This demo demonstrates MRI reconstruction model described in https://arxiv.org/abs/1810.12473 and implemented in https://github.com/rmsouza01/Hybrid-CS-Model-MRI/. The model is used to restore undersampled MRI scans which is useful for data compression.
- hybrid-cs-model-mri
NOTE: Refer to the tables Intel's Pre-Trained Models Device Support and Public Pre-Trained Models Device Support for the details on models inference support at different devices.
- Running the application with the -h option yields the following usage message:
$ python3 mri_reconstruction_demo.py -h
usage: mri_reconstruction_demo.py [-h] -i INPUT -p PATTERN -m MODEL
[-d DEVICE] [--no_show]
MRI reconstrution demo
optional arguments:
-h, --help show this help message and exit
-i INPUT, --input INPUT
Path to input .npy file with MRI scan data.
-p PATTERN, --pattern PATTERN
Path to sampling mask in .npy format.
-m MODEL, --model MODEL
Path to .xml file of OpenVINO IR.
-d DEVICE, --device DEVICE
Optional. Specify the target device to infer on; CPU,
GPU, HDDL or MYRIAD is acceptable. Default value is
CPU.
--no_show Disable results visualization- To run the demo, you need to have
- A sample scan from Calgary-Campinas Public Brain MR Dataset
- Sampling mask