Article
Computer Science, Artificial Intelligence
Zhimeng Han, Muwei Jian, Gai-Ge Wang
Summary: This paper proposes an efficient model called ConvUNeXt, based on the classical UNet, for medical image segmentation with a low number of parameters. The model incorporates improvements such as larger convolution kernels, depth-wise separable convolution, residual connections, and a lightweight attention mechanism to enhance segmentation performance. Experimental results demonstrate superior performance compared to the standard UNet, particularly with limited data.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Pradip Paithane, Sangeeta Kakarwal
Summary: The LMNS-net deep learning model is a fast and accurate approach for automatic pancreatic segmentation in clinical abdominal CT images. It utilizes a lightweight multiscale module to reduce computation time and achieve high accuracy. The model takes only 1-3 seconds for segmentation in the testing process, making it faster and more efficient than other approaches.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Shukai Yang, Xiaoqian Zhang, Yufeng Chen, Youtao Jiang, Quan Feng, Lei Pu, Feng Sun
Summary: In recent years, there has been much attention on precise medical image segmentation methods based on the encoder-decoder structure. However, there are still limitations, such as increasingly complex network structures and insufficient multiscale information fusion ability. To address these issues, a novel lightweight precise medical image segmentation network called UcUNet was designed, which has a large receptive field and multiscale information fusion ability with a low parameter count.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Feiyue Wang, Shaopeng Hu, Kohei Shimasaki, Idaku Ishii
Summary: In this study, a software-based fingertip velocimeter using high-frame-rate video processing is proposed to estimate when and where an operator taps with his/her finger. The system combines digital image correlation (DIC) with convolution neural network (CNN)-based object detection to estimate fingertip velocities in real time. Experimental results demonstrate the effectiveness of the fingertip velocimeter as a finger tapping interface for multiple fingers.
IEEE SENSORS JOURNAL
(2023)
Article
Optics
Yin Wang, Jiaqing Zhao
Summary: Digital image correlation (DIC) is a non-contact optical method that tracks the speckle pattern on the specimen surface to calculate displacement and strain. Traditional DIC methods have limitations, but deep learning-based DIC methods (Deep-DIC) show promising performance. In this paper, a new Hermite dataset and a new network architecture designed for DIC tasks are proposed, and their superiority over other Deep-DIC methods is demonstrated.
OPTICS AND LASERS IN ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Zhijie Wang, Ran Song, Peng Duan, Xiaolei Li
Summary: Semantic segmentation, as a challenging task in computer vision, has been effectively improved by the design of an enhancement-fusion network (EFNet). EFNet enhances input images to provide diversified features for pixel-wise labeling. Experimental results demonstrate that the combination of EFNet and CNN-based segmentation networks significantly enhances segmentation performance.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Erjing Zhou, Xiang Xu, Baomin Xu, Hongwei Wu
Summary: This paper proposes a novel semantic segmentation model called Ince-DResAsppNet based on dense convoluted separation convolution. The model aims to reduce semantic information loss and enhance detailed information in order to improve pixel-level semantic understanding. Experimental results demonstrate that our model outperforms existing semantic segmentation models in terms of segmentation accuracy on the PASCAL VOC 2012 and CityScapes datasets.
APPLIED INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Wenfeng Luo, Meng Yang, Weishi Zheng
Summary: The LSISU model utilizes saliency and incremental supervision updating to dynamically update segmentation supervision, complementing classification loss with an image saliency objective. By generating high-quality initial mask estimation through class-wise pooling strategy, LSISU effectively deals with object co-occurrence problem and achieves superior segmentation performance on benchmark datasets.
