检索结果-内蒙古大学图书馆

Delay-Minimization and Back-Off Aware Q-Learning With Advanced Bio-Inspired CH Selection for Multi-Hop Communication in Vehicular Ad-Hoc Networks

学校读者我要写书评

暂无评论

Radio Science 2025年第6期60卷

作者： Rashid, Sami Abduljabbar Audah, Lukman Hamdi, Mustafa Maad Abood, Mohammed Salah Abbas, Ghassan Raad Mohammed, Bassim Sayed Elwi, Taha A. Khan, Salahuddin Virdee, Bal S. Krasniqi, Astrit Kouhalvandi, Lida Alibakhshikenari, Mohammad Biomedical Engineering Research Center University of Anbar Ramadi Iraq Faculty of Electrical and Electronic Engineering Universiti Tun Hussein Onn Malaysia Batu Pahat Malaysia Department of Computer Science College of Computer Science and IT University of Anbar Ramadi Iraq School of Information and Electronics Beijing Institute of Technology Beijing China Institute of Mathematics and Information Technologies Volgograd State University Volgograd Russia Technical College of Engineering Al-Bayan University Baghdad Iraq Department of Automation and Artificial Intelligence Engineering College of Information Engineering at Al-Nahrain University Baghdad Iraq College of Engineering King Saud University Riyadh Saudi Arabia Center for Communications Technology London Metropolitan University London United Kingdom Department of Electrical and Electronics Engineering Dogus University Istanbul Turkey Electronics Engineering Department University of Rome "Tor Vergata" Rome Italy

The increasing significance of Vehicular Ad-hoc Networks (VANETs) in intelligent transportation systems has introduced challenges related to high mobility, network congestion, and energy efficiency. To address these challenges, this paper proposes a new approach based on Delay-Minimization and Back-Off Aware Q-Learning with Advanced Bio-Inspired Cluster Head (CH) Selection (DBACH) to enhance multi-hop data transmission in VANETs. The DBACH framework features network formation, latency minimization, a back-off Q-learning model, and an improved dragonfly algorithm-based CH selection. This method reduces transmission delay, routing overhead, and power consumption to enhance VANET QoS. DBACH was evaluated against RCDC, DCPA, and WCAM for effectiveness. The simulated vehicle numbers and speeds (km/h) were used to assess energy efficiency, throughput, packet delivery ratio, data loss ratio, computation time, and routing overhead. The DBACH boosts energy efficiency to 85 J, throughput to 160–200 Kbps, and packet delivery ratio to 11%–13%. Data loss drops to 7%–15%, latency is 60–94 ms, and routing overhead is 170–300 packets. When DBACH is a promising option for enhancing VANET communication dependability and energy economy due to its efficiency, communication stability, and success rates. © 2025. American Geophysical Union. All Rights Reserved.

关键词： Backpropagation

Robotic knee tracking control to mimic the intact human knee profile based on actor-critic reinforcement learning

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Huang, He Wu, Ruofan Yao, Zhikai Si, Jennie School of Electrical Computer and Energy Engineering Arizona State University TempeAZ85287 United States Department of Biomedical Engineering North Carolina State University RaleighNC27695 United States University of North Carolina at Chapel Hill Chapel HillNC27599 United States

We address a state-of-the-art reinforcement learning (RL) control approach to automatically configure robotic prosthesis impedance parameters to enable end-to-end, continuous locomotion intended for transfemoral amputee subjects. Specifically, our actor-critic based RL provides tracking control of a robotic knee prosthesis to mimic the intact knee profile. This is a significant advance from our previous RL based automatic tuning of prosthesis control parameters which have centered on regulation control with a designer prescribed robotic knee profile as the target. In addition to presenting the complete tracking control algorithm based on direct heuristic dynamic programming (dHDP), we provide an analytical framework for the tracking controller with constrained inputs. We show that our proposed tracking control possesses several important properties, such as weight convergence of the learning networks, Bellman (sub)optimality of the cost-to-go value function and control input, and practical stability of the human-robot system under input constraint. We further provide a systematic simulation of the proposed tracking control using a realistic human-robot system simulator, the OpenSim, to emulate how the dHDP enables level ground walking, walking on different terrains and at different paces. These results show that our proposed dHDP based tracking control is not only theoretically suitable, but also practically useful. Copyright © 2021, The Authors. All rights reserved.

