Enhanced Systems and Methods for Phase Unwrapping in Dense MRI using Deep Learning

This inventive concept presents improved systems and methods for phase unwrapping in Dense MRI using deep learning, addressing the limitations of traditional techniques and enhancing the accuracy, efficiency, and real-time capabilities of strain analysis in cardiac imaging.

The original patent disclosed systems and methods for phase unwrapping in Dense MRI using deep learning, which had limitations in terms of noise and artifacts in the phase image, computational time required for phase unwrapping, and real-time strain analysis capabilities. This inventive concept addresses these limitations by introducing novel deep learning-based techniques for phase unwrapping, post-processing, and real-time strain analysis.

The new claims describe a system for phase unwrapping in DENSE MRI, comprising a deep learning module configured to generate a wrapping label map and a corresponding phase image, and a post-processing module to refine the phase image based on spatial and temporal consistency. The inventive concept also includes a method for improving the accuracy of strain analysis in DENSE MRI, which involves acquiring displacement encoded MRI data, generating a phase image using a deep learning-based phase unwrapping technique, and quantifying global and segmental strain associated with the heart of a subject. Additionally, the inventive concept encompasses a computer-readable medium storing instructions for performing a method of phase unwrapping in DENSE MRI, and a system for real-time strain analysis in DENSE MRI, comprising a deep learning module and a strain analysis module.

The new claims introduce novel deep learning-based techniques for phase unwrapping, post-processing, and real-time strain analysis, which are not obvious from the original patent. The inventive concept's novelty lies in the integration of these techniques to provide improved accuracy, efficiency, and real-time capabilities for strain analysis in cardiac imaging.

Alternative embodiments of the inventive concept could include the use of different deep learning architectures, such as convolutional neural networks (CNNs) or recurrent neural networks (RNNs), for phase unwrapping and strain analysis. Additionally, the inventive concept could be adapted for use in other medical imaging modalities, such as computed tomography (CT) or ultrasound.

The inventive concept has significant commercial potential in the medical imaging industry, particularly in the field of cardiac imaging. The improved accuracy, efficiency, and real-time capabilities of the inventive concept could lead to widespread adoption in hospitals and clinics, and could also enable new applications in personalized medicine and medical research.