检索结果-内蒙古大学图书馆

Improving Interpretability of Radiology Report-based Pediatric Brain Tumor Pathology Classification and Key-phrases Extraction Using Large Language Models

Improving Interpretability of Radiology Report-based Pediatr...

引用

2024 IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2024

作者： Zhao, Chen Kudus, Kareem Ketabi, Sara Namdar, Khashayar Wagner, Matthias W. Ertl-Wagner, Birgit B. Khalvati, Farzad University of Toronto Department of Mechanical and Industrial Engineering TorontoON Canada Neurosciences & Mental Health Research Program Research Institute The Hospital for Sick Children TorontoON Canada Institute of Diagnostic and Interventional Neuroradiology University Hospital Augsburg Germany The Hospital for Sick Children Department of Diagnostic & Interventional Radiology TorontoON Canada University of Toronto Department of Medical Imaging TorontoON Canada University of Toronto Department of Computer Science TorontoON Canada Institute of Medical Science University of Toronto TorontoON Canada

ISBN: (纸本)9798350351552

Radiology reports are crucial for bridging the expertise of radiologists and other clinicians. Machine Learning models trained on these reports have shown promising performance in various downstream clinical tasks, such as predicting the necessity of future follow-up procedures, based on past radiology reports. However, for clinicians to adopt these models and for radiologists to validate the results, interpretability of the model is essential. In this study, we train BERT models on radiology reports to classify pediatric brain tumor pathologies. These large language models enable accurate report-level classification, without the need for costly word-level annotations. To identify and extract keywords and key-phrases related to distinct pathologies from radiology reports, we used a modified Term Frequency-Inverse Document Frequency to determine phrase importance based on prevalence and attributions scores. We achieved an overall multiclass Area Under Receiver Operating Characteristic Curve (AUROC) of 79.57% using ClinicaiBERT. Moreover, the per-class AUROC values were 86%, 71.2%, and 81.5%, for 'Pilocytic Astrocytoma', 'Low-Grade Astrocytoma', and 'Other' pathologies, respectively. Our explainability analysis identified hypotonia and mesencephalon as the most important terms for 'Pilocytic Astrocytoma' and 'Low-Grade Astrocytoma', respectively. © 2024 IEEE.

关键词： Radiology

来源：评论

学校读者我要写书评

暂无评论

Multi-channel Convolutional Neural Network for Hate Speech Detection in Social Media 9th

Multi-channel Convolutional Neural Network for Hate Speech D...

引用

9th EAI International Conference on Advancement of science and Technology, ICAST 2021

作者： Abebaw, Zeleke Rauber, Andreas Atnafu, Solomon IT Doctoral Program Addis Ababa University Addis Ababa Ethiopia Institute of Information Systems Engineering Technical University of Vienna Vienna Austria Department of Computer Science Addis Ababa University Addis Ababa Ethiopia

ISBN: (纸本)9783030937089

As online social media content continues to grow, so does the spread of hate speech. Hate speech has devastating consequences unless it is detected and monitored early. Recently, deep neural network-based hate speech detection models, particularly conventional single-channel Convolutional Neural Network (CNN), have achieved remarkable performance. However, the effectiveness of the models depends on the type of language they are trained on and the training data size. We argue that the effectiveness of the models could further be enhanced if we use multi-channel CNN models even for under-resourced languages that have limited training data size. This is because the single-channel CNN might fail to consider the potential effect of multiple channels to generate better features, which is not well investigated for hate speech detection. Therefore, in this work, we explore the use of multi-channel CNN to extract better features from different channels in an end-to-end manner on top of a word2vec embedding layer. Tested on a new small-scale Amharic hate speech dataset containing 2000 annotated social media comments, the experimental results show that the proposed multi-channel CNN model outperforms the single-channel CNN models but underperform from the baseline Support Vector Machine (SVM) with an average F-score of 81.3%, 78.2%, and 92.5% respectively. The finding of the study implies that the proposed MC-CNN model can be used as an alternative solution for hate speech detection using a deep learning approach when dataset scarcity is an issue. © 2022, ICST Institute for computer sciences, Social Informatics and Telecommunications engineering.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Convolutional Neural Network Using Res-Net For Organic and Anorganic Waste Classification

Convolutional Neural Network Using Res-Net For Organic and A...

