This repository serves as the main hub for all code & resources related to the tabletop MRI systems at Stanford University.
To get started, these videos and files can serve as basic orientation to the systems.
- https://www.youtube.com/watch?v=-oz5ecPS6VQ&t=561s
- https://www.youtube.com/watch?v=9qmofVIGHZ4&t=3s
- https://www.sciencedirect.com/science/article/pii/S1090780719302642?via%3Dihub
- https://zeugmatographix.org/ocra/
March 9: updated to latest release of Relax 2.0 software
February 28: gradient channels y and z2 not working
- As evidenced by shim calibration returning flat lines for these channels
- TODO Ariel - pick up Tabletop from Q-Bio
February 28: Relax 2.0 software version predates latest release
Shim calibration prone to freezing
- This has only been observed on ocra2, so may be related to shim channels malfunctioning
Certain button clicks will crash the Relax 2.0 program
- appears to be sequence-dependent; full extent of issue is unknown, but certainly includes some 2D imaging sequences
- Commenting lines of code with parameter ”frameon=” resolves the issue
- TODO identify the specific button/sequence combinations that trigger these errors
Reducing FOV does not shorten scan time as expected (fewer phase encode lines)
- Observed by Daniel
- Alex suspects this is more of a suboptimal UI design than a bug: the “Resolution” field does not set pixel size directly, but rather the number of pixels (the image matrix size). Hence changes to the field of view result in inversely proportional changes to the pixel size in order to preserve the matrix size.
Unexpected bimodal signal observed in FID spectrum
- Observed by Daniel & Mark using agar/oil phantom
- TODO further investigation is needed comparing FID vs spin echo for single- and multi-component phantoms and recording the parameters used
- TODO check whether poor shim could explain
Time signal becomes zero at long sampling times
- Observed by Daniel
Correct typos and misleading labels in Parameters menu
- Generell -> General
- Resolution -> Pixel Count or Image Matrix Size
Synchronization of code between tabletop systems
- Current status quo results in bug fixes not always distributed to other systems
Visualize pulse sequences to help with debugging and learning
Force quit a scan without killing the server
Automatic pre-scan
- Run center frequency calibration prior to a scan
- Create option in main menu to toggle this feature on/off
- Consider adding other abbreviated calibration routines e.g. shimming
Improve shim calibration
- Further automation to reduce manual effort
- Improve algorithm to achieve lower B0 inhomogeneity (currently 20-40ppm)
- We may find that hardware is the limiting factor here
Facilitate pulse sequence development & deployment through Pulseq integration
Web app to allow remote access bypassing Stanford network
Automate phantom loading/unloading for remote use with multiple phantoms
- Test latest release of firmware and software (March 8). Revise Bugs and Feature Requests accordingly.
- Communicate with Marcus and Thomas about residual issues pertaining to the software and hardware.
- Decide on software platform moving forward.
- Tend to bugs and feature requests.