PATTERN RECOGNITION
(2021)
Article
Biology
Jingyuan Li, Wenfang Sun, Karen M. von Deneen, Xiao Fan, Gang An, Guangbin Cui, Yi Zhang
Summary: In this study, we propose a deep learning network with enhanced global-awareness for automatic thymoma segmentation in preoperative contrast-enhanced computed tomography (CECT) images. The network enhances the global-awareness of convolutional neural networks (CNNs) through multi-level feature interaction and integration. Evaluation results show that the network has superior segmentation performance and generalization ability compared to other state-of-the-art models.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Geochemistry & Geophysics
Chengxin Liu, Shuaiyuan Du, Hao Lu, Dehui Li, Zhiguo Cao
Summary: This letter introduces a deep encoder-decoder network for semantic land cover segmentation in aerial imagery, addressing the challenges specific to this domain. Experimental results show that the proposed method achieves compelling performance.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Engineering, Multidisciplinary
A. Brodecki, M. Kopec, Z. L. Kowalewski
Summary: This research compares the long-time degradation effect on two different states of 10CrMo9-10 (10H2M) power engineering steel using various experimental and analytical approaches. Fatigue loading was applied to specimens machined from the as-received steel and the same material after 280,000 hours of exploitation at 540 degrees C and 2.9 MPa internal pressure, while monitoring with Digital Image Correlation (DIC) technique. The impact of long-time degradation on the mechanical response of 10H2M steel was studied through fractographic observations, characterized by the fatigue damage measure, phi, and the fatigue damage parameter D.
INTERNATIONAL JOURNAL OF PRESSURE VESSELS AND PIPING
(2023)
Article
Mechanics
Ryan Spencer, Ahmed Arabi Hassen, Justin Baba, John Lindahl, Lonnie Love, Vlastimil Kunc, Suresh Babu, Uday Vaidya
Summary: As additive manufacturing (AM) advances, there is a growing need for in-situ monitoring of thermal residual stress. This study demonstrates the use of a novel digital image correlation (DIC) adaptation to effectively monitor the impact of thermal residual stress on large-scale AM, particularly in evaluating warpage of printed components.
COMPOSITE STRUCTURES
(2021)
Article
Engineering, Civil
Mengxu Lu, Zhenxue Chen, Q. M. Jonathan Wu, Nannan Wang, Xuewen Rong, Xinghe Yan
Summary: This paper proposes a real-time network for semantic segmentation called FRNet, which achieves a trade-off between accuracy and inference speed by employing Factorized and Regular (FR) blocks and an asymmetric encoder-decoder architecture. Experimental results on multiple datasets demonstrate that our network outperforms other state-of-the-art networks.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Review
Optics
Ghulam Mubashar Hassan
Summary: Digital Image Correlation (DIC) has been successful in measuring deformation remotely, but has limitations in measuring discontinuous deformation, restricting its application to lab environment. This study systematically analyzes different methodologies to overcome the limitations of DIC and presents future research directions in this area.
OPTICS AND LASERS IN ENGINEERING
(2021)
Article
Geochemistry & Geophysics
Li Chen, Xin Dou, Jian Peng, Wenbo Li, Bingyu Sun, Haifeng Li
Summary: The paper introduces an end-to-end ensemble fully convolutional network (EFCNet) with adaptive fusion module (AFM) and separable convolutional module (SCM), which can effectively enhance semantic segmentation performance of high-resolution remote sensing images and reduce model complexity.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Engineering, Multidisciplinary
Mahmoud Alfouneh, Van-Nam Hoang, Zhen Luo, Quantian Luo
Summary: This article investigates the topology optimization of multi-layer multi-material composite structures under static loading. The moving iso-surface threshold optimization method, previously defined for single or cellular materials, is extended to multi-layer multi-material structures using a physical response function discrepancy scheme. It is also integrated with an alternating active-phase algorithm as an alternative procedure. The proposed methods are applied to three types of objective functions, namely, minimizing compliance, maximizing mutual strain energy, and minimizing full-stress designs. Examples are presented and compared with existing literature to verify the derived formulations for topology optimization of multi-layer multi-material structures.