引用

International Conference of computer science and Information Technology

作者： Abba Suganda Girsang Rezky Yunanda Muhammad Edo Syahputra Ezra Peranginangin Computer Science Department BINUS Graduate Program - Master of Computer Science Bina Nusantara University Jakarta Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia Industrial Engineering Department Faculty of Engineering Bina Nusantara University Jakarta Indonesia

ISBN: (纸本)9798350399080

Convolutional Neural network is state of the art of image recognition or image classification. However to build the robust model using CNN needs many parameters adjusted, and choosing the good combination hyperparameter which impacts taking much computation time. Genetic algorithm is one method metaheuristic which is robust for choosing the combinatorial possible hyperparameter. This model also uses ResNet-50 which is a pretrained model of convolutional neural network that consists 50 layers. By using a pretrained like ResNet-50, it will increase the performance model. CNN-ResNet with efficient genetic algorithm (EGA) to optimize the hyperparameter. The EGA algorithm utilizes transfer learning techniques in its algorithm so that the optimization process on CNN can achieve unified accuracy values quickly. The best performance model optimized using EGA outperformed the ResNet-50 model and the model optimized using GA and VLGA in classifying organic and inorganic materials. The accuracy value obtained from EGA is 97.53% with a loss of 0.08.

关键词： Image recognition Computational modeling Transfer learning Metaheuristics Inorganic materials Iron Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

Message from program Committee

ICCQ 2022 - Proceedings of the 2nd International Conference ...

引用

ICCQ 2022 - Proceedings of the 2nd International Conference on Code Quality 2022年 2022-January卷 9-9页

作者： Succi, Giancarlo Program Committee Faculty of Computer Science and Software Engineering Innopolis University Russi Italy

来源：评论

学校读者我要写书评

暂无评论

Impact of fabrication methods on binder distribution and charge transport in composite cathodes of all-solid-state batteries

引用

Materials Futures 2023年第4期2卷 127-135页

作者： Benjamin Emley Chaoshan Wu Lihong Zhao Qing Ai Yanliang Liang Zhaoyang Chen Liqun Guo Tanguy Terlier Jun Lou Zheng Fan Yan Yao Materials Science and Engineering Program and Texas Center for Superconductivity at the University of Houston University of Houston4726 Calhoun RdHoustonTX 77204United States of America Department of Electrical and Computer Engineering University of Houston4726 Calhoun RdHoustonTX 77204United States of America Department of Materials Science and Nano Engineering Rice University6100 Main StreetHoustonTX 77005United States of America SIMS Laboratory Shared Equipment AuthorityRice UniversityHoustonTX 77005United States of America Department of Engineering Technology University of HoustonHoustonTX 77204United States of America

The manufacturing process of all-solid-state batteries necessitates the use of polymer ***,these binders,being ionic insulators by nature,can adversely affect charge transport within composite cathodes,thereby impacting the rate performance of the *** this work,we aim to investigate the impact of fabrication methods,specifically the solvent-free dry process versus the slurry-cast wet process,on binder distribution and charge transport in composite cathodes of solid-state *** the dry process,the binder forms a fibrous network,while the wet process results in binder coverage on the surface of cathode active *** difference in microstructure leads to a notable 20-fold increase in ionic conductivity in the dry-processed ***,the cells processed via the dry method exhibit higher capacity retention of 89%and 83%at C/3 and C/2 rates,respectively,in comparison to 68%and 58%for the wet-processed cells at the same *** findings provide valuable insights into the influence of fabrication methods on binder distribution and charge transport,contributing to a better understanding of the binder’s role in manufacturing of all-solid-state batteries.