关键词： Dynamic programming

Classification of skin-cancer lesions based on Fluorescence Lifetime Imaging

学校读者我要写书评

暂无评论

Classification of skin-cancer lesions based on Fluorescence ...

Medical Imaging 2020: biomedical Applications in Molecular, Structural, and Functional Imaging

作者： Vasanthakumari, Priyanka Romano, Renan A. Rosa, Ramon G.T. Salvio, Ana G. Yakovlev, Vladislav Kurachi, Cristina Jo, Javier A. Department of Biomedical Engineering Texas A&M University College StationTX United States São Carlos Institute of Physics University of São Paulo São Paulo Brazil School of Electrical and Computer Engineering University of Oklahoma NormanOK United States Skin Department of Amaral Carvalho Hospital São Paulo Jau Brazil

ISBN: (纸本)9781510634015

Every year more than 5.4 million new cases of skin cancer are reported in the US. Melanoma is the most lethal type with only 5% occurrence rate, but accounts for over 75% of all skin cancer deaths. Non-melanoma skin cancer, especially basal cell carcinoma (BCC) is the most commonly occurring and often curable type that affects more than 3 million people and causes about 2000 deaths in the US annually. The current diagnosis involves visual inspection, followed by biopsy of the lesions. The major drawbacks of this practice include difficulty in border detection causing incomplete treatment and, the inability to distinguish between clinically similar lesions. Melanoma is often mistaken for the benign lesion pigmented seborrheic keratosis (pSK), making it extremely important to differentiate benign and malignant lesions. In this work, a novel feature extraction algorithm based on phasors was performed on the Fluorescence Lifetime Imaging (FLIM) images of the skin to reliably distinguish between benign and malignant lesions. This approach, unlike the standard FLIM data processing method that requires time-deconvolution of the instrument response from the measured time-resolved fluorescence signal, is computationally much simpler and provides a unique set of features for classification. Subsequently, FLIM derived features were selected using a double step cross validation approach that assesses the reliability and the performance of the resultant trained classifier. Promising FLIM-based classification performance was attained for detecting benign from malignant pigmented (sensitivity: ~80%, specificity: 79%, overall accuracy: ~79%) and non-pigmented (sensitivity: ~88%, specificity: 83%, overall accuracy: ~87%) lesions. © 2020 SPIE

关键词： Machine learning

Low cost carbon fiber as potential lightning strike protection for wind turbine blades 35

学校读者我要写书评

暂无评论

Low cost carbon fiber as potential lightning strike protecti...

35th Annual American Society for Composites Technical Conference, ASC 2020

作者： Kumar, Vipin Kore, Surbhi Subash Theodore, Merlin Wang, Yeqing Yousefpour, Kamran Lin, Wenhua Park, Chanyeop Vaidya, Uday Kunc, Vlastimil KnoxvilleTN37932 United States Department of Mechanical Aerospace and Biomedical Engineering University of Tennessee Estabrook Rd KnoxvilleTN37996 United States Carbon Fiber Technology Facility Oak Ridge National Laboratory Oak RidgeTN37830 United States Department of Aerospace Engineering Mississippi State University MS39762 United States Department of Electrical and Computer Engineering Mississippi State University MS39762 United States

