CEST Contrast Image Reconstruction via PROPELLER Undersampling and Deep Neural Network
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current CEST imaging techniques face challenges in achieving rapid acquisition of high-quality contrast images due to long scanning times and unsatisfactory signal-to-noise ratios, with existing methods either failing to shorten scanning time effectively or being unsuitable for complex samples.
Innovation Solution
A method and system utilizing PROPELLER undersampling combined with a deep neural network for reconstructing CEST contrast images, which generates training samples to train the network for efficient image reconstruction from undersampled data, accounting for magnetization transfer and noise effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If FSE readout module is used to acquire CEST contrast image, then image quality is improved, but scanning time is excessively long
Solution Approach 1:
The patent applies PROPELLER (Periodically Rotated Overlapping Parallel Lines with Enhanced Reconstruction) technology to segment and rotate the acquisition of k-space data. Instead of acquiring the entire image sequentially as in FSE, the imaging space is divided into multiple rotating blades that are acquired periodically and overlapped, enabling parallel processing and significantly reducing scanning time while maintaining image quality through the overlapping reconstruction process.
2Loss of time
If EPI readout module is used to shorten scanning time, then scanning time is reduced, but image distortion increases
Solution Approach 1:
The patent employs dynamic rotating blade acquisition where the PROPELLER blades are periodically rotated and overlapped during data acquisition. This dynamic approach allows the system to capture k-space data from multiple angular orientations, enabling correction of distortion through the overlapping regions while maintaining the fast acquisition speed of EPI-like sequences.
3Loss of time
If CEST-FISP method is used to shorten scanning time, then scanning time is reduced, but signal-to-noise ratio decreases
Solution Approach 1:
The PROPELLER method implements continuous useful action by periodically rotating and overlapping the acquisition blades throughout the scanning process. This continuous overlapping acquisition ensures that each region of the image is sampled multiple times from different angular positions, accumulating signal while averaging out noise, thereby maintaining high signal-to-noise ratio even with accelerated scanning.
4Loss of time
If Radial UCI method is used to shorten scanning time, then scanning time is reduced, but applicability to complex samples is limited
Solution Approach 1:
The PROPELLER method provides universal applicability by using periodically rotated overlapping parallel lines that can effectively sample complex sample structures regardless of their orientation or complexity. The multiple angular orientations of the rotating blades ensure comprehensive coverage of complex samples, making the method versatile for various sample types while maintaining fast acquisition speed.
Data Source
AI summary
Disclosed is a method for reconstruction of a Chemical Exchange Saturation Transfer (CEST) contrast image. The method includes: generating training samples for a deep neural network; training the deep neural network with the training samples to obtain a trained deep neural network; and reconstructing a CEST contrast image by using the trained deep neural network and PROPELLER undersampled CEST images. The method for reconstruction of a CEST contrast image can effectively shorten the experimental time of a CEST contrast imaging and can obtain a smoother and more accurate CEST contrast image. Further disclosed is a system for reconstruction of a CEST contrast image to implement the method for reconstruction.