关键词： solid-state battery solvent-free dry process polymer binder lithium thiophosphate electrolyte

来源：评论

学校读者我要写书评

暂无评论

GraphT5: Unified Molecular Graph-Language Modeling via Multi-Modal Cross-Token Attention

arXiv

引用

arXiv 2025年

作者： Kim, Sangyeup Kim, Nayeon Piao, Yinhua Kim, Sun Computer Science and Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of AIGENDRUG Co. Ltd Korea Republic of

Molecular language modeling tasks such as molecule captioning have been recognized for their potential to further understand molecular properties that can aid drug discovery or material synthesis based on chemical reactions. Unlike the common use of molecule graphs in predicting molecular properties, most methods in molecular language modeling rely heavily on SMILES sequences. This preference is because the task involves generating a sequence of multiple tokens using transformer-based models. Therefore, a main challenge is determining how to integrate graph data, which contains structural and spatial information about molecules, with text data. In addition, simply using both 1D SMILES text and 2D graph as inputs without addressing how they align and represent the molecule structure in different modalities makes it challenging to fully utilize structural knowledge about molecules. To this end, we propose GraphT5, a multi-modal framework that integrates 1D SMILES text and 2D graph representations of molecules for molecular language modeling. Specifically, we introduce a novel cross-token attention module in GraphT5 to bridge the gap arising from the fundamental differences between the two modalities of molecule representations. Cross-token attention exploits implicit information between SMILES and graphs of molecules, resulting from their interactions at a fine-grained token level that benefits molecular language modeling. Extensive experiments including molecule captioning, IUPAC name prediction tasks, and case studies show that our GraphT5 outperforms the latest baseline approaches, which validates the effectiveness of our GraphT5 in sufficiently utilizing 1D SMILES text and 2D graph representations. © 2025, CC BY-NC-ND.

关键词： Unified Modeling Language

来源：评论

学校读者我要写书评

暂无评论

Meal Image Recognition and Healthy Meal Combination Recommendation System Integrated with Generative Artificial Intelligence

Meal Image Recognition and Healthy Meal Combination Recommen...

引用

IEEE Eurasia Conference on Biomedical engineering, Healthcare and Sustainability (ECBIOS)

作者： Chin-Chih Chang Chi-Hung Wei Sean Hsiao Chyuan-Huei Thomas Yang Department of Computer Science and Information Engineering Chung Hua University Hsinchu Taiwan Ph.D. Program in Engineering Science Chung Hua University Hsinchu Taiwan Bachelor's Program in Applied Foreign Language and Smart Commerce Chung Hua University Hsinchu Taiwan School of Information Engineering Shandong Vocational and Technical University of International Studies Shandong China

ISBN: (数字)9798350396133

ISBN: (纸本)9798350396140

People increasingly prioritize a balanced diet to enhance well-being, yet making informed dietary choices remains challenging amidst the abundance of options. To address this, we developed a meal image recognition and healthy meal combination recommender system, integrating generative artificial intelligence (AI). Convolutional neural networks were used for precise meal image recognition to identify diverse food items. The generative AI-augmented recommendation engine offers personalized meal suggestions aligned with nutritional goals and dietary preferences, utilizing a nutrition knowledge base to ensure overall well-being. The system's feasibility was validated, illustrating its excellence in meal recognition accuracy, recommendation diversity, and user engagement. By integrating generative AI, the system shows its potential to enhance dietary recommendations and public health.

关键词： Image recognition Accuracy Generative AI Knowledge based systems Convolutional neural networks Sustainable development Public healthcare Recommender systems Engines Biomedical engineering

来源：评论

学校读者我要写书评

暂无评论

Optimal deployment of private 5G multi-access edge computing systems at smart factories: Using hybrid crow search algorithm

引用

Journal of Network and computer Applications 2024年 227卷

作者： Lin, Chun-Cheng Deng, Der-Jiunn Hsieh, Li-Tsung Pan, Pei-Tzu Department of Industrial Engineering and Management National Yang Ming Chiao Tung University Hsinchu300 Taiwan International Master Program of Smart Manufacturing and Applied Information Science National Chin-Yi University of Technology Taichung411 Taiwan Department of Computer Science and Information Engineering National Changhua University of Education Changhua500 Taiwan