ISBN: (纸本)9781605956657

Until recently, glass fiber composites (GFRP) were the preferred choice to prepare wind turbine blades due to their low cost compared to their counterpart carbon fiber composites (CFRP). However, to harvest the maximum wind energy, ever larger wind turbine blades are being manufactured. To support such a large structure carbon fiber composites CFRP have become the integral part of load bearing structures in the blade. In this work, we are proposing to utilize the low cost carbon fiber (LCCF), manufactured at Carbon Fiber Technology Facility (CFTF) of Oak Ridge National Laboratory (ORNL) as not only the cost effective alternative to the currently used carbon fiber (CF) but also as the lightning strike protection of the wind turbine blades. A textile-grade precursor was used to prepare LCCF. Wind turbines often get hit by lightning strikes due to their operating locations. LCCF can provide structural integrity to these gigantic structures and mitigate the effect of lightning strike on them by effectively dissipating the current. Two composite panels made of LCCF were tested against artificial lightning strikes of 100 kA and 200 kA (component A of lightning waveform SAE ARP 5412-B). The results showed 100 % residual mechanical properties of LCCF composite due to their high electrical conductivity both in-plane (38 S/cm) and in through-thickness direction (0.27 S/cm). There was no significant damage (fiber breakage, resin evaporation or delamination) in both the cases. © ASC 2020.

关键词： Wind turbine blades

Knowledge-Guided Reinforcement Learning Control for Robotic Lower Limb Prosthesis

学校读者我要写书评

暂无评论

Knowledge-Guided Reinforcement Learning Control for Robotic ...

IEEE International Conference on Robotics and Automation (ICRA)

作者： Xiang Gao Jennie Si Yue Wen Minhan Li He Helen Huang Department of Electrical Computer and Energy Engineering Arizona State University Tempe AZ USA Department of Biomedical Engineering North Carolina State University Raleigh NC USA

ISBN: (数字)9781728173955

ISBN: (纸本)9781728173962

Robotic prostheses provide new opportunities to better restore lost functions than passive prostheses for trans-femoral amputees. But controlling a prosthesis device automatically for individual users in different task environments is an unsolved problem. Reinforcement learning (RL) is a naturally promising tool. For prosthesis control with a user in the loop, it is desirable that the controlled prosthesis can adapt to different task environments as quickly and smoothly as possible. However, most RL agents learn or relearn from scratch when the environment changes. To address this issue, we propose the knowledge-guided Q-learning (KG-QL) control method as a principled way for the problem. In this report, we collected and used data from two able-bodied (AB) subjects wearing a RL controlled robotic prosthetic limb walking on level ground. Our ultimate goal is to build an efficient RL controller with reduced time and data requirements and transfer knowledge from AB subjects to amputee subjects. Toward this goal, we demonstrate its feasibility by employing OpenSim, a well-established human locomotion simulator. Our results show the OpenSim simulated amputee subject improved control tuning performance over learning from scratch by utilizing knowledge transfer from AB subjects. Also in this paper, we will explore the possibility of information transfer from AB subjects to help tuning for the amputee subjects.

关键词： Prosthetics Task analysis Impedance Knee Legged locomotion Tuning

The FIB-4 scores in the emergency department to predict the outcomes of COVID-19 patients in taiwan

学校读者我要写书评

暂无评论

Heliyon 2024年第4期10卷 e25649页

作者： Liu, Chia-Yu Chou, San-Fang Chiang, Pei-Ying Sun, Jen-Tang Tsai, Kuang-Chau Jaw, Fu-Shan Chang, Chung-Ta Fan, Chieh-Min Wu, Yuan-Hui Lee, Peng-Yu Hsieh, Chia-Ying Chen, Jie-Ming Hsieh, Chien-Chieh Department of Radiology Taipei Veterans General Hospital Taipei City Taiwan Department of Medical Research Far Eastern Memorial Hospital New Taipei City Taiwan Division of Hospital Medicine Department of Internal Medicine Far Eastern Memorial Hospital Taiwan Department of Emergency Medicine Far Eastern Memorial Hospital New Taipei City Taiwan Department of Biomedical Engineering National Taiwan University Taipei City Taiwan Graduate Institute of Medicine Yuan Ze University Taoyuan Taiwan Department of Medical Education Taipei Tzu Chi Hospital Buddhist Tzu Chi Medical Foundation New Taipei City Taiwan Department of Emergency Medicine Ten Chan General Hospital Zhongli Dist Taoyuan City Taiwan International Bachelor Program in Electrical and Communication Engineering Yuan Ze University Taoyuan Taiwan

