检索结果-内蒙古大学图书馆

EPR analysis and spectral deconvolution of alanine dosimetric properties

Journal of Radioanalytical and Nuclear Chemistry 2025年第4期334卷 2657-2670页

作者： Zahiri, Fatna Mikou, Mohammed Mamadou Saidou, Abdoul Karim Belahmar, Ayoub El Baydaoui, Redouane Laboratory of Mathematics Computer Science and Engineering Sciences Faculty of Science and Technology Hassan First University of Settat B.P. 577 Settat26000 Morocco Hassan First University of Settat B.P. 555 Settat26000 Morocco

Precise dose monitoring in radiotherapy is essential to ensure treatment efficacy and the protection of healthy tissues. Although the Alanine/EPR dosimetry system has demonstrated dosimetric properties, further analysis of this dosimetry system, taking into account the characteristics of the different contributions to the EPR spectrum of irradiated alanine, will be important in determining the system's performance and ensuring its reliability for dosimetric applications in radiotherapy and hadrontherapy. In this work, we used the deconvolution method, based on the Levenberg–Marquardt algorithm, to determine the contributions constituting the irradiated alanine spectrum. Fourteen contributions were revealed. Analysis of the comportment of these contributions as a function of irradiation dose and post-irradiation storage time revealed quasi-linear evolution according to irradiation dose, and good stability of the produced free radicals during the post-irradiation storage period. This result approves the potential application of the Alanine/EPR system for dose control in radiotherapy. © Akadémiai Kiadó Zrt 2025.

关键词： Radiotherapy

来源：评论

学校读者我要写书评

暂无评论

Evaluating BERT-based language models for detecting misinformation

引用

Neural Computing and Applications 2025年第16期37卷 9937-9968页

作者： Anggrainingsih, Rini Hassan, Ghulam Mubashar Datta, Amitava Information Technology and Data Science Faculty Sebelas Maret University Ir Sutami 36A Central Java Surakarta57126 Indonesia Computer Science and Software Engineering The University of Western Australia 35 Stirling Highway PerthWA6009 Australia

Online misinformation poses a significant challenge due to its rapid spread and limited supervision. To address this issue, automated rumour detection techniques are essential for countering the negative impact of false information. Previous research primarily focussed on extracting text features, which proved time-consuming and less effective. In this study, we contribute substantially to two domains: rumour detection on Twitter and the evaluation of text embeddings. We thoroughly analyse rumour detection models and compare the quality of text embeddings generated by various fine-tuned BERT-based models. Our findings indicate that our proposed models outperform existing techniques. Notably, when we test these models on combined datasets, we observe significant performance improvements with larger training and testing data sizes. We conclude that carefully considering the dataset, data splitting, and classification techniques is crucial for evaluating solution performance. Additionally, we find that differences in the quality of text embeddings between RoBERTa, BERT, and DistilBERT are insignificant. This challenges existing assumptions and highlights the need for future research to explore these nuances further. © The Author(s) 2025.

关键词： BERT-based models Comparative analysis Rumour detection Text embeddings

来源：评论

学校读者我要写书评

暂无评论

End-to-end optimized ensemble model for Alzheimer’s disease detection 19

End-to-end optimized ensemble model for Alzheimer’s disease...

引用

19th International Conference on Ubiquitous Information Management and Communication, IMCOM 2025

作者： Hassan, Abdelrahman Abuhmed, Tamer El-Sappagh, Shaker Faculty of Computer Science and Engineering Galala University Egypt College of Computing and Informatics Sungkyunkwan University Korea Republic of

ISBN: (纸本)9798331507817

Alzheimer's disease (AD) is a progressive neurodegenerative disorder with an increasing prevalence among the elderly, making early and accurate diagnosis critical for effective intervention and management. This paper introduces an end-to-end machine learning pipeline optimized for detecting AD using multimodal data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The study proposes an ensemble model that integrates key modalities—demographics, cognitive scores, neuropsychological assessments, and MRI data—via early fusion to enhance AD detection. Various machine learning models, including Logistic Regression, Gradient Boosting, AdaBoost, XGBoost, Support Vector Machine, and Neural Networks, are compared, and feature selection is optimized through Recursive Feature Elimination, ANOVA F-value, and mutual information methods, with mutual information showing the most efficacy. With further hyperparameter tuning, the highest accuracy is achieved at 95.85%, with Gradient Boosting as the top-performing model. Additional experiments assess and evaluate individual modality contributions, underscoring the predictive value of cognitive scores and neuropsychological features. To enhance clinical interpretability, we incorporate model explainability using LIME and SHAP, which identify key features influencing model predictions across different AD stages. Explainable AI (XAI) is essential in healthcare for providing transparency and supporting clinician confidence by highlighting which biomarkers, such as MMSE, CDRSB, and PTAU, contribute most to diagnostic outcomes. This study demonstrates classical machine learning's potential in AD diagnostics, offering a pathway to cost-effective and clinically viable models for neurodegenerative disease management. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

USING BLACK-SUN EFFECT TO IMPROVE UNSCENTED KALMAN FILTERING OF TRISAT-R SATELLITE ATTITUDE PREDICITON

USING BLACK-SUN EFFECT TO IMPROVE UNSCENTED KALMAN FILTERING...

