnosis, autonomous driving, or criminal justice, where misclassification can have significant consequences. To address these challenges, we propose an innovative two-stage approach aimed at mitigating bias more effectively and efficiently. © 2024 IEEE.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

A Novel Human Interaction Framework Using Quadratic Discriminant Analysis with HMM

引用

computers, Materials & Continua 2023年第11期77卷 1557-1573页

作者： Tanvir Fatima Naik Bukht Naif Al Mudawi Saud S.Alotaibi Abdulwahab Alazeb Mohammed Alonazi Aisha Ahmed AlArfaj Ahmad Jalal Jaekwang Kim Department of Computer Science Air UniversityIslamabad44000Pakistan Department of Computer Science College of Computer Science and Information SystemNajran UniversityNajran55461Saudi Arabia Information Systems Department Umm Al-Qura UniversityMakkahSaudi Arabia Department of Information Systems College of Computer Engineering and SciencesPrince Sattam bin Abdulaziz UniversityAl-Kharj16273Saudi Arabia Department of Information Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Convergence Program for Social Innovation Sungkyunkwan UniversitySuwon03063Korea

Human-human interaction recognition is crucial in computer vision fields like surveillance,human-computer interaction,and social *** enhances systems’ability to interpret and respond to human behavior *** research focuses on recognizing human interaction behaviors using a static image,which is challenging due to the complexity of diverse *** overall purpose of this study is to develop a robust and accurate system for human interaction *** research presents a novel image-based human interaction recognition method using a Hidden Markov Model(HMM).The technique employs hue,saturation,and intensity(HSI)color transformation to enhance colors in video frames,making them more vibrant and visually appealing,especially in low-contrast or washed-out *** filters reduce noise and smooth imperfections followed by silhouette extraction using a statistical *** extraction uses the features from Accelerated Segment Test(FAST),Oriented FAST,and Rotated BRIEF(ORB)*** application of Quadratic Discriminant Analysis(QDA)for feature fusion and discrimination enables high-dimensional data to be effectively analyzed,thus further enhancing the classification *** ensures that the final features loaded into the HMM classifier accurately represent the relevant human *** impressive accuracy rates of 93%and 94.6%achieved in the BIT-Interaction and UT-Interaction datasets respectively,highlight the success and reliability of the proposed *** proposed approach addresses challenges in various domains by focusing on frame improvement,silhouette and feature extraction,feature fusion,and HMM *** enhances data quality,accuracy,adaptability,reliability,and reduction of errors.

关键词： Human interaction recognition HMM classification quadratic discriminant analysis dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

CAWA: Correlation-Based Adaptive Weight Adjustment via Effective Fields for Debiasing

CAWA: Correlation-Based Adaptive Weight Adjustment via Effec...

引用

International Symposium on Soft Computing and Intelligent Systems (SCIS)

作者： Kyoungrae Noh San Kim Jaekwang Kim Computer Science and Engineering Sungkyunkwan University Suwon South Korea Convergence Program for Social Innovation Sungkyunkwan University Suwon South Korea

ISBN: (数字)9798350373332

ISBN: (纸本)9798350373349

关键词： Adaptation models Correlation Accuracy Neural networks Training data Data models Medical diagnosis Autonomous vehicles Image classification

来源：评论

学校读者我要写书评

暂无评论

LMCG-Net: Integrating LLMs and ECG for LVEF-Based Classification for Pacing Patients

LMCG-Net: Integrating LLMs and ECG for LVEF-Based Classifica...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Shim, Wonkyeong Park, Namjun Ko, Donggeun Gwag, Hyebin Park, Youngjun Park, Seungjung Kim, Jaekwang Sungkyunkwan University Dept. of Computer Science and Engineering Suwon Korea Republic of Aim Future Ai Seoul Korea Republic of Samsung Changwon Hospital Changwon Korea Republic of Wonju Severance Christian Hospital Kangwon-do Korea Republic of Samsung Medical Center Seoul Korea Republic of Sungkyunkwan University Convergence Program for Social Innovation School of Convergence Seoul Korea Republic of

ISBN: (纸本)9798350386226

Detecting left ventricular systolic dysfunction (LVSD) traditionally relies on expensive and specialized echocardiography, limiting accessibility for many patients. To address this, researchers explore the potential of electrocardiography (ECG), a more affordable and widely available alternative, despite its historically limited performance in cardiac dysfunction detection. In this study, we present a novel approach that integrates a 1D convolutional neural networks (CNNs) with a large-scale language model (LLM) to simultaneously analyze sequential ECG data and non-sequential clinical metadata. To validate our model's effectiveness, we conducted rigorous comparative experiments on both specially collected clinical data and public datasets, achieving an impressive AUROC of 0.97 across both. Our findings underscore the capability of ECG-based AI to accurately predict LVSD including pacemaker patients, offering a rapid and cost-effective alternative to traditional echocardiography. This innovative approach could significantly enhance early diagnosis and management of cardiac dysfunction. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Base on GAN Combined with CNN Architecture to Generate Brain Stroke CT Images 24

Base on GAN Combined with CNN Architecture to Generate Brain...

