检索结果-内蒙古大学图书馆

Multimodal E-Commerce Product Classification Using Hierarchical Fusion 2

学校读者我要写书评

暂无评论

Multimodal E-Commerce Product Classification Using Hierarchi...

2nd IEEE Conference on Information Technology and Data Science, CITDS 2022

作者： Tashu, Tsegaye Misikir Fattouh, Sara Kiss, Peter Horvath, Tomas Elte - Eötvös Loránd University Faculty of Informatics Department of Data Science and Engineering Pázmány Péter sétány 1/C Budapest1117 Hungary Faculty of Science Institute of Computer Science Pavol Jozef Šafárik University Jesenná5 Košice040 01 Slovakia College of Informatics Kombolcha Institute of Technology Wollo University Kombolcha208 Ethiopia

ISBN: (数字)9781665496537

ISBN: (纸本)9781665496537

In this work, we present a multi-modal model for commercial product classification, that combines features extracted by multiple neural network models from textual (Camem-BERT and FlauBERT) and visual data (SE-ResNeXt-50), using simple fusion techniques. The proposed method significantly outperformed the performance of the unimodal models, as well as the reported performance of similar models on our specific task. We made experiments with multiple fusing techniques, and found, that the best preforming technique to combine the individual embedding of the unimodal network is based on the combination of concatenation and averaging the feature vectors. Each modality complemented the shortcomings of the other modalities, demonstrating that increasing the number of modalities can be an effective method for improving the performance of multi-label and multimodal classification problems. © 2022 IEEE.

关键词： Neural network models

Integrating Dynamic Correlation Shifts and Weighted Benchmarking in Extreme Value Analysis

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Panagoulias, Dimitrios P. Sarmas, Elissaios Marinakis, Vangelis Virvou, Maria Tsihrintzis, George A. Department of Informatics University of Piraeus Karaoli ke Dimitriou 80 Piraeus18534 Greece Decision Support Systems Lab School of Electrical and Computer Engineering National Technical University of Athens Ir. Politechniou 9 Zografou Athens157 73 Greece

This paper presents an innovative approach to Extreme Value Analysis (EVA) by introducing the Extreme Value Dynamic Benchmarking Method (EVDBM). EVDBM integrates extreme value theory to detect extreme events and is coupled with the novel Dynamic Identification of Significant Correlation (DISC)Thresholding algorithm, which enhances the analysis of key variables under extreme conditions. By integrating return values predicted through EVA into the benchmarking scores, we are able to transform these scores to reflect anticipated conditions more accurately. This provides a more precise picture of how each case is projected to unfold under extreme conditions. As a result, the adjusted scores offer a forward-looking perspective, highlighting potential vulnerabilities and resilience factors for each case in a way that static historical data alone cannot capture. By incorporating both historical and probabilistic elements, the EVDBM algorithm provides a comprehensive benchmarking framework that is adaptable to a range of scenarios and contexts. The methodology is applied to real PV data, revealing critical low - production scenarios and significant correlations between variables, which aid in risk management, infrastructure design, and long-term planning, while also allowing for the comparison of different production plants. The flexibility of EVDBM suggests its potential for broader applications in other sectors where decision-making sensitivity is crucial, offering valuable insights to improve outcomes. © 2024, CC BY.

关键词： Predictive analytics

Advancing Medical Imaging with Language Models: A Journey from N-grams to ChatGPT

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hu, Mingzhe Pan, Shaoyan Li, Yuheng Yang, Xiaofeng Department of Computer Science and Informatics Emory University AtlantaGA United States Department of Biomedical Engineering Emory University AtlantaGA United States Department of Radiation Oncology Winship Cancer Institute School of Medicine Emory University AtlantaGA United States

In this paper, we aimed to provide a review and tutorial for researchers in the field of medical imaging using language models to improve their tasks at hand. We began by providing an overview of the history and concepts of language models, with a special focus on large language models. We then reviewed the current literature on how language models are being used to improve medical imaging, emphasizing different applications such as image captioning, report generation, report classification, finding extraction, visual question answering, interpretable diagnosis and more for various modalities and organs. The ChatGPT was specially highlighted for researchers to explore more potential applications. We covered the potential benefits of accurate and efficient language models for medical imaging analysis, including improving clinical workflow efficiency, reducing diagnostic errors, and assisting healthcare professionals in providing timely and accurate diagnoses. Overall, our goal was to bridge the gap between language models and medical imaging and inspire new ideas and innovations in this exciting area of research. We hope that this review paper will serve as a useful resource for researchers in this field and encourage further exploration of the possibilities of language models in medical imaging. Copyright © 2023, The Authors. All rights reserved.