引用

2024 Small Satellites Systems and Services Symposium, 4S 2024

作者： Kosanič, Nejc Kramberger, Iztok University of Maribor Faculty of Electrical Engineering and Computer Science Koroška cesta 46 Maribor2000 Slovenia

ISBN: (纸本)9781510688940

Due to the nature of monetary and spatial constrictions, larger systems on small satellites are getting replaced by smaller but more inaccurate sensors. To improve the satellite orientation estimate, multiple different optical sensors may be used. In the case of TRISAT-R, the satellite is equipped with photodiodes and small cameras (CMOS sensors) in parallel, to establish the position of the Sun and calculate the satellite attitude. Due to the intensity of solar radiation, the "black-Sun" effect can be clearly distinguished: a black circle in a photo of a CMOS sensor. The black circle is smaller than the whole Sun on the photo, which provides a more accurate estimate of Suns position. In this paper we will discuss the use of an Unscented Kalman Filtering, which uses magnetometer, gyro meter and optical sensor data, and then calculates the satellites attitude and error estimation. The accuracy of attitude estimation between photodiodes and CMOS sensor images will be compared and conclusion that CMOS sensors can be used to calibrate photodiodes in orbit will be presented. © 2025 The Authors.

关键词： CMOS integrated circuits

来源：评论

学校读者我要写书评

暂无评论

Comparative analysis of carbon dioxide and hydrogen plume migration in aquifers inspired by the FluidFlower benchmark study

引用

International Journal of Hydrogen Energy 2025年 135卷 56-68页

作者： Misaghi Bonabi, Amin van Rooijen, Willemijn Al Kobaisi, Mohammed Vuik, Cornelis Hajibeygi, Hadi Department of Applied Mathematics Faculty of Electrical Engineering Mathematics and Computer Science Delft University of Technology Delft Netherlands Department of Geoscience and Engineering Faculty of Civil Engineering and Geosciences Delft University of Technology Delft Netherlands

Large-scale geological storages of hydrogen (H2) and carbon dioxide (CO2) in saline aquifers present feasible options for a sustainable energy future. We compared the plume migration of CO2 and H2 in aquifers using the FluidFlower benchmark, incorporating the state-of-the-art thermophysical and petrophysical properties. The H2 plume, with its higher buoyancy and mobility compared to CO2, remains predominantly in the gas phase due to its lower solubility, increasing the chances of escaping through fractures or migration to distant regions. This additionally leads to a higher pressurized reservoir, which, along with higher buoyancy, increases the chance of caprock penetration. Dissolution trapping of CO2 into brine increases over time due to its fingering, while H2 does not show fingering. Our findings show that while geological carbon storage (GCS) benefits significantly from all structural, dissolution, and residual trapping, underground hydrogen storage (UHS) relies mainly on structural trapping, making the integrity of sealing elements of the system a key factor in its performance. © 2025 The Authors

关键词： Two phase flow

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Lung Cancer Histopathology Classification: Emphasizing Accuracy and Customisation

引用

Biomedical Materials and Devices 2025年 1-13页

作者： Kashyap, Seema Shukla, Arvind Kumar Naim, Iram Pal, Saurabh School of Computer Science and Engineering IFTM University Uttar Pradesh Moradabad India Faculty of Engineering and Technology MJP Rohilkhand University Bareilly India Faculty of Engineering and Technology VBS Purvanchal University Jaunpur India

Lung cancer is a dangerous disease that can be fatal, and a correct diagnosis is essential for figuring out the best way to treat it. The optimum treatment for people with lung cancer requires the classification of the disease into its histological types, such as adenocarcinoma (ADC), small cell lung cancer (SCLC), and squamous cell carcinoma (SCC). Each histological subtype has its features and may react differently to different types of medicine. So, knowing the exact subtype helps guide treatment choices and improve the patient's outcome. Lung cancer subtypes are necessary for personalized treatment. It helps doctors choose tumor-specific treatments such as surgery, radiation, chemotherapy, targeted drugs, and immunotherapies. Precise categorization improves prognosis, avoids needless medicines, and lets patients participate in clinical studies targeting their cancer subtype. Precision medicine improves lung cancer outcomes with accurate categorization. The current algorithms in this domain have shown deficiencies in performance criteria such as specificity, F-score, sensitivity, and precision in recognition. These limitations may stem from challenges such as the complexity and heterogeneity of histopathological images, variations in staining techniques, and the presence of confounding factors. Deep learning methods have made it easier to look at histopathology slides of cancer and see what's going on. Several studies have shown that convolutional neural networks (CNN) are essential for classifying histopathological pictures of different kinds of cancer, like brain, skin, breast, lung, and colon cancer. This study divides lung cancer images into three groups: normal, adenocarcinoma, and squamous cell carcinoma. We have been training deep learning algorithms to identify lung cancer in histopathology slides better, and utilizing deep learning strategies and cutting-edge algorithms such as VGG-19, ResNet-50 v2, EfficientNetB1, and others indicates a comprehensive ap

