Why not Matlab? If the reason is the price tag, you could try Octave, an open source Matlab clone...

Anyway, C++ would be the most obvious option, but not the cheaper (in terms of programmer hours). If you are familiar with OpenGL, you could use Java or Python too.

Dear Leiner,

At my opinion, it's impossible to give simple answer. As mentioned C++, Java or Python will be good, but you must find approciate software library for it, like OpenCV. Also you can look for R language.

Also you can look for Interactive Data Language (IDL), but it's proprietary software.

regards

In my University there is a software development center for medicine that works with C, C ++ and Java. regards

These days nearly all software that is intended to be used by someone other than its own author requires a multi-language paradigm. If you need a "friendly user interface", you want to choose a programming system that interfaces with the applications UI most easily. In Windows, that is often C#. But for the computational kernels, you'll choose or build libraries that are designed and optimized for speed. C and C++ are the most popular for that purpose. Where correctness is important, typed functional languages like Haskell, OCaml and even Scala provde significant advantages. For general developer productivity, i.e. quickly delivering lots of functionality, Python often turns out to be quite effective.

But if your interaction model is web-based with a multi-server backend, a framework that is oriented to that type of architecture may be better, like Ruby on Rails. Web frontends are almost invariably in Javascript.

In the end, the most productive programming language is the one that the developer is the most fluent in. Use whatever you're most comfortable with for most of the system, and use highly optimized tools only for the most critical portions.

Hello!

I would suggest C++. And here are the reasons for such a suggestion:

• powerfull Visualisation Tool Kit (VTK) library that provides some instruments for volumetric data visualisation (https://www.vtk.org/).
• Insight Segmentation and Registration Toolkit (ITK) that provides implementation of different algorithms for medical volume segmentation (https://itk.org/).
• Qt - one of the best (in my opinion - THE best) library for GUI and much more. Free and open source. Commercial license also available. You may need it if you are going to create closed source application and you are using some modules like QChart. (https://www.qt.io/)
• Common Tool Kit (CTK) that provides some useful UI elements for use with VTK and CTK as well as some utility classes (http://www.commontk.org/index.php/Main_Page).
• Grassroots DICOM (GDCM) library to work with DICOM files. I cannot say that I like the way it is implemented (some parts are strange), but it works pretty well and has support for a lot of different compression formats, used in DICOM files. For me it just works. (https://sourceforge.net/projects/gdcm/)
• Boost.Units for type safe dimensional analysis. Unfortunately, I did not use it yet, but it allows you to write code that has information about measurement units you use and may be checked at compile time to prevent problems with different implementation of floating point values. (http://www.boost.org/doc/libs/release/doc/html/boost_units.html)

These are the options that may be useful when it comes to medical imaging, segmentation, etc.

I work on image processing since 10 years, and i always use C++, and if you want win time, you have to use matlab.

3D Slicer is an excellent option. It is free, versatile and has a strong visualitation cappabilities.

In my university, reserchers use C, Matlab, C++ and Java

Best regards,

Wiem

Suggesting C++ for its visualisation/fast performance if you have grip on it, opencv can be use with any of the language.

C/C++ . Most of developed medical libraries (ITK, DICOM read/write standards, PACS, etc.) written in c/c++. Or matlab for fast development :)