ENGINEERING OPTIMIZATION
(2023)
Article
Acoustics
Feng Zhao, Shuqian Cao, Quantian Luo, Jinchen Ji
Summary: This paper presents an improved design of the quasi-zero stiffness isolator with three pairs of oblique springs to increase the amplitude of the excitation. Theoretical formulations are derived for stiffness and force, and the influences of three independent parameters on the quasi-zero stiffness region are studied to obtain optimal design parameters. The experimental results show that the enhanced design of the quasi-zero stiffness isolator with three pairs of oblique springs can achieve lower displacement transmissibility and deal with the displacement excitation with higher amplitude.
JOURNAL OF VIBRATION AND CONTROL
(2023)
Article
Geriatrics & Gerontology
Na Li, Runan Luo, Wenlong Zhang, Yu Wu, Chaojie Hu, Manli Liu, Diya Jiang, Ziran Jiang, Xinxin Zhao, Yiping Wang, Qing Li
Summary: The study reveals that IL-17A can promote endothelial cell aging by activating the JNK signaling pathway and upregulating FTO expression. This discovery is significant for the identification of new therapeutic targets against endothelial cell aging and related vascular complications.
Article
Mechanics
Wen Zuo, Quantian Luo, Qing Li, Guangyong Sun
Summary: Thin-walled structures made of fiber reinforced composites are commonly used in engineering practice, but there is limited research on their residual properties after high temperature and hygrothermal aging. This experimental investigation aims to study the effects of moisture absorption and high temperatures on the mechanical characteristics of fiber reinforced plastic composite tubes. The study found that crashworthiness characteristics decrease significantly with increased temperature and moisture absorption rate. The failure modes varied and were influenced by the glass transition temperature of the matrix. Moisture absorption had two stages and was affected by temperature. Microscopically, the morphology and bonding conditions between fiber and resin changed significantly due to temperature and hydrothermal aging.
COMPOSITE STRUCTURES
(2023)
Article
Mechanics
Erdong Wang, Chao Chen, Guangzhou Zhang, Quantian Luo, Qing Li, Guangyong Sun
Summary: Open-cell Kelvin lattice structures (Kelvin foams) are fabricated through the SLM process and the multiaxial mechanical behaviors of these foams are studied. It is found that the yield surface of the Kelvin foams gradually shrinks with increasing dimensional tolerance induced by the SLM process, especially under hydrostatic compression. The influence of foam filler on the yield surface is weakened when experiencing hydrostatic compression.
COMPOSITE STRUCTURES
(2023)
Article
Engineering, Biomedical
Ali Entezari, Nai-Chun Liu, Zhongpu Zhang, Jianguang Fang, Chi Wu, Boyang Wan, Michael Swain, Qing Li
Summary: Despite advances in bone scaffold design optimization, their functionality remains suboptimal due to uncertainties caused by the manufacturing process. A novel multi-objective robust optimization approach is proposed to minimize the effects of uncertainties on the optimized design. This study presents the first non-deterministic optimization of tissue scaffold, shedding light on the significant topic of scaffold design and additive manufacturing.
JOURNAL OF THE MECHANICAL BEHAVIOR OF BIOMEDICAL MATERIALS
(2023)
Review
Engineering, Mechanical
Yaozhong Wu, Jianguang Fang, Chi Wu, Cunyi Li, Guangyong Sun, Qing Li
Summary: Lightweight materials and structures have been extensively studied for design and manufacturing of more sustainable products with reduced materials and energy consumption, while maintaining proper mechanical and energy absorption characteristics. Additive manufacturing techniques have offered more freedom for designing novel lightweight materials and structures, but the rational design for desired mechanical properties remains challenging. This review comprehensively discusses the recent advances in additively manufactured materials and structures, focusing on their mechanical properties and energy absorption applications. It also addresses challenges, future directions, and optimization techniques in this field.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2023)
Article
Mechanics
Yu Lu, Qiang Liu, Zengbo Zhang, Liuye Qin, Qing Li
Summary: This study aimed to investigate the responses of riveted-bonded hybrid joints connecting CFRP and Al under tensile and cross tensile loads. Different locking modes were fabricated and analyzed. The mechanical properties and failure mechanisms were studied and compared. Numerical models were established to replicate the failure behaviors and identify damaged areas.
