|
Motion Correction with Locally Linear Embedding for Helical Photon-Counting CT
RPI ID: 2022-041-301
Innovation Summary:
This invention introduces a novel motion correction framework for photon-counting computed tomography (PCCT) using locally linear embedding (LLE). The method addresses motion artifacts by modeling the high-dimensional projection data as a low-dimensional manifold and applying LLE to recover motion-free representations....
Published: 7/21/2025
|
Updated: 7/8/2025
|
Inventor(s): Mengzhou Li, Ge Wang
Keywords(s): helical CT scan, locally linear embedding, motion estimation, photon-counting detectors (PCDs), Rigid motion correction
Category(s): Media, Arts, Science and Technology
|
|
Translation Based Low-end Computed Tomography
RPI ID: 2014-031-201Innovation Summary:This invention introduces a linear scanning architecture for computed tomography (CT) that eliminates the need for rotational gantries. The system uses linearly translating X-ray sources and detectors moving in opposite or near-opposite directions. This arrangement enables tomographic imaging in environments where...
Published: 7/21/2025
|
Updated: 7/7/2025
|
Inventor(s): Wenxiang Cong, Fenglin Liu, Ge Wang, Hengyong Yu
Keywords(s):
Category(s): Biotechnology and the Life Sciences
|
|
Rapid Filtration Methods for Dual-energy X-ray CT
RPI ID: 2016-056-401, 2016-056-601, 2016-056-602, 2016-056-603Innovation Summary:This portfolio features a suite of advanced filtration technologies developed to enhance the performance of dual-energy X-ray computed tomography (DECT) systems. These innovations address key limitations in spectral separation, motion artifact reduction, and computational...
Published: 7/21/2025
|
Updated: 7/7/2025
|
Inventor(s): Wenxiang Cong, Yan Xi, Ge Wang
Keywords(s): Biomedical Engineering, computed tomography (CT), Diagnosis, Dual-energy X-ray CT, dual-layer x-ray detection, dual-source scanning, Imaging, kVp-switching, Measurement, Safety
Category(s): Biotechnology and the Life Sciences
|
|
X-ray Phase-contrast and Dark-field Information Extraction with Electric Fringe Scanning and/or Active Pixel Processing
RPI ID: 2016-003-401Innovation Summary:This invention introduces a next-generation X-ray imaging system that extracts both phase-contrast and dark-field data without requiring mechanical grating translation. By integrating the analyzer grating (G2) functionality directly into the detector and controlling it electrically, the system eliminates the need...
Published: 7/21/2025
|
Updated: 7/7/2025
|
Inventor(s): Shuo Pang, Zaifeng Shi, Wenxiang Cong, Ge Wang
Keywords(s): image reconstruction, x-ray dark-field imaging, X-ray grating imaging, x-ray phase imaging
Category(s): Biotechnology and the Life Sciences
|
|
Dual Network Architecture for Few-view CT – Trained on ImageNet Data and Transferred for Medical Imaging
RPI ID: 2020-001-401Innovation Summary:This invention introduces a deep learning-based system for reconstructing CT images from a limited number of views. The model is trained to infer missing projection data and generate high-quality images from sparse inputs. It significantly reduces radiation exposure while maintaining diagnostic accuracy. The system...
Published: 7/21/2025
|
Updated: 7/3/2025
|
Inventor(s): Ge Wang, Hongming Shan, Wenxiang Cong, Huidong Xie
Keywords(s): deep learning, Dual network architecture (DNA), few-view CT, generative adversarial network (GAN), machine-learning, sparse-view CT
Category(s): Biotechnology and the Life Sciences
|
|
Modularized Adaptive Processing Neural Network (MAP-NN) for Low-dose CT with Radiologists-in-the-loop
RPI ID: 2019-023-201Innovation Summary:This invention introduces a modularized adaptive processing neural network architecture designed for efficient and scalable AI applications. The system dynamically adjusts its structure and processing pathways based on input complexity and task requirements. It supports real-time learning and inference, making...
Published: 7/21/2025
|
Updated: 7/3/2025
|
Inventor(s): Ge Wang, Hongming Shan
Keywords(s): deep learning, denoising, machine-learning, radiologists-in-the-loop, tomography reconstruction
Category(s): Biotechnology and the Life Sciences
|
|
CT Super-resolution GAN Constrained by the Identical, Residual, and Cycle Learning Ensemble (GAN-CIRCLE)
RPI ID: 2019-005-201 / 2019-005-601Innovation Summary:This invention introduces a deep learning framework called GAN-Circle for super-resolution reconstruction in computed tomography (CT) imaging. The system is based on a generative adversarial network (GAN) architecture constrained by three synergistic learning strategies: identical mapping, residual...
Published: 7/21/2025
|
Updated: 7/3/2025
|
Inventor(s): Ge Wang, Chenyu You, Wenxiang Cong, Hongming Shan, Guang Li
Keywords(s): High-resolution CT image, Image super resolution, low-resolution CT image
Category(s): Computational Science and Engineering
|
|
A Synergized Pulsing-Imaging Network (SPIN)
RPI ID: 2018-029-401Innovation Summary:This invention introduces a synergized pulsing-imaging network (SPIN) that jointly optimizes MRI pulse sequences and image reconstruction. The system uses a deep learning framework to co-train the pulse generation and image reconstruction modules, improving image quality and acquisition efficiency. It adapts to...
Published: 7/21/2025
|
Updated: 7/3/2025
|
Inventor(s): Ge Wang, Qing Lyu, Tao Xu
Keywords(s): deep learning, machine-learning, medical imaging
Category(s): Biotechnology and the Life Sciences
|
|
Training a CNN with pseudo ground truth for CT metal artifact reduction
RPI ID: 2018-021-201 / 2018-021-601Innovation Summary:This invention introduces a method for training convolutional neural networks (CNNs) to reduce artifacts in CT images using pseudo ground truth data. The system includes an estimated ground truth generator that produces reference images from artifact-laden scans, enabling the CNN to learn artifact...
Published: 7/21/2025
|
Updated: 7/3/2025
|
Inventor(s): Ge Wang, Lars Gjesteby, Hongming Shan
Keywords(s): Biomedical Engineering, machine-learning, medical imaging
Category(s): Computational Science and Engineering
|
|
A Deep Neural Network based Corrector for Pulse Pileup Effect of Photon Counting Detectors
RPI ID: 2018-020-401 / 2018-020-601Innovation Summary:This invention introduces a neural network-based corrector for photon-counting computed tomography (PCCT) systems. Designed to address spectral distortions and detector non-linearities, the system processes raw photon data to generate corrected images with enhanced spectral fidelity and quantitative...
Published: 7/21/2025
|
Updated: 7/3/2025
|
Inventor(s): Ge Wang, Ruibin Feng, David Rundle
Keywords(s): deep learning, machine-learning, photon-counting Data
Category(s): Computational Science and Engineering
|