关键词： Image enhancement

Enhancing Generalization via Sharpness-Aware Trajectory Matching for Dataset Condensation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Gao, Boyan Zhao, Bo Gowda, Shreyank N. Xing, Xingrun Yang, Yibo Hospedales, Timothy Clifton, David A. Department of Engineering Science University of Oxford United Kingdom School of Artificial Intelligence Shanghai Jiao Tong University China School of Computer Science University of Nottingham United Kingdom Institute of Automation Chinese Academy of Sciences China School of Informatics University of Edinburgh United Kingdom

Dataset condensation aims to synthesize datasets with a few representative samples that can effectively represent the original datasets. This enables efficient training and produces models with performance close to those trained on the original sets. Most existing dataset condensation methods conduct dataset learning under the bilevel (inner- and outer-loop) based optimization. However, the preceding methods perform with limited dataset generalization due to the notoriously complicated loss landscape and expensive time-space complexity of the inner-loop unrolling of bilevel optimization. These issues deteriorate when the datasets are learned via matching the trajectories of networks trained on the real and synthetic datasets with a long horizon inner-loop. To address these issues, we introduce Sharpness-Aware Trajectory Matching (SATM), which enhances the generalization capability of learned synthetic datasets by optimising the sharpness of the loss landscape and objective simultaneously. Moreover, our approach is coupled with an efficient hypergradient approximation that is mathematically well-supported and straightforward to implement along with controllable computational overhead. Empirical evaluations of SATM demonstrate its effectiveness across various applications, including in-domain benchmarks and out-of-domain settings. Moreover, its easy-to-implement properties afford flexibility, allowing it to integrate with other advanced sharpness-aware minimizers. Our code will be released. © 2025, CC BY.

关键词：

The role of input imaging combination and ADC threshold on segmentation of acute ischemic stroke lesion using U-Net

学校读者我要写书评

暂无评论

European Radiology 2023年第9期33卷 6157-6167页

作者： Li, Ya-Hui Lin, Shao-Chieh Chung, Hsiao-Wen Chang, Chia-Ching Peng, Hsu-Hsia Huang, Teng-Yi Shen, Wu-Chung Tsai, Chon-Haw Lo, Yu-Chien Lee, Tung-Yang Juan, Cheng-Hsuan Juan, Cheng-En Chang, Hing-Chiu Liu, Yi-Jui Juan, Chun-Jung Graduate Institute of Biomedical Electronics and Bioinformatics National Taiwan University Taipei Taiwan Department of Medical Imaging China Medical University Hsinchu Hospital No. 199 Sec. 1 Xinglong Rd. Zhubei City Hsinchu County 302 Hsinchu Taiwan Ph.D. Program in Electrical and Communication Engineering Feng Chia University Taichung Taiwan Department of Electrical Engineering National Taiwan University Taipei Taiwan Department of Management Science National Yang Ming Chiao Tung University Hsinchu Taiwan Department of Biomedical Engineering and Environmental Sciences National Tsing Hua University Hsinchu Taiwan Department of Electrical Engineering National Taiwan University of Science and Technology Taipei Taiwan Department of Radiology School of Medicine College of Medicine China Medical University Taichung Taiwan Department of Medical Imaging Medical University Hospital Taichung Taiwan Department of Neurology China Medical University Hospital Taichung Taiwan Cheng Ching Hospital Taichung Taiwan Master’s Program of Biomedical Informatics and Biomedical Engineering Feng Chia University Taichung Taiwan Department of Automatic Control Engineering Feng Chia University No. 100 Wenhwa Rd. Seatwen Taichung 40724 Taiwan Department of Biomedical Engineering The Chinese University of Hong Kong ERB1112 11/F William M.W. Mong Engineering Building N.T Shatin Hong Kong Multi-Scale Medical Robotics Center The Chinese University of Hong Kong N.T Shatin Hong Kong Department of Biomedical Engineering National Defense Medical Center Taipei Taiwan Department of Computer Science and Information Engineering National Taiwan University Taipei Taiwan