引用

8th International Conference on Medical and Health Informatics, ICMHI 2024

作者： Shih, Guan-Yu Ting, Hsien-Wei Chen, Yi-Xiang Chien, Ting-Ying Department of Computer Science and Engineering Yuan Ze University Taoyuan City Taiwan Graduate Program inBiomedical Informatics Yuan Ze University Taoyuan City Taiwan Department of Neurosurgery Taipei Hospital Ministry of Health and Welfare New Taipei City242 Taiwan Innovation Center for Big Data and Digital Convergence Yuan Ze University Taoyuan City Taiwan

ISBN: (纸本)9798400716874

In recent years, many fields have expanded their research methods through the integration of artificial intelligence. In the current medical field, it is widely used in image recognition to diagnose patient symptoms, train clinical prediction models, and improve clinical workflow processes to avoid missing treatment opportunities. These methods have all achieved good results so far. Now, generative AI is also impacting the development of the medical field, not only generating prescriptions that match patient conditions but also being applied in generating medical images for data augmentation and simulating various pathological conditions, thereby aiding in training deep learning models. Stroke is a leading cause of death or disability, ranking as the second leading cause of death worldwide, next to cancer. If cancers were categorized individually, stroke would surpass all single diseases, making it a primary focus of research in brain disease treatment. There are already studies where generative AI has been used to create medical images, which are then manually evaluated by doctors to determine if they were generated by AI. In this project, we plan to generate CT images of strokes using generative AI and use a trained model (to diagnose stroke) to improve accuracy. Through this project, we hope to solve the problem of scarce stroke CT images while also providing support for training stroke-related models. © 2024 Copyright ACM.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Constructing a Segmentation Model for Patients with Leukoaraiosis Using Deep Learning Algorithms 24

Constructing a Segmentation Model for Patients with Leukoara...

引用

8th International Conference on Medical and Health Informatics, ICMHI 2024

作者： Lo, Li-Ting Tsai, Yi-Chia Chang, Yi-Jou Ting, Hsien-Wei Chien, Ting-Ying Department of Computer Science and Engineering Yuan Ze University Taoyuan City Taiwan Graduate Program in Biomedical Informatics Yuan Ze University Taoyuan City Taiwan Department of Neurosurgery Taipei Hospital Ministry of Health and Welfare New Taipei City242 Taiwan Innovation Center for Big Data and Digital Convergence Yuan Ze University Taoyuan City Taiwan

ISBN: (纸本)9798400716874

Many patients with leukoaraiosis (LA) exhibit mild and difficult-to-detect symptoms in the early stages, and due to the lack of effective detection methods, the optimal timing for treatment is often missed. This not only makes later treatment more difficult but, in severe cases, can even lead to mild to severe dementia. Therefore, in addition to relying on radiologists to manually interpret magnetic resonance imaging (MRI) scans, finding more effective detection methods has become extremely urgent. This study aimed to combine image processing and deep learning technologies to develop a model capable of accurately identifying and segmenting white matter hyperintensities (WMHs) to assist doctors in the early diagnosis of LA. We first performed efficient preprocessing on MR image slices from two different series and used the ResNet18 model to identify key slices containing WMHs, ultimately achieving precise segmentation of WMH areas with the U-Net model. This research focused on developing innovative preprocessing techniques, improving the accuracy by 12.3% with a limited dataset through preprocessing. Ultimately, the ResNet18 model achieved an F1 score of 0.829, while the U-Net model’s intersection over union (IoU) and Dice coefficient reached 0.750 and 0.857, respectively, showing the potential of this architecture. We propose improvements in the final chapter to construct a more comprehensive model. © 2024 Copyright ACM.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Predicting Types of Urinary Crystals and Diameter Using Image Processing Techniques and Deep Learning Models 24

Predicting Types of Urinary Crystals and Diameter Using Imag...

引用

8th International Conference on Medical and Health Informatics, ICMHI 2024

作者： Huang, Yu-Ju Tseng, Yi-Shiou Chen, Tzu-Jung Chien, Ting-Ying Department of Computer Science and Engineering Yuan Ze University Taoyuan City Taiwan Department of Surgery Far Eastern Memorial Hospital Graduate Institute of Medicine Yuan Ze University Taiwan Graduate Program in Biomedical Informatics Yuan Ze University Taoyuan City Taiwan Innovation Center for Big Data and Digital Convergence Yuan Ze University Taoyuan City Taiwan

ISBN: (纸本)9798400716874

Kidney stones are primarily crystals formed from ion oversaturation in urine. Currently, the diagnosis of kidney stones involves experienced professionals manually interpreting images of urinary crystals under a microscope. In recent years, with the rapid development of artificial intelligence (AI), many researchers have proposed various methods to assist in diagnosis, hoping to improve medical quality. This study explored the application of AI in the medical field, aiming to automate the interpretation of urinary crystal images, such as red blood corpuscles (RBCs), calcium oxalate monohydrate (COM), calcium oxalate dihydrate octahedron (COD-OCT), calcium oxalate dihydrate dodecahedron (COD-DOD), debris, and others, and to use image processing techniques to classify urinary crystal types. This research is expected to assist medical personnel in quickly determining the categories of urinary crystals. Ultimately, our model achieved an F1-score of 0.93, indicating that this architecture has significant potential. © 2024 Copyright ACM.

关键词： Deep learning

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：