A5101968853- Pituitary Gland Disorders and Treatments
- VLSI and FPGA Design Techniques
- Image Processing and 3D Reconstruction
- 3D Surveying and Cultural Heritage
- Advanced Neural Network Applications
- Handwritten Text Recognition Techniques
- 3D Shape Modeling and Analysis
- Meningioma and schwannoma management
- Computer Graphics and Visualization Techniques
- VLSI and Analog Circuit Testing
- Advancements in Photolithography Techniques
- Cerebral Venous Sinus Thrombosis
- Hand Gesture Recognition Systems
- Robot Manipulation and Learning
- Advanced Multi-Objective Optimization Algorithms
- Additive Manufacturing and 3D Printing Technologies
- Generative Adversarial Networks and Image Synthesis
- Advanced Control Systems Design
- IoT and Edge/Fog Computing
- Blockchain Technology Applications and Security
- Power Systems and Renewable Energy
- Glioma Diagnosis and Treatment
- Additive Manufacturing Materials and Processes
- Injection Molding Process and Properties
- Image and Object Detection Techniques
Peking Union Medical College Hospital
2021-2024
Chinese Academy of Medical Sciences & Peking Union Medical College
2021-2024
Carnegie Mellon University
2018-2023
Beijing Institute of Technology
2021
Zhongyuan University of Technology
2013
Abstract Global routing has been a historically challenging problem in the electronic circuit design, where challenge is to connect large and arbitrary number of components with wires without violating design rules for printed boards or integrated circuits. Similar problems also exist complex hydraulic systems, pipe logistic networks. Existing solutions typically consist greedy algorithms hard-coded heuristics. As such, existing approaches suffer from lack model flexibility usually fail...
While millions of scanned engineering drawings are received every year, the online quotation companies for custom mechanical parts have experienced a surging need to increase their processing efficiency by replacing currently manual inspection process with an automatic system. Previous work has used traditional, and data-driven computer-vision approaches detect symbols text information from drawings. However, there lacks unified framework determine associated manufacturing processes as...
The resection plan of pituitary adenoma (PA) needs preoperative observation the sellar region. Radiomics prediction requires high-quality segmentations. Manual delineation is time-consuming and subject to rater variability.This work aims create an automated segmentation method for region, several tools extract invasiveness-related features, evaluate their clinical usefulness by predicting tumor consistency.Patients included were diagnosed with at Peking Union Medical College Hospital. A deep...
Abstract We propose a data-driven 3D shape design method that can learn generative model from corpus of existing designs, and use this to produce wide range new designs. The approach learns an encoding the samples in training using unsupervised variational autoencoder-decoder architecture, without need for explicit parametric representation original To facilitate generation smooth final surfaces, we develop based on distance transformation data, rather than commonly utilized binary voxel...
Abstract In the physical design of integrated circuits, global and detailed routing are critical stages involving determination interconnected paths each net on a circuit while satisfying constraints. Existing actual routers as well routability predictors either have to resort expensive approaches that lead high computational times, or use heuristics do not generalize well. Even though new, learning-based methods been proposed address this need, requirements labelled data difficulties in...
Background There are no established accurate models that use machine learning (ML) methods to preoperatively predict immediate remission after transsphenoidal surgery (TSS) in patients diagnosed with histology-positive Cushing’s disease (CD). Purpose Our current study aims devise and assess an ML-based model TSS CD. Methods A total of 1,045 participants CD who received at Peking Union Medical College Hospital a 20-year period (between February 2000 September 2019) were enrolled the present...
Convolutional neural network (CNN) is a deep-learning method for image classification and recognition based on multi-layer NN. In this study, CNN was used to accurately assess cavernous sinus invasion (CSI) in pituitary adenoma (PA).A total of 371 patients with PA were enrolled the retrospective study. The cohort divided into invasive (n = 102) non-invasive groups 269) surgically confirmed CSI. Images selected T1-enhanced imaging MR scans. underwent fivefold division randomized datasets...
In the physical design of integrated circuits, global and detailed routing are critical stages involving determination interconnected paths each net on a circuit while satisfying constraints. Existing actual routers as well routability predictors either have to resort expensive approaches that lead high computational times, or use heuristics do not generalize well. Even though new, learning-based methods been proposed address this need, requirements labelled data difficulties in addressing...
We propose a data-driven 3D shape design method that can learn generative model from corpus of existing designs, and use this to produce wide range new designs. The approach learns an encoding the samples in training using unsupervised variational autoencoder-decoder architecture, without need for explicit parametric representation original To facilitate generation smooth final surfaces, we develop based on distance transformation data, rather than commonly utilized binary voxel...
The ability to track human operators' hand usage when working in production plants and factories is critically important for developing realistic digital factory simulators as well manufacturing process control. We propose a proof-of-concept instrumented glove with only few strain gage sensors microcontroller that continuously tracks records the configuration during actual use. At heart of our approach trainable system can predict fourteen joint angles using small set sensors. First, ten...
Abstract We present a new data generation method to facilitate an automatic machine interpretation of 2D engineering part drawings. While such drawings are common medium for clients encode design and manufacturing requirements, lack computer support automatically interpret these necessitates manufacturers resort laborious manual approaches which, in turn, severely limits processing capacity. Although recent advances trainable vision methods may enable interpretation, it remains challenging...
In this paper, the system identification toolbox is used to identify model of DC motor. The results show that method simple and effective. view optimization problem controller, an automatic design PID parameters based on Matlab environment presented. This applies active set algorithm realize parameters. simulation simple, simulated effects are ideal stability good.
The ability to track human operators’ hand usage when working in production plants and factories is critically important for developing realistic digital factory simulators as well manufacturing process control. We propose an instrumented glove with only a few strain gauge sensors micro-controller that continuously tracks records the configuration during actual use. At heart of our approach trainable system can predict fourteen joint angles using small set sensors. First, ten gauges are...
Abstract With the advancement of technology and economy, scale smart home industry has exploded. While improving people’s lives, systems are facing threats to privacy leaks, malicious attacks, structural security. Effective security mechanisms very important for protecting valuable data in systems. Access control helps information prevent access, thereby reducing risk leakage. Blockchain can provide support Internet Things system due its advantages decentralization immutability. Therefore,...
The access position and capacity of distribution generation (DG) affect the static voltage stability micro-grid, thus affecting renewable energy utilization. In current reform supply side, a multi-objective optimization model is established, aiming at abandoning wind light problem. This has three advantages, which are largest utilization, micro-grid minimum cost DG investment considering environmental benefits. It can effectively promote use power, photovoltaic power other sources. this...
Global routing has been a historically challenging problem in electronic circuit design, where the challenge is to connect large and arbitrary number of components with wires without violating design rules for printed boards or integrated circuits. Similar problems also exist complex hydraulic systems, pipe systems logistic networks. Existing solutions typically consist greedy algorithms hard-coded heuristics. As such, existing approaches suffer from lack model flexibility non-optimum...
We present a data-driven framework to automate the vectorization and machine interpretation of 2D engineering part drawings. In industrial settings, most manufacturing engineers still rely on manual reads identify topological requirements from drawings submitted by designers. The process is laborious time-consuming, which severely inhibits efficiency quotation tasks. While recent advances in image-based computer vision methods have demonstrated great potential interpreting natural images...