关键词： And images from histopathology Classification Deep learning EfficientNetB1 Lung cancer ResNet-50 v2 VGG-19

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing FANETs with Reinforcement Learning: Optimized Data Forwarding and Real-Time Adaptability

IEEE Open Journal of the Communications Society

引用

IEEE Open Journal of the Communications Society 2025年 6卷 4295-4310页

作者： Mohammed, Yasir I. Hassan, Rosilah Hasan, Mohammad Kamrul Islam, Shayla Abbas, Huda Saleh Khan, Muhammad Asghar Khan, Muhammad Attique Center for Cyber Security Faculty of Information Science and Technology the National University of Malaysia Selangor Bangi43600 Malaysia Institute of Computer Science and Digital Innovation UCSI University Kuala Lumpur56000 Malaysia Department of Computer Science Computer Science and Engineering College Taibah University Madinah Saudi Arabia Department of EE College of Engineering Prince Mohammad Bin Fahd University Al-Khobar Saudi Arabia Department of AI College of Computer Engineering and Science Prince Mohammad Bin Fahd University Al-Khobar Saudi Arabia

Unmanned Aerial Vehicles (UAVs), commonly known as drones, have significantly advanced wireless communication frameworks by enabling the formation of Flying Ad-Hoc Networks (FANETs). FANETs facilitate autonomous collaboration among UAVs through decentralized and self-organizing communication protocols, proving especially effective in dynamic applications such as military surveillance, disaster management, and environmental monitoring. Nevertheless, traditional routing algorithms, initially developed for terrestrial networks, often fail to meet the unique challenges of FANETs, notably their high mobility and frequently changing network topologies. A framework was proposed to address these challenges;this paper formulates a multi-objective optimization problem aimed at optimizing UAV trajectories, enhancing energy efficiency, and maximizing communication range to improve overall data forwarding performance. A Reinforcement Learning (RL)-based agent is created that constantly enhances its decision-making capacity by utilizing real-time feedback and dynamically chooses best forwarding tactics. This work also combines developments in large-scale data collecting from Wireless Sensor Networks (WSNs), using mobile sinks supported by FANETs in conjunction with multi-objective optimization approaches to improve data collecting efficiency greatly. Experimental tests show that the suggested RL-based techniques outperform conventional routing protocols by properly lowering delays and raising the Packet Delivery Ratio (PDR). Moreover, simulation findings show the better scalability and adaptability of RL-enabled UAV networks, stressing its possible use in dynamic real-world situations such as disaster relief operations and environmental monitoring tasks. © 2020 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Comprehensive multimodal approach for Parkinson’s disease classification using artificial intelligence: insights and model explainability

引用

Soft Computing 2025年第3期29卷 1845-1877页

作者： Balaha, Hossam Magdy Hassan, Asmaa El-Sayed Ahmed, Rawan Ayman Balaha, Magdy Hassan Bioengineering Department J.B. Speed School of Engineering University of Louisville LouisvilleKY United States Computer Science and Control Systems Engineering Department Faculty of Engineering Mansoura University Mansoura Egypt Mathematics and Engineering Physics Department Faculty of Engineering Mansoura University Mansoura Egypt Department of Obstetrics and Gynecology Faculty of Medicine Tanta University Tanta Egypt

