Search Results - chuang+niu

13 Results Sort By:
HOME: High-Order Mixed-Moment-based Embedding for Representation Learning
RPI ID: 2023-002-301, 2023-002-401 Innovation Summary: This technology introduces a novel embedding framework for representation learning that leverages high-order mixed statistical moments to capture complex data relationships. The system enhances feature extraction by encoding richer structural and distributional information from input data. It...
Published: 3/10/2026   |   Inventor(s): Chuang Niu, Ge Wang
Keywords(s): Artificial Intelligence, deep learning, self-supervised reapresentation learning
Category(s): Computational Science and Engineering
X-Ray Dissectography
RPI ID: 2022-016-401 / 2022-016-301 Innovation Summary: X-ray dissectography enhances radiographic imaging by dissecting overlapping anatomical layers using contrast modulation and spatial filtering. The technique improves depth resolution and diagnostic clarity in complex regions. It is particularly effective for identifying subtle pathologies that...
Published: 3/10/2026   |   Inventor(s): Chuang Niu, Ge Wang
Keywords(s): Atificial Intellegnece, Desectography, X-ray Imaging
Category(s): Biotechnology and the Life Sciences
Self-Supervised Representation Learning Through Multi-Segmental Discriminative Coding
RPI ID: 2022-058-301, 2022-058-401 Innovation Summary: This invention introduces a self-supervised representation learning (SSRL) system that utilizes multi-segmental informational coding to enhance data representation. The SSRL circuitry includes transformer-based components designed to process input data without requiring labeled examples. By leveraging...
Published: 3/10/2026   |   Inventor(s): Ge Wang, Chuang Niu
Keywords(s): Artificial Intelligence, deep learning, self-supervised reapresentation learning
Category(s): Computational Science and Engineering
AI Enables Ultra-Low-Dose CT Reconstruction
RPI ID: 2021-072-401 Innovation Summary: AI-enabled ultra-low-dose CT reconstruction systems use deep learning to restore image quality from low-radiation scans. The models are trained on paired datasets to enhance resolution and contrast while minimizing noise. This approach allows for safer imaging without compromising diagnostic accuracy. It is...
Published: 3/10/2026   |   Inventor(s): Mannudeep Kalra, Chuang Niu, Hengyong Yu, Shadi Ebrahimian, Weiwen Wu, Ge Wang
Keywords(s): deep learning, Deep Tomographic Reconstruction, Few-view, Ultra-low-dose
Category(s): Computational Science and Engineering
Noise2Sim – Similarity-based Self-Learning for Image Denoising
RPI ID: 2021-034-401 Innovation Summary: Noise2Sim is a self-supervised image denoising method that leverages similarity-based learning without requiring clean reference images. The algorithm identifies consistent patterns across noisy inputs and uses them to reconstruct cleaner outputs. It is particularly effective in medical and scientific imaging...
Published: 3/10/2026   |   Inventor(s): Chuang Niu, Ge Wang
Keywords(s): deep learning, Image Denoising, self-learning, self-similarity
Category(s): Computational Science and Engineering
Low-dimensional manifold constrained disentanglement network for metal artifact reduction in CT images
RPI ID: 2020-111-201 Innovation Summary: A deep learning network is designed to reduce metal artifacts in CT imaging by constraining feature disentanglement within a low-dimensional manifold. The model separates anatomical structures from artifact-induced distortions, improving image clarity. It is trained on synthetic and real-world datasets and supports...
Published: 3/10/2026   |   Inventor(s): Chuang Niu, Wenxiang Cong, Ge Wang
Keywords(s): disentanglement network, low-dimensional manifold model (LDMM), metal artifact reduction, unpaired learning
Category(s): Computational Science and Engineering
Parallelizing the Diffusion Model for Breast CT Image Reconstruction from Sparse Cone-beam Projections
RPI ID: 2023-078-301 Innovation Summary: This technology introduces a tomographic image reconstruction system that leverages parallel denoising diffusion probabilistic models (DDPMs) to recover high-quality images from sparse projection data. The system uses a probabilistic framework to iteratively refine image estimates, enabling accurate reconstructions...
Published: 3/10/2026   |   Inventor(s): Wenjun Xia, Chuang Niu, Wenxiang Cong, Ge Wang
Keywords(s): Breast CT, deep reconstruction, denoising diffusion probabilistic model (DDPM), distributed computing, dual domain, parallel computing, sub-volume
Category(s): Computational Science and Engineering
Synthesizing Big Data of High Quality without Privacy Leakage –Competitive Performance of Deep CT Denoising Networks Trained on Diffusion Model-generated Data
RPI ID: 2023-075-301 Innovation Summary: This technology introduces systems and methods for synthesizing image data using super-resolution (SR) techniques. It leverages multiple low-resolution (LR) images to generate high-resolution (HR) outputs by extracting and combining fine-grained features across frames. The system is designed to enhance image...
Published: 3/10/2026   |   Inventor(s): Wenjun Xia, Yongyi Shi, Chuang Niu, Ge Wang
Keywords(s): Diffusion Models, Low-Dose CT, Privacy Protection
Category(s): Computational Science and Engineering
Medical Multi-Task Learning with a Large Image-Language Model
RPI ID: 2023-052-301 Innovation Summary: This technology presents a foundation model for medical applications that integrates multimodal data (e.g., imaging, text, signals) and supports multitask learning. The model is trained on diverse clinical datasets to perform diagnostics, prognostics, and treatment recommendations. It leverages deep learning...
Published: 3/10/2026   |   Inventor(s): Ge Wang, Chuang Niu
Keywords(s): biomedical analysis, biomedical imaging, CAD, large lanuage models, radiology reprot generation, visual-language models
Category(s): Computational Science and Engineering
Processing Technique for Ceramic- based Energy Conversion and Storage Devices
Due to rising carbon emissions, extreme weather events and other environmental threats, the demand for alternative energy sources is greater than ever. Therefore, businesses and communities are asking for three things: more affordable electricity, more resilient power, and cleaner energy. Fuel cells, electrolysis cells, membrane reactors, and solid-state...
Published: 10/27/2025   |   Inventor(s): Hai Xiao, Jianhua Tong, Fei Peng, Kyle Brinkman, Jincheng Lei, Yuzhe Hong, Hua Huang, Shenglong Mu, Rajendra Bordia
Keywords(s):  
Category(s): Manufacturing
1 2