COMPOSITE STRUCTURES
(2023)
Article
Engineering, Multidisciplinary
Dapeng Wang, Dequan Zhang, Yuan Meng, Meide Yang, Chuizhou Meng, Xu Han, Qing Li
Summary: With the increasing complexity of engineering problems, traditional reliability analysis methods face challenges in terms of computational efficiency and accuracy. The Kriging model, a surrogate model, has been widely used in reliability analysis due to its advantages in computational efficiency and numerical accuracy. However, there are still significant issues with the Kriging model-assisted reliability analysis, such as the need for a large candidate sample pool and excessive local prediction accuracy. To address these issues, a new method called AK-HRn, which combines adaptive Kriging and n-hypersphere rings, is proposed in this study. The AK-HRn method demonstrates high efficiency and robustness in solving complex reliability analysis problems.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2023)
Article
Engineering, Mechanical
Cunyi Li, Jianguang Fang, Yuheng Wan, Na Qiu, Grant Steven, Qing Li
Summary: This study aims to develop a phase field framework for simulating the complex mechanical behaviors of laser powder bed fusion printed metallic materials. By considering the microstructural orientation induced by laser powder bed fusion, transversely isotropic Hill48 and modified Mohr-Coulomb constitutive models are incorporated to describe plasticity and fracture behaviors respectively. The proposed phase field model is able to better reproduce force-displacement responses of all specimens by considering the stress state-dependent crack initiation. Moreover, applying a transversely isotropic fracture model is necessary to accurately predict the crack path and global force-displacement responses.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2023)
Review
Engineering, Civil
Ruyang Yao, Tong Pang, Bei Zhang, Jianguang Fang, Qing Li, Guangyong Sun
Summary: This article provides a comprehensive overview of recent advances in the development of thin-walled multi-cell structures and materials (TWMCSM) for crashworthiness and protection applications in various vehicles. It covers the classification of TWMCSM, commonly-used manufacturing methods, energy absorption mechanism and characteristics, experimental testing and numerical modeling techniques, key parameters affecting crashworthiness, analytical modeling methods, design optimization procedures, typical applications and future research directions. It aims to provide informative references and a comprehensive landscape for researchers and engineers in designing new TWMCSM for better energy absorption and crashworthiness.
THIN-WALLED STRUCTURES
(2023)
Article
Engineering, Civil
Yang Wei, Quantian Luo, Qing Li, Guangyong Sun
Summary: In this study, the mixed mode failure behaviors of adhesive bonded joints were investigated using the modified Arcan fixture. The interfacial displacement field was obtained using digital image correlation technique, and the mode mixture was decoupled using the virtual crack closure technique. With the help of these techniques, a refined cohesive zone model was established and validated through experiments and finite element analysis.
THIN-WALLED STRUCTURES
(2023)
Article
Dentistry, Oral Surgery & Medicine
Pongsakorn Poovarodom, Chaiy Rungsiyakull, Jarupol Suriyawanakul, Qing Li, Keiichi Sasaki, Nobuhiro Yoda, Pimduen Rungsiyakull
Summary: This study aimed to evaluate the influence of subcrestal implant placement depth on bone remodeling using time-dependent finite element analysis (FEA) with a bone-remodeling algorithm. The study found that deeper implant placement can increase bone density, but it also increases the maximum von Mises stress and overloading elements.
