-
Notifications
You must be signed in to change notification settings - Fork 5
Installation and Use
The following sections provide instructions on installing, configuring, and running the TomoTherapy® Failure Modes and Effects Analysis (FMEA) Tool.
To install the TomoTherapy FMEA Simulation Tool, copy all MATLAB .m, .fig files, and submodules (dicom_tools, tomo_extract, structure_atlas, and gamma) into a directory with read/write access. If using git, execute the following command:
git clone --recursive https://github.com/mwgeurts/systematic_error
Next, copy the following beam model files in a folder named GPU in the same directory. These files will be copied to the computation server along with the plan files at the time of program execution. To change the location of this folder, edit the line modeldir = './GPU'; in the function AutoSystematicError().
- dcom.header
- lft.img
- penumbra.img
- kernel.img
- fat.img
The TomoTherapy FMEA Simulation Tool must be configured to either calculate dose locally or communicate with a dose calculation server. If using local calculation, gpusadose must be installed in an execution path available to MATLAB. If using a remote server, open CalcDose(), find the statement ssh2 = ssh2_config('tomo-research', 'tomo', 'hi-art');, and enter the IP/DNS address of the dose computation server (tomo-research, for example), a user account on the server (tomo), and password (hi-art). This user account must have SSH access rights, rights to execute gpusadose, and finally read/write acces to the temp directory. See Accuray Incorporated to see if your research workstation includes this feature. For additional information, see the tomo_extract submodule.
Refer to the Software Compatibility wiki page for more information on Operating System and MATLAB version compatibility.
Global configuration variables such as Gamma criteria and the expected beam on time can be modified by changing the values in AutoSystematicError() prior to execution. A full list of variables is provided in the following table.
| Global Variable | Default Value | Description |
|---|---|---|
| anon | false | Anonymous flag. If true, only the archive XML SHA1 signature and MVCT UID will be stored, and no additinal patient identifying information will be included in the results |
To run this application, call the function AutoSystematicError() from MATLAB with no input arguments. As described below, the application will find all patient archives (filename appended with "_patient.xml") within a specified input directory and run the FMEA simulation. To change the search directory, edit the inputDir declaration statement in the function AutoSystematicError(). When searching the input directory, the folder lists are intentionally randomized such that the plans are processed by the tool in a random order.
Because this application runs without a user interface, it may also be executed via a terminal, as shown in the following example:
/Applications/MATLAB_R2015b.app/bin/matlab -nodesktop -r AutoSystematicError
Home | Installation and Use | Troubleshooting | ©2015 University of Wisconsin Board of Regents