In smart factories, an increasing number of mobile intelligent devices are deployed to meet the growing demands for flexible manufacturing. These devices, equipped with various sensors, synchronize a substantial amount of data with cloud servers for real-time monitoring and control. Fifth generation mobile networks (5G) combined with multi-access edge computing (MEC) provide the capabilities for multiple connections, large-scale data transmission, and efficient response times, becoming the mainstream network architectures for the needs. Additionally, with the emerging trend of energy-saving, recent works have addressed energy consumption as a significant cost factor. However, most previous studies tended to focus solely on cloud servers, neglecting energy consumption at edge computing and underestimating the impact of heterogeneous mobile device. As the scope of smart factory networks continues to expand, the challenges become increasingly critical. This study presents a private 5G MEC system architecture tailored for smart panel factories. The concerned problem is formulated as an integer programming model, which determines deployment locations and quantities for MEC servers and 5G small cells, including picocells and femtocells, so as to minimize the overall deployment costs while meeting the constraints of connectivity, capacity, latency, coverage, serviceability, and energy consumption. Since the edge server deployment problem is NP-hard, this study further proposes a hybrid metaheuristic algorithm, CSAVNS, which combines the strengths of crow search algorithm (CSA) and variable neighborhood search (VNS). In the global search phase of CSA, the VNS local search is introduced to enhance the algorithm's capabilities. Experimental analysis demonstrates that the proposed CSAVNS outperforms other algorithms in terms of solution solving capabilities. © 2024 Elsevier Ltd

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

A Study on Medicinal Plant Leaf Recognition Using Artificial Intelligence 3

A Study on Medicinal Plant Leaf Recognition Using Artificial...

引用

3rd International Conference on Informatics, Multimedia, Cyber, and Information System, ICIMCIS 2021

作者： Ayumi, Vina Ermatita, Ermatita Abdiansah, Abdiansah Noprisson, Handrie Purba, Mariana Utami, Marissa Universitas Sriwijaya Doctoral Program in Engineering Palembang Indonesia Universitas Sriwijaya Faculty of Computer Science Palembang Indonesia Universitas Mercu Buana Faculty of Computer Science Jakarta Indonesia Program of Informatics Management Politeknik Anika Palembang Indonesia Muhammadiyah University of Bengkulu Faculty of Engineering Bengkulu Indonesia

ISBN: (纸本)9781665427333

Medicinal plant recognition manually takes a lot of time and money. Moreover, to reduce these resources, some researchers propose to implement artificial intelligence technology. This paper aims are to conduct a systematic literature review of medicinal plant leaf recognition published in the last two years (2019-2020) from IEEE, Springer and science Direct. We obtained 15 studies in the field of medicinal plant leaf recognition using artificial intelligence. The dataset used for medicinal plant leaf recognition is mostly used private dataset, however, there are public dataset named Leaf, Flavia, Swedish dataset. We also found robust method that can be used for medicinal plant leaf recognition is Multichannel Modified Local Gradient Pattern (MCMLGP) and Gray Level Co-Occurrence Matrix (GLCM) as feature extraction;and Convolutional Neural Network (CNN), Multi-Layer Perceptron trained with Backpropagation algorithm (MLP-BP), Support Vector Machine (SVM), and Transfer Learning (VGG19) as classifier. © 2021 IEEE.

关键词： Image processing

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Brain Cortical Connectivity in Drug Abusers Using EEG Data

Evaluation of Brain Cortical Connectivity in Drug Abusers Us...

引用

Iranian Conference of Biomedical engineering (ICBME)

作者： Behzad Yousefipour Sepehr Shirani Maryam Mohebbi Peyman Hasani Abharian Department of Biomedical Engineering K.N.Toosi University of Technology Tehran Iran Department of Computer Science Nottingham Trent University Nottingham UK Cognitive Neuroscience Institute for Cognitive Science Studies Tehran Iran

This study implements Electroencephalogram (EEG) signals in a data-driven approach based on functional brain connectivity to identify brain abnormal cortical connectivity in opioid abusers. The prime objective here is to identify opioid addiction. A 19-channel resting-state EEG signal was recorded from 22 opioid addicts and 22 normal, closely matched individuals. First, brain networks associated with two groups were constructed using the Mutual Information (MI) metric in frequency bands of the delta, theta, alpha, beta, gamma, and wideband. The groups were then divided using discriminative graph features. The results from the Support Vector Machine (SVM) classifier, evaluated by the Leave-one-out cross-validation method, achieved accuracy, sensitivity, and specificity of 100% and an F-score of 1.00 in the gamma frequency band using the AdaBoost subsampling approach. The findings indicate that the implemented pipeline, based on MI quantity between EEG data channels and combined with graph metrics, can help detect opioid addiction.

关键词： Support vector machines Measurement Drugs Addiction Pipelines Sensitivity and specificity Electroencephalography

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：