Objective: We aimed to determine the reliability of using the Fibrosis-4 (FIB-4) index in COVID-19 patients without underlying liver illness. Method: We employed multivariate logistic regression to identify variables that exhibited statistically significant influence on the ultimate outcome. Multilayer perceptron analysis was employed to develop a prediction model for the FIB-4 index concerning ICU admission and intubation rates. However, the scarcity of cases rendered the assessment of the mortality rate unfeasible. We plotted ROC curves to analyze the predictive strength of the FIB-4 index across various age groups. Result: In univariate logistic regression, only the FIB-4 index and respiratory rate demonstrated statistical significance on all poor outcomes. The FIB-4 index for mortality prediction had an ROC and AUC of 0.863 (95% CI: 0.781–0.9444). It demonstrates predictive power across age groups, particularly for age ≥65 (AUC: 0.812, 95% CI: 0.6571–0.9673) and age <65 (AUC: 0.878, 95% CI: 0.8012–0.9558). Its sensitivity for intubation and ICU admission prediction is suboptimal. Conclusion: FIB-4 index had promising power in prediction of mortality rate in all age groups. © 2024 The Authors

关键词： COVID-19 Critical care medicine Respiratory distress

Interpretable Factorization for Neural Network ECG Models

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Snyder, Christopher Vishwanath, Sriram Department of Biomedical Engineering University of Texas at Austin AustinTX United States Department of Electrical and Computer Engineering University of Texas at Austin AustinTX United States

The ability of deep learning (DL) to improve the practice of medicine and its clinical outcomes faces a looming obstacle: model interpretation. Without description of how outputs are generated, a collaborating physician can neither resolve when the model’s conclusions are in conflict with his or her own, nor learn to anticipate model behavior. Current research aims to interpret networks that diagnose ECG recordings, which has great potential impact as recordings become more personalized and widely deployed. A generalizable impact beyond ECGs lies in the ability to provide a rich test-bed for the development of interpretive techniques in medicine. Interpretive techniques for Deep Neural Networks (DNNs), however, tend to be heuristic and observational in nature, lacking the mathematical rigor one might expect in the analysis of math equations. The motivation of this paper is to offer a third option, a scientific approach. We treat the model output itself as a phenomenon to be explained through component parts and equations governing their behavior. We argue that these component parts should also be "black boxes" – additional targets to interpret heuristically with clear functional connection to the original. We show how to rigorously factor a DNN into a hierarchical equation consisting of black box variables. This is not a subdivision into physical parts, like an organism into its cells;it is but one choice of an equation into a collection of abstract functions. Yet, for DNNs trained to identify normal ECG waveforms on PhysioNet 2017 Challenge data, we demonstrate this choice yields interpretable component models identified with visual composite sketches of ECG samples in corresponding input regions. Moreover, the recursion distills this interpretation: additional factorization of component black boxes corresponds to ECG partitions that are more morphologically pure. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks

A Transformer-based Diffusion Probabilistic Model for Heart Rate and Blood Pressure Forecasting in Intensive Care Unit

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Chang, Ping Li, Huayu Quan, Stuart F. Lu, Shuyang Wung, Shu-Fen Roveda, Janet Li, Ao Department of Electrical & Computer Engineering The University of Arizona TucsonAZ United States Division of Sleep and Circadian Disorders Departments of Medicine and Neurology Brigham and Women’s Hospital Harvard Medical School BostonMA United States Asthma and Airway Disease Research Center College of Medicine The University of Arizona TucsonAZ United States Department of Cardiovascular Surgery Zhongshan Hospital Fudan University Shanghai China The Shanghai Institute of Cardiovascular Diseases Shanghai China Bio5 Institute The University of Arizona TucsonAZ United States College of Nursing The University of Arizona TucsonAZ United States Department of Biomedical Engineering The University of Arizona TucsonAZ United States

