A5019640194- Advanced MRI Techniques and Applications
- Medical Imaging Techniques and Applications
- MRI in cancer diagnosis
- Magnetic Properties and Applications
- Advanced X-ray and CT Imaging
- Intelligent Tutoring Systems and Adaptive Learning
- Reservoir Engineering and Simulation Methods
- Advanced NMR Techniques and Applications
- Quantum Chromodynamics and Particle Interactions
- Radiomics and Machine Learning in Medical Imaging
- Lanthanide and Transition Metal Complexes
- Dark Matter and Cosmic Phenomena
- Particle physics theoretical and experimental studies
- Seismic Imaging and Inversion Techniques
Tsinghua University
2024
Central China Normal University
2006
Institute of High Energy Physics
2006
Using a data sample of 58×106 J/ψ decays collected with the Beijing Spectrometer II detector at Electron Positron Collider, searches for invisible η and η′ in to ϕη ϕη′ are performed. The ϕ signals, which reconstructed K+K− final states, used tag decays. No signals found either or η′, upper limits 90% confidence level determined be 1.65×10−3 ratio B(η→invisible)B(η→γγ) 6.69×10−2 B(η′→invisible)B(η′→γγ). These first into states.Received 4 July...
Abstract Purpose To develop a highly accelerated CEST Z‐spectral acquisition method using specifically‐designed k‐space sampling pattern and corresponding deep‐learning‐based reconstruction. Methods For down‐sampling, customized was proposed for CEST, with the randomized probability following frequency‐offset‐dependent (FOD) function in direction of saturation offset. reconstruction, convolution network (CNN) enhanced Partially Separable (PS) to optimize spatial domain frequency separately....
Here we introduced a model-based deep learning approach which enabled joint optimization of both sampling and reconstruction for CEST MRI. The main purpose is to investigate an efficient undersampling pattern acceleration. Retrospective results (4X) showed that the proposed workflow capable leveraging redundancy information from optimized pattern, further reconstructing gold-standard-consistent contrast maps.
CEST MRI is a promising molecular imaging technique, but the in vivo quantification inherently faces challenges because of multiple types signal contaminations. Herein we introduced QUEST-like 10-point fitting method for simultaneous NOE and amide at 3T with reduced T1 contamination. The was evaluated by simulations phantom experiments. Furthermore, glioblastoma patients, highlighted regions on fitted images are consistent hyperintensities Gd-T1W maps, which smaller enhanced area than those...
The reasoning steps generated by LLMs might be incomplete, as they mimic logical leaps common in everyday communication found their pre-training data: underlying rationales are frequently left implicit (unstated). To address this challenge, we introduce RATIONALYST, a model for process-supervision of based on vast collection rationale annotations extracted from unlabeled data. We extract 79k web-scale unlabelled dataset (the Pile) and combination datasets with minimal human intervention....
Motivation: The two-shot SNAP MRI is effective for carotid plaque diagnosis with extended scan time. To accelerate the scan, under-sampling reconstruction and optimization of sampling locations are considered. Goal(s): optimize masks IR-TFE REF-TFE respectively to reconstruct under-sampled images higher quality. Approach: After parameterization ky-kz two shots, a model-based deep learning framework was utilized achieve goals. Results: demonstrated superior performance compared other methods....
Motivation: As an exciting ‘label-free' molecular imaging technique, CEST workflow is always time-consuming, because of the seconds-long TR and multiple frequency repetitions in acquisition, iteration reconstruction, pixel-by-pixel B0 correction quantification. Goal(s): To achieve rapid high-quality sampling, reconstruction quantification CEST-MRI. Approach: We constructed a data-driven framework, by joint optimization k-space Results: Retrospective experiments on human brain...
Motivation: Chemical Exchange Saturation Transfer (CEST) acceleration requires robust contrast recovery from under-sampled k-space data. Goal(s): To achieve accelerated CEST-MRI with well-preserved among different tissue types. Approach: Herein we proposed a reconstruction method that iteratively decomposed both K-space and Image domains into Low-rank plus Sparse components, termed as KILS. Results: Retrospective experiments the healthy adults brain tumor patients indicated KILS could an 8X...
Motivation: Non-ideal RF slice-profile can affect the accuracy of MRF. However, correcting this effect in dictionary simulation is time-consuming. Goal(s): To propose an improved algorithm which correct MRF accurately and efficiently. Approach: We method combines spinor rotation representation extended-phase-graph(EPG) algorithm. The proposed validated by retrospective experiments on brain T1 T2 maps, compared with EPG Results: non-ideal slice profile much faster computational speed. Impact:...