A5068855808- Additive Manufacturing Materials and Processes
- Advanced machining processes and optimization
- Metallurgy and Material Forming
- Industrial Vision Systems and Defect Detection
- Model Reduction and Neural Networks
- Microstructure and Mechanical Properties of Steels
- Metal Alloys Wear and Properties
- Welding Techniques and Residual Stresses
- Manufacturing Process and Optimization
- Additive Manufacturing and 3D Printing Technologies
- Robotic Locomotion and Control
- Robotics and Automated Systems
- Metal Forming Simulation Techniques
- Topology Optimization in Engineering
- Advanced Surface Polishing Techniques
- Prosthetics and Rehabilitation Robotics
- COVID-19 diagnosis using AI
- Soft Robotics and Applications
- Non-Destructive Testing Techniques
- Robotic Path Planning Algorithms
- Optical measurement and interference techniques
- High Entropy Alloys Studies
- Winter Sports Injuries and Performance
- 3D Printing in Biomedical Research
- Control and Dynamics of Mobile Robots
Tsinghua University
2015-2025
Nankai University
2024
Tianjin People's Hospital
2024
University of Science and Technology of China
2024
State Key Laboratory of Tribology
2016-2023
Chinese PLA General Hospital
2020-2021
Carnegie Mellon University
2019-2020
Georgia Institute of Technology
2017-2019
Case Western Reserve University
2016
The demand for fast and accurate structural analysis is becoming increasingly more prevalent with the advance of generative design topology optimization technologies. As one step toward accelerating analysis, this work explores a deep learning based approach predicting stress fields in 2D linear elastic cantilevered structures subjected to external static loads at its free end using convolutional neural networks (CNN). Two different architectures are implemented that take as input structure...
Abstract In topology optimization using deep learning, the load and boundary conditions represented as vectors or sparse matrices often miss opportunity to encode a rich view of design problem, leading less than ideal generalization results. We propose new data-driven model called TopologyGAN that takes advantage various physical fields computed on original, unoptimized material domain, inputs generator conditional generative adversarial network (cGAN). Compared baseline cGAN, achieves...
Abstract Using deep learning to analyze mechanical stress distributions is gaining interest with the demand for fast analysis. Deep approaches have achieved excellent outcomes when utilized speed up computation and learn physical nature without prior knowledge of underlying equations. However, most studies restrict variation geometry or boundary conditions, making it difficult generalize methods unseen configurations. We propose a conditional generative adversarial network (cGAN) model...
The ready-to-use, structure-supporting hydrogel bioink can shorten the time for ink preparation, ensure cell dispersion, and maintain preset shape/microstructure without additional assistance during printing. Meanwhile, with high permeability might facilitate uniform growth in biological constructs, which is beneficial to homogeneous tissue repair. Unfortunately, current bioinks are hard meet these requirements simultaneously a simple way. Here, based on fast dynamic crosslinking of aldehyde...
Precise quantification of protein-ligand interaction is critical in early-stage drug discovery. Artificial intelligence (AI) has gained massive popularity this area, with deep-learning models used to extract features from ligand and protein molecules. However, these often fail capture intermolecular non-covalent interactions, the primary factor influencing binding, leading lower accuracy interpretability. Moreover, such overlook spatial structure complexes, resulting weaker generalization....
Abstract The Periprosthetic Acetabular Osteotomy (PAO) is a commonly used technique in orthopedics for treating developmental hip dysplasia and dislocation, as the most effective treatment of (DDH). However, performing PAO can be challenging surgeons due to limited visibility difficulty detecting any deformations osteotome chisels when they are deeply immersed pelvis. These challenges result serious complications, such excessive bleeding nerve injuries. We propose novel precision tracking...
Abstract This research presents a method of optimizing the consolidation parts in an assembly using metal additive manufacturing (MAM). The generates candidates for consolidation, filters them feasibility and structural redundancy, finds optimal build layout parts, optimizes which to consolidate genetic algorithm. Results are presented both minimal production time costs, respectively. cost models consider each step process, including MAM build, post-processing steps such as support structure...
Abstract In topology optimization using deep learning, load and boundary conditions represented as vectors or sparse matrices often miss the opportunity to encode a rich view of design problem, leading less than ideal generalization results. We propose new data-driven model called TopologyGAN that takes advantage various physical fields computed on original, unoptimized material domain, inputs generator conditional generative adversarial network (cGAN). Compared baseline cGAN, achieves...
Abstract In design for forming, it is becoming increasingly significant to develop surrogate models of high-fidelity finite element analysis (FEA) simulations forming processes achieve effective component feasibility assessment as well process and optimizations. However, using traditional scalar-based machine learning methods (SBMLMs) fall short on accuracy generalizability. This because SBMLMs fail harness the location information available from simulations. To overcome this shortcoming,...
Both of the long-term fidelity and cell viability three-dimensional (3D)-bioprinted constructs are essential to precise soft tissue repair. However, shrinking/swelling behavior hydrogels brings about inadequate constructs, bioinks containing excessive polymer detrimental viability. Here, we obtained a facile hydrogel by introducing 1% aldehyde hyaluronic acid (AHA) 0.375%
Abstract Proliferation of HPSCs in vitro can promote its broad clinical therapeutic use. For co-culture, interaction between the stem cell and feeder as well their spatial position are essential. To imitate natural microenvironment, a 3D engineered scaffold for CD34 + cells co-culture was established via bioprinting. Herein, concentration hydrogel ratio two kinds were optimized. Flow cytometry, real time PCR RNA-seq technology applied to analyze effect on expanded cells. After 10 days with...
Mesenchymal stem cell (MSC)-derived extracellular vesicles (EVs) are promising candidates for regenerative medicine; however, the lack of scalable methods high quantity EV production limits their application. In addition, signature EV-derived proteins shared in 3D environments and 2D surfaces, remain mostly unknown. Herein, we present a platform combining MSC microfiber culture with ultracentrifugation purification yield. Within this platform, solution (∼3 × 108total cells) is encapsulated...
Brain tumor segmentation using neural networks presents challenges in accurately capturing diverse shapes and sizes while maintaining real-time performance. Additionally, addressing class imbalance is crucial for achieving accurate clinical results. To tackle these issues, this study proposes a novel N-shaped lightweight network that combines multiple feature pyramid paths U-Net architectures. Furthermore, we ingeniously integrate hybrid attention mechanisms into various locations of...
Negative pressure (NP) therapy is effective in managing chronic lymphedema of the extremities. However, seal formation general head (GH) can fail due to interspace between lip and irregular skin surface on limbs before suction, resulting inefficiency therapy, prolonging time required for physiotherapy, increasing workload physiotherapists. In this letter, we present a bio-inspired (BIH) that uses adaptive control NP unstructured surfaces. Its designed with soft material inspired by bloodworm...
Abstract Background Compliance with colonoscopy among elderly individuals participating in colorectal cancer (CRC) screening programs is unsatisfactory, despite a high detection rate of bowel‐related diseases. In this study, our aim was to analyze the impact risk factors on trends compliance and rates high‐risk aged 60–74. Methods A retrospective study conducted 60–74 2021 CRC program Tianjin, China. Logistic regression analyses, including both univariate multivariate were performed explore...