Parkinson’s disease (PD) is a debilitating neurodegenerative disorder affecting millions worldwide. Early detection is vital for effective management, yet remains challenging. In this study, we investigated four distinct datasets for PD detection. Through comprehensive experimentation employing ensemble methods and feature selection, we achieved high classification accuracies across the datasets. For the Oxford Parkinson’s Disease Detection Dataset, an accuracy of 95.67%, precision of 97.59%, recall of 84.5%, specificity of 99.32%, and F1-score of 90.57% were achieved. For the Alzheimer Parkinson Diseases 3 Class Dataset, the "Stacking" approach surpasses individual models, reaching an accuracy of 99.85%, precision of 99.81%, recall of 99.81%, specificity of 99.86%, and F1 of 99.81%. For the NewHandPD dataset, Regarding the Spiral category, The "Base-P32-384" model surpasses others with an accuracy of 97.35%, precision of 96.50%, recall of 98.57%, and F1-score of 97.53%. The collective "Stacking" approach proves highly effective regarding the Circle category, achieving 100% across all performance metrics. Regarding the Meander category, the "Base-P16-224" model achieves an accuracy of 97.35%, precision of 99.26%, recall of 95.71%, specificity of 99.19%, and F1 of 97.45%. The Mobile Device Voice Recordings at King’s College London (MDVR-KCL) dataset contains two datasets. Regarding the "SpontaneousDialogue" dataset, accuracy, BAC, precision, recall, specificity, and F1-score were computed, resulting in values of 94.03%, 92.83%, 90.78%, 100.0%, and 85.67%, respectively. Regarding the "ReadText" dataset, accuracy, BAC, precision, recall, specificity, and F1-score were computed, resulting in values of 91.89%, 90.62%, 87.5%, 100.0%, and 81.25%, respectively. Our findings highlight the efficacy of leveraging diverse data sources and advanced machine learning techniques to enhance PD detection accuracy. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germa

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

A Prediction Model for Tumbler Strength Based on Deep Temporal Clustering Across Multiple Operating Conditions 26

A Prediction Model for Tumbler Strength Based on Deep Tempor...

引用

26th International Conference on Industrial Technology, ICIT 2025

作者： Hu, Yifeng Chen, Ao Chen, Xiaoxia Wang, Pengqi Cui, Bingbing Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo China

ISBN: (纸本)9798331521950

Iron ore sintering is a critical step in the blast furnace ironmaking process and tumbler strength is a key physical parameter for evaluating sintered ore quality. Accurate prediction of tumbler strength aids in adjusting process parameters to ensure high strength and stability, fulfilling the blast furnace's needs. This paper proposes a tumbler strength prediction model based on deep temporal clustering in multiple work modes. First, work modes are identified using feature space mapping and deep clustering techniques, addressing the limitations of traditional clustering methods in condition identification. Next, a spatio-temporal modeling approach is applied to independently model each work mode, considering the spatio-temporal characteristics of the sintering process. To mitigate the impact of sample size imbalances across work modes, a weighted strategy is introduced, assigning different weights to data from each condition. Experimental results show that the proposed method achieves high prediction accuracy for tumbler strength. © 2025 IEEE.

关键词： Ore sintering

来源：评论

学校读者我要写书评

暂无评论

Enhancing Privacy in IoT Networks: A Comparative Analysis of Classification and Defense Methods

引用

IEEE Access 2025年 13卷 71611-71646页

作者： Emre Ergun, Ahmet Can, Ozgu Kantarcioglu, Murat Ege University Graduate School of Natural and Applied Sciences Bornova35100 Turkey Izmir Katip Celebi University Faculty of Engineering and Architecture Department of Computer Engineering Cigli35620 Turkey Ege University Department of Computer Engineering Bornova35100 Turkey Virginia Tech Department of Computer Science BlacksburgVA24061 United States

The rapid proliferation of Internet of Things (IoT) devices has led to a substantial increase in network packet traffic, raising significant privacy concerns. Although traffic encryption is employed to protect the privacy of IoT devices, attackers can still leverage Machine Learning (ML) and Deep Learning (DL) techniques to classify device types by analyzing packet characteristics, such as size and timing. The main challenges in the state of the art are the lack of effective methods for exposing privacy violations in encrypted IoT traffic, and the absence of robust defense mechanisms to mitigate privacy breaches caused by network traffic analysis. Considering these challenges, this study presents two key contributions: (i) a novel vector-based classification method that enhances device-type identification from encrypted IoT traffic using advanced ML and DL techniques, and (ii) a robust defense mechanism based on Differential Privacy (DP) and advanced padding techniques against traffic analysis attacks. Therefore, the study examines privacy risks associated with sequential IoT device data and evaluates the effectiveness of ML algorithms using two datasets. The results demonstrate that the proposed vector-based classification method significantly improves the attacker’s classification accuracy, even when privacy-preserving techniques, such as padding, are used to obscure device-type classification. For this purpose, the study evaluates eXtreme Gradient Boosting (XGBoost), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) for IoT traffic classification, achieving an accuracy rate of 99.61% with XGBoost, 96.74% with LSTM, and 96.94% with GRU. Additionally, the Decision Tree (DT), Random Forest (RF), k-Nearest Neighbors (kNN), and GRU classification algorithms are also evaluated and compared with the XGBoost and LSTM classifiers for the proposed attack model. As a defense mechanism, DP is applied using the Fourier Perturbation Algorithm (FPA) to optimize padd

关键词： Decision trees

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：