Background: To evaluate the effect of the weighting of input imaging combo and ADC threshold on the performance of the U-Net and to find an optimized input imaging combo and ADC threshold in segmenting acute ischemic stroke (AIS) lesion. Methods: This study retrospectively enrolled a total of 212 patients having AIS. Four combos, including ADC-ADC-ADC (AAA), DWI-ADC-ADC (DAA), DWI-DWI-ADC (DDA), and DWI-DWI-DWI (DDD), were used as input images, respectively. Three ADC thresholds including 0.6, 0.8 and 1.8 × 10^–3 mm²/s were applied. Dice similarity coefficient (DSC) was used to evaluate the segmentation performance of U-Nets. Nonparametric Kruskal–Wallis test with Tukey–Kramer post-hoc tests were used for comparison. A p <.05 was considered statistically significant. Results: The DSC significantly varied among different combos of images and different ADC thresholds. Hybrid U-Nets outperformed uniform U-Nets at ADC thresholds of 0.6 × 10^–3 mm²/s and 0.8 × 10^–3 mm²/s (p <.001). The U-Net with imaging combo of DDD had segmentation performance similar to hybrid U-Nets at an ADC threshold of 1.8 × 10^–3 mm²/s (p =.062 to 1). The U-Net using the imaging combo of DAA at the ADC threshold of 0.6 × 10^–3 mm²/s achieved the highest DSC in the segmentation of AIS lesion. Conclusions: The segmentation performance of U-Net for AIS varies among the input imaging combos and ADC thresholds. The U-Net is optimized by choosing the imaging combo of DAA at an ADC threshold of 0.6 × 10^–3 mm²/s in segmentating AIS lesion with highest DSC. Key Points: • Segmentation performance of U-Net for AIS differs among input imaging combos. • Segmentation performance of U-Net for AIS differs among ADC thresholds. • U-Net is optimized using DAA with ADC = 0.6 × 10^–3mm²/s. © 2023, The Author(s), under exclusive licence to European Society of Radiology.

关键词： Deep Learning Diffusion Magnetic Resonance Imaging Ischemic Stroke Neural Networks, computer Retrospective Study

Machine Learning for Predicting the Severity of Road Accident Victims at a University Hospital Emergency Center

学校读者我要写书评

暂无评论

Machine Learning for Predicting the Severity of Road Acciden...

International Conference on Big Data and Smart Computing (BIGCOMP)

作者： Wiriya Mahikul Thamonwan Thongbun Asama Tungparamutsakul Pornwanut Kitudom Phond Phunchongharn Shinanang Promkaew Nipon Singkam Athita Onuean Chulabhorn Royal Academy Princess Srisavangavadhana College of Medicine Bangkok Thailand Department of Computer Engineering King Mongkut's University of Technology Thonburi Bangkok Thailand Faculty of Medicine Burapha University Chonburi Thailand Faculty of Informatics Burapha University Chonburi Thailand

Increasingly frequent road accidents in Thailand have resulted in overcrowding in hospital emergency rooms, forcing medical personnel to work under urgent conditions. This can lead to difficulties in triaging and prioritizing accident victims, potentially affecting patient outcomes. In this study, machine learning techniques were applied to predict the severity of road accident victims in Thailand using personal information, injury location, and risk behaviors. Support Vector Machine (SVM), Random Forest, and XGBoost models were compared for prediction accuracy. The XGBoost model was found to be the most suitable and effective, with an accuracy of 0.53, precision of 0.55, recall of 0.53, and an F -score of 0.53 However, the most accurate prediction was for emergencies, with an accuracy of 0.75, a precision of 0.62, a recall of 0.75, and an F-score of 0.68. The most influential factor in the model was head injuries. This XGBoost model was then used to develop a web application called “ForCrash”, which provides predictions of accident severity. User satisfaction with the application was high, as it was found to be visually appealing, accurate, trustworthy, and user-friendly. The application also proved to be a valuable tool for medical personnel, especially in hospitals lacking emergency medical expertise. Additionally, it facilitated the systematic collection of patient data for future studies and research.

关键词：

A high granularity state-space method for contaminant detection and isolation in intelligent buildings 16

学校读者我要写书评

暂无评论

A high granularity state-space method for contaminant detect...