Background and Objective: Vital sign monitoring in the Intensive Care Unit (ICU) is crucial for enabling prompt interventions for patients. This underscores the need for an accurate predictive system. Therefore, this study proposes a novel deep learning approach for forecasting Heart Rate (HR), Systolic Blood Pressure (SBP), and Diastolic Blood Pressure (DBP) in the ICU. Methods: We extracted 24,886 ICU stays from the MIMIC-III database which contains data from over 46 thousand patients, to train and test the model. The model proposed in this study, Transformer-based Diffusion Probabilistic Model for Sparse Time Series Forecasting (TDSTF), merges Transformer and diffusion models to forecast vital signs. The TDSTF model showed state-of-the-art performance in predicting vital signs in the ICU, outperforming other models’ ability to predict distributions of vital signs and being more computationally efficient. The code is available at https://***/PingChang818/TDSTF. Results: The results of the study showed that TDSTF achieved a Standardized Average Continuous Ranked Probability Score (SACRPS) of 0.4438 and a Mean Squared Error (MSE) of 0.4168, an improvement of 18.9% and 34.3% over the best baseline model, respectively. The inference speed of TDSTF is more than 17 times faster than the best baseline model. Conclusion: TDSTF is an effective and efficient solution for forecasting vital signs in the ICU, and it shows a significant improvement compared to other models in the field. Copyright © 2023, The Authors. All rights reserved.

关键词： Forecasting

Lattice Transformation from 2-D to Quasi 1-D and Phonon Properties of Exfoliated ZrS2 and ZrSe2

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Alsulami, Awsaf Alharbi, Majed Alsaffar, Fadhel Alolaiyan, Olaiyan Aljalham, Ghadeer Albawardi, Shahad Alsaggaf, Sarah Alamri, Faisal Tabbakh, Thamer A. Amer, Moh R. Center of Excellence for Green Nanotechnologies Joint Centers of Excellence Program King Abdulaziz City for Science and Technology Riyadh Saudi Arabia Department of Mechanical and Aerospace Engineering University of California Los Angeles Los AngelesCA90095 United States National Center for Nanotechnology Materials Science Institute King Abdulaziz City for Science and Technology Riyadh Saudi Arabia Department of Electrical and Computer Engineering University of California Los Angeles Los AngelesCA90095 United States

Recent reports on thermal and thermoelectric properties of emerging 2-Dimensional (2D) materials have shown promising results. Yet, most of these emerging materials are underrepresented or lack a consistent and thorough investigation. Among these materials are Zirconium based chalcogenides such as Zirconium Disulfide (ZrS2) and Zirconium Dieselenide (ZrSe2). Here, we investigate the thermal properties of these Zirconium based materials using confocal Raman spectroscopy. We observed 2 different and distinctive Raman signatures for exfoliated ZrX2 (where X = S or Se). These Raman modes generally depend on the shape of the exfoliated nanosheets, regardless of the incident laser polarization. These 2 shapes are divided into 2D- ZrX2 and quasi 1D- ZrX2. For 2D- ZrX2, Raman modes are in alignment with those reported in literature. However, for quasi 1D-ZrX2, we show that Raman modes are identical to exfoliated ZrX3 nanosheets, indicating a major lattice transformation from 2D to quasi-1D. We also measure thermal properties of each resonant Raman mode for each ZrX2 shape. Based on our measurements, most Raman modes exhibit a linear downshift dependence with temperature. However, for ZrS2, we see an upshift (blueshift) with temperature for A1g mode, which is attributed to non-harmonic effects caused by dipolar coupling with IR-active modes. Moreover, the observed temperature dependence coefficient for some phonon modes of quasi 1D-ZrX2 differ dramatically, which can be caused by the quasi 1D lattice. Finally, we measure phonon dynamics under optical heating for each of 2D-ZrX2 and quasi 1D-ZrX2 and show phonon confinement in quasi 1D-ZrX2 nanosheets. We extract the thermal conductivity and the interfacial thermal conductance for each of 2D-ZrX2 and quasi 1D-ZrX2 nanosheets. Our calculations indicate lower interfacial thermal conductance for quasi 1D-ZrX2 compared to 2D-ZrX2, which can be attributed to the phonon confinement in 1D. Based on our model, we show low thermal con

关键词： Thermal conductivity