IMJENSE: Scan-specific Implicit Representation for Joint Coil Sensitivity and Image Estimation in Parallel MRI
Authors:
Ruimin Feng,
Qing Wu,
Jie Feng,
Huajun She,
Chunlei Liu,
Yuyao Zhang,
Hongjiang Wei
Abstract:
Parallel imaging is a commonly used technique to accelerate magnetic resonance imaging (MRI) data acquisition. Mathematically, parallel MRI reconstruction can be formulated as an inverse problem relating the sparsely sampled k-space measurements to the desired MRI image. Despite the success of many existing reconstruction algorithms, it remains a challenge to reliably reconstruct a high-quality im…
▽ More
Parallel imaging is a commonly used technique to accelerate magnetic resonance imaging (MRI) data acquisition. Mathematically, parallel MRI reconstruction can be formulated as an inverse problem relating the sparsely sampled k-space measurements to the desired MRI image. Despite the success of many existing reconstruction algorithms, it remains a challenge to reliably reconstruct a high-quality image from highly reduced k-space measurements. Recently, implicit neural representation has emerged as a powerful paradigm to exploit the internal information and the physics of partially acquired data to generate the desired object. In this study, we introduced IMJENSE, a scan-specific implicit neural representation-based method for improving parallel MRI reconstruction. Specifically, the underlying MRI image and coil sensitivities were modeled as continuous functions of spatial coordinates, parameterized by neural networks and polynomials, respectively. The weights in the networks and coefficients in the polynomials were simultaneously learned directly from sparsely acquired k-space measurements, without fully sampled ground truth data for training. Benefiting from the powerful continuous representation and joint estimation of the MRI image and coil sensitivities, IMJENSE outperforms conventional image or k-space domain reconstruction algorithms. With extremely limited calibration data, IMJENSE is more stable than supervised calibrationless and calibration-based deep-learning methods. Results show that IMJENSE robustly reconstructs the images acquired at 5$\mathbf{\times}$ and 6$\mathbf{\times}$ accelerations with only 4 or 8 calibration lines in 2D Cartesian acquisitions, corresponding to 22.0% and 19.5% undersampling rates. The high-quality results and scanning specificity make the proposed method hold the potential for further accelerating the data acquisition of parallel MRI.
△ Less
Submitted 21 November, 2023;
originally announced November 2023.
Directed Energy Interception of Satellites
Authors:
Harrison She,
Will Hettel,
Philip Lubin
Abstract:
High power Earth and orbital-based directed energy (DE) systems pose a potential hazard to Earth orbiting spacecraft. The use of very high power, large aperture DE systems to propel spacecraft is being pursued as the only known, feasible method to achieve relativistic flight in our NASA Starlight and Breakthrough Starshot programs. In addition, other beamed power mission scenarios, such as orbital…
▽ More
High power Earth and orbital-based directed energy (DE) systems pose a potential hazard to Earth orbiting spacecraft. The use of very high power, large aperture DE systems to propel spacecraft is being pursued as the only known, feasible method to achieve relativistic flight in our NASA Starlight and Breakthrough Starshot programs. In addition, other beamed power mission scenarios, such as orbital debris removal and our NASA program using DE for powering high performance ion engine missions, pose similar concerns. It is critical to quantify the probability and rates of interception of the DE beam with the approximately 2000 active Earth orbiting spacecraft. We have modeled the interception of the beam with satellites by using their orbital parameters and computing the likelihood of interception for many of the scenarios of the proposed systems we are working on. We are able to simulate both the absolute interception as well as the distance and angle from the beam to the spacecraft, and have modeled a number of scenarios to obtain general probabilities. We have established that the probability of beam interception of any active satellite, including its orbital position uncertainty, during any of the proposed mission scenarios is low ($\approx10^{-4}$). The outcome of this work gives us the ability to predict when to energize the beam without intercept, as well as the capability to turn off the DE as needed for extended mission scenarios. As additional satellites are launched, our work can be readily extended to accommodate them. Our work can also be used to predict interception of astronomical adaptive optics guide-star lasers as well as more general laser use.
△ Less
Submitted 27 February, 2019; v1 submitted 24 September, 2018;
originally announced September 2018.
