Skip to content

Importing to CERR

Aditya P. Apte edited this page Sep 16, 2021 · 14 revisions
Clone this wiki locally

CERR can import images and radiation therapy objects from various formats like DICOM, RTOG, mha. Various DICOM modalities that can be imported include CT, MR, PET, US, SPECT, RTSTRUCT, RTPLAN, RTDOSE, GSPS.

Instructions

  • Type 'CERR' into MATLAB's command window to launch the CERR startup screen.

  • Select the appropriate option (e.g. 'DICOM') under 'Import Study' and click Import.

  • Choose a directory to scan for DICOM files.

  • Once the scan is complete, select the patient data to be imported to CERR. Then select 'Import Patient'. In case different objects belonging to the same patient have different Study UIDs, select the "Import All" option to combine and import all associated data for a single patient.

  • Where multiple sequences are available, select 'Yes' to merge into a single sequence or "No" to keep store them separately.

  • When prompted, select the output folder and provide a file name, then click 'Save'. Choose to compress ("Yes") the file if required.

Support for Multiple series'

CERR identifies and supports the import of multiple series' contained in the chosen directory. If available, CERR provides an option at the end of import, to merge different scan sets, or store them separately. CERR provides the ability to separate MRI scan sets by b-value and acquisition time, and identifies them as distinct series'.

Batch-importing to CERR

The following code can be used to convert DICOM / RTOG files in a directory (including any files in subdirectories) to CERR format, and place the output in a specified folder

    % Define data source and destination paths 
    srcDir = 'Path\to\dataFolders';    
    dstDir = 'Path\to\outputFolder';

    %Set flags for compression and merging
    zipFlag = 'No'; %Set to 'Yes' for compression to bz2 zip 
    mergeFlag = 'No'; %Set to 'Yes' to merge all scans into a single series

    % For each sub-directory in srcDir, combine its contents into a planC.
    srcDir --
           dir1 ---
               file1
               file2
               subdir1
               subdir2
           dir2 ---
               ...
               ...
    dstDir (result of import) -- 
           dir1.mat
           dir2.mat

    init_ML_DICOM
    singleCerrFileFlag = 'no';
    batchConvertWithSubDirs(srcDir,dstDir,zipFlag,mergeFlag,singleCerrFileFlag)

    %For each sub-directory in srcDir, create separate planC's.
    srcDir --
           dir1 ---
               file1
               file2
               subdir1
               subdir2
           dir2 ---
               ...
               ...
    dstDir (result of import) -- 
           dir1.mat
           subdir1.mat
           subdir2.mat
           dir2.mat

    init_ML_DICOM;
    batchConvert (srcDir,dstDir,zipFlag,mergeFlag);

This also generates a spreadsheet listing the conversion status for each file, placed in the output folder.