Is the fastest way to learn MATLAB DICOM manipulation (e.g. structure translation) through working with CERR?

What is the fastest way to learn DICOM manipulation in MATLAB, shift a structure, and write and manage DICOM files?

I have been given the task to shift a contoured structure in radiation therapy patient DICOM files in MATLAB. (It is for research, not clinical use.) I am new to working with DICOM in MATLAB. First I tried reading the MathWorks documentation for the provided DICOM commands, but the documentation is limited such that use of these commands seems to require familiarity with DICOM.

I then moved to working with CERR, but I am struggling to get it working with MATLAB R2016a. It has obsolete webpages online, inapplicable documentation, and the code appears simply broken, e.g. apparently referencing a variable or class before creating it. The last version appears one year old, with documentation even older, so I wonder if it has been superseded by other MATLAB packages or simply abandoned.

I searched the available DICOM packages uploaded by others, but found only viewers and content not relevant to my work.

Is the fastest way to try to get CERR working? (I have attempted to join the CERR Google Group to discuss encountered errors, e.g. the undefined variable or class mentioned above, but my membership is still pending, possibly due to Google administration limitations at my university.)

Or is the fastest way to resume using the provided DICOM functions together with analyzing the code provided in File Exchange and MathWorks Examples? (I must learn what the data used to visualize the structures is, where it is stored, how to access it and manipulate it, etc.)

Are there other resources you would recommend? (textbooks, publications, webpages)

Currently, I am inclined to proceed based on the MathWorks documentation and build my code and knowledgebase "from the ground up" reading examples like Josh Schaefferkoetter's, but I would like to know if there is a faster way. I am concerned that proceeding with CERR may "waste time" debugging others' code (e.g. learning about the dcm4che2 DICOM Toolkit and figuring out why CERR's invocation of it isn't working) and waiting for collaborators' responses, but since I lack experience working with DICOM, I am aware that it could be a massive undertaking, and that it might be much faster to "get CERR working" rather than "start from scratch".

Edit: I am currently proceeding through reading documentation and others' miscellaneous code, e.g. ulrikls's dicomrt2matlab and Jun Li's EsmeProcess.

Answer 1

[Daniel Bridges]

In my experience, working with CERR and other DICOM scripts in the File Exchange required analyzing others' code, often learning other MATLAB commands, and dozens of different files and thousands of lines of code were too much to sift through. I found it faster to begin "coding from scratch", looking at MATLAB documentation, the official DICOM support forum, and the DICOM Standard. Working with its PDF publication, parts 3 and 6 have been most useful.

Answer 2

[Classic]

Can you use CERR and MAtlab????

[Daniel Bridges]

No, I encountered runtime errors. I was not able to use the CERR software to accomplish my tasks.