16th Conference of the International Society of Indoor Air Quality and Climate: Creative and Smart Solutions for Better Built Environments, Indoor Air 2020

作者： Kyriacou, Alexis Michaelides, Michalis Panayiotou, Christos Polycarpou, Marios KIOS Research and Innovation Center of Excellence The Department of Electrical and Computer Engineering University of Cyprus Nicosia Cyprus Department of Electrical Engineering Computer Engineering and Informatics Cyprus University of Technology Limassol Cyprus

ISBN: (纸本)9781713823605

Conservation of the Indoor Air Quality is essential in modern, energy efficient, intelligent buildings. However, accidents or malicious acts that often result in various airborne contaminant releases can endanger the occupants' wellbeing. In this work, an indoor air quality monitoring methodology is presented for detecting and localizing a contaminant source in the 3D indoor environment. Specifically, a state-space model is used to describe the contaminant dispersion in a zone which is discretised into multiple cuboid cells. The airflow exchange between the cells is computed based on a 3D discretized coarse-grid CFD analysis. A contaminant detection and localization methodology that considers modelling uncertainty and measurement noise, is adapted and applied to the CFD-based state-space model for detecting the existence of a possible contaminant source and estimating its location within the zone. The performance of the approach is illustrated through simulation examples. © 2020 16th Conference of the International Society of Indoor Air Quality and Climate: Creative and Smart Solutions for Better Built Environments, Indoor Air 2020. All rights reserved.

关键词： Uncertainty analysis

BreastSAM: A Study of Segment Anything Model for Breast Tumor Detection in Ultrasound Images

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hu, Mingzhe Li, Yuheng Yang, Xiaofeng Department of Computer Science and Informatics Emory University AtlantaGA United States Department of Biomedical Engineering Emory University AtlantaGA United States Department of Radiation Oncology Winship Cancer Institute School of Medicine Emory University AtlantaGA United States

Breast cancer is one of the most common cancers among women worldwide, with early detection significantly increasing survival rates. Ultrasound imaging is a critical diagnostic tool that aids in early detection by providing real-time imaging of the breast tissue. We conducted a thorough investigation of the Segment Anything Model (SAM) for the task of interactive segmentation of breast tumors in ultrasound images. We explored three pre-trained model variants: ViT_h, ViT_l, and ViT_b, among which ViT_l demonstrated superior performance in terms of mean pixel accuracy, Dice score, and IoU score. The significance of prompt interaction in improving the model's segmentation performance was also highlighted, with substantial improvements in performance metrics when prompts were incorporated. The study further evaluated the model's differential performance in segmenting malignant and benign breast tumors, with the model showing exceptional proficiency in both categories, albeit with slightly better performance for benign tumors. Furthermore, we analyzed the impacts of various breast tumor characteristics - size, contrast, aspect ratio, and complexity - on segmentation performance. Our findings reveal that tumor contrast and size positively impact the segmentation result, while complex boundaries pose challenges. The study provides valuable insights for using SAM as a robust and effective algorithm for breast tumor segmentation in ultrasound images. Copyright © 2023, The Authors. All rights reserved.

关键词： Aspect ratio

Indoor Navigation Using Augmented Reality For PENS Postgraduate Building

学校读者我要写书评

暂无评论

Indoor Navigation Using Augmented Reality For PENS Postgradu...

International Electronics Symposium (IES)

作者： Nike Suryawati Sritrusta Sukaridhoto Hestiasari Rante Evianita Dewi Fajrianti Oktafian Sultan Hakim Creative Multimedia Department Politeknik Elektronika Negeri Surabaya Jl. Raya ITS Surabaya Indonesia Department of Information and Communication Systems Okayama University Tsushimanaka Kita Ward Okayama Japan Department of Informatics and Computer Engineering Politeknik Elektronika Negeri Surabaya Indonesia

The use of navigation systems is widespread in assisting individuals with finding their desired destination, visualizing the route, and measuring the distance traveled. In the specific context of navigating indoors, the postgraduate building at Politeknik Elektronika Negeri Surabaya lacks real-time digital directions, which can lead to confusion among users. To address this issue, we developed an Augmented Reality (AR) based navigation system that aims to provide a user-friendly solution and eliminate any potential confusion. Our research focuses on the development of an interior navigation system that leverages AR technology to enhance the user experience. We implemented QR codes to establish the initial position of each room, and conducted a confusion matrix test to validate the accuracy of scanning each QR code, resulting in a high accuracy rate of 96.6 percent. These findings confirm the reliability and effectiveness of our navigation system.

关键词：