JOURNAL OF PROSTHODONTIC RESEARCH
(2023)
Article
Engineering, Mechanical
Rosaria Del Toro, Maria Laura De Bellis, Marcello Vasta, Andrea Bacigalupo
Summary: This article presents a multifield asymptotic homogenization scheme for analyzing Bloch wave propagation in non-standard thermoelastic periodic materials. The proposed method derives microscale field equations, solves recursive differential problems within the unit cell, establishes a down-scaling relation, and obtains average field equations. The effectiveness of this approach is validated by comparing dispersion curves with those from the Floquet-Bloch theory.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)
Article
Engineering, Mechanical
Yue Bao, Zhengcheng Yao, Yue Zhang, Xueman Hu, Xiandong Liu, Yingchun Shan, Tian He
Summary: This paper proposes a novel triple-gradient phononic acoustic black hole (ABH) beam that strategically manipulates multiple gradients to enhance its performance. The study reveals that the ABH effect is not solely brought about by the thickness gradient, but also extends to the power-law gradients in density and modulus. The synergistic development of three different gradient effects leads to more pronounced and broader bandgaps in PCs.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)
Article
Engineering, Mechanical
Matthias Ryser, Jason Steffen, Bekim Berisha, Markus Bambach
Summary: This study investigates the feasibility of replacing complex experiments with multiple simpler ones to determine the anisotropic yielding behavior of sheet metal. The results show that parameter identifiability and accuracy can be achieved by combining multiple specimen geometries and orientations, enhancing the understanding of the yield behavior.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)
Article
Engineering, Mechanical
Wenjun Li, Pengfei Zhang, Siyong Yang, Shenling Cai, Kai Feng
Summary: This study presents a novel two-dimensional non-contact platform based on Near-field Acoustic Levitation (NFAL), which can realize both one-dimensional and two-dimensional transportation. Numerical and experimental results prove the feasibility and ease of this method.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)
Article
Engineering, Mechanical
Shuo Liu, Lu Che, Guodong Fang, Jun Liang
Summary: This study presents a novel lamina conjugated bond-based peridynamic (BB-PD) model that overcomes the limitations of material properties and is applicable to composite laminates with different stacking sequences. The accuracy and applicability of the model are validated through simulations of elastic deformation and progressive damage behavior, providing an explanation of the damage modes and failure mechanisms of laminated composite materials subjected to uniaxial loading.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)
Article
Engineering, Mechanical
Omar El-Khatib, S. Kumar, Wesley J. Cantwell, Andreas Schiffer
Summary: Sandwich-structured honeycombs (SSHCs) are hierarchical structures with enhanced mass-specific properties. A model capable of predicting the elastic properties of hexagonal SSHCs is presented, showing superior in-plane elastic and shear moduli compared to traditional honeycombs, while the out-of-plane shear moduli are reduced.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)
Article
Engineering, Mechanical
Zhi-Jian Li, Hong-Liang Dai, Yuan Yao, Jing-Ling Liu
Summary: This paper proposes a process-performance prediction model for estimating the yield strength and ultimate tensile strength of metallic parts fabricated by powder bed fusion additive manufacturing. The effect of main process variables on the mechanical performance of printed metallic parts is analyzed and the results can serve as a guideline for improvement. The accuracy of the proposed model is validated by comparison with literature.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)
Article
Engineering, Mechanical
Saman A. Bapir, Kawa M. A. Manmi, Rostam K. Saeed, Abdolrahman Dadvand
Summary: This study numerically investigates the behavior of an ultrasonically driven gas bubble between two parallel rigid circular walls with a cylindrical micro-indentation in one wall. The primary objective is to determine the conditions that facilitate the removal of particulate contamination from the indentation using the bubble jet. The study found that the bubble jet can effectively remove contamination from the indentation for certain ranges of indentation diameter, but becomes less effective for larger indentation diameters.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)
Article
Engineering, Mechanical
E. Polyzos, E. Vereroudakis, S. Malefaki, D. Vlassopoulos, D. Van Hemelrijck, L. Pyl
Summary: This research investigates the elastic and damage characteristics of individual composite beads used in 3D printed composites. A new analytical probabilistic progressive damage model (PPDM) is introduced to capture the elastic and damage attributes of these beads. Experimental results show strong agreement with the model in terms of elastic behavior and ultimate strength and strain.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2024)