New RADS backend support (#9)

* Improving code clarity (separate widgets), upgrading to match with Raidionics v1.3
* Pure segmentation update
* Reporting for neuro and mediastinum applications
* Update to the structure
* Small fix to model download

Author: dbouget

CMakeLists.txt

@@ -0,0 +1,27 @@
+cmake_minimum_required(VERSION 3.5)
+
+project(Raidionics)
+
+#-----------------------------------------------------------------------------
+# Extension meta-information
+set(EXTENSION_HOMEPAGE "http://slicer.org/slicerWiki/index.php/Documentation/Nightly/Extensions/Raidionics")
+set(EXTENSION_CATEGORY "AI")
+set(EXTENSION_CONTRIBUTORS ", David Bouget (Medical Image Analysis group, Health Research Department, SINTEF Digital)")
+set(EXTENSION_DESCRIPTION "Module for pre-/post-operative central nervous system tumor segmentation and standardized reporting")
+set(EXTENSION_ICONURL "http://www.example.com/Slicer/Extensions/Raidionics.png")
+set(EXTENSION_SCREENSHOTURLS "http://www.example.com/Slicer/Extensions/Raidionics/Screenshots/1.png")
+set(EXTENSION_DEPENDS "NA") # Specified as a space separated string, a list or 'NA' if any
+
+#-----------------------------------------------------------------------------
+# Extension dependencies
+find_package(Slicer REQUIRED)
+include(${Slicer_USE_FILE})
+
+#-----------------------------------------------------------------------------
+# Extension modules
+add_subdirectory(Raidionics)
+## NEXT_MODULE
+
+#-----------------------------------------------------------------------------
+include(${Slicer_EXTENSION_GENERATE_CONFIG})
+include(${Slicer_EXTENSION_CPACK})

README.md

@@ -5,6 +5,7 @@
 [![License](https://img.shields.io/badge/License-BSD%202--Clause-orange.svg)](https://opensource.org/licenses/BSD-2-Clause)
 [![Paper](https://zenodo.org/badge/DOI/10.1038/s41598-023-42048-7.svg)](https://doi.org/10.1038/s41598-023-42048-7)
 [![Paper](https://zenodo.org/badge/DOI/10.3389/fneur.2022.932219.svg)](https://www.frontiersin.org/articles/10.3389/fneur.2022.932219/full)
+[![GitHub release](https://img.shields.io/github/v/release/raidionics/raidionics-slicer?sort=semver)](https://github.com/raidionics/raidionics-slicer/releases)
 
 
 Plugin for 3D Slicer to use the segmentation models and clinical reporting techniques (RADS) packaged in Raidionics.
@@ -13,14 +14,14 @@ The plugin was first introduced in the article _"Brain tumor preoperative surger
 segmentation and standardized reporting"_, published in [Frontiers in Neurology](https://www.frontiersin.org/articles/10.3389/fneur.2022.932219/full).  
 
 <p align="center">
-<img src="Raidionics/Raidionics/Resources/Icons/Raidionics-Slicer.gif" width="85%">
+<img src="Raidionics/Resources/Icons/Raidionics-Slicer.gif" width="85%">
 </p>
 </div>
 
 
 ## [Installation](https://github.com/raidionics/Raidionics-Slicer#installation)
 
-The plugin has been tested with the stable release 5.6.1 of 3D Slicer, and the upcoming release 5.7.0.  
+The plugin has been tested with the stable release 5.8.1 of 3D Slicer, and the upcoming release 5.9.0.  
 A [step-by-step video](https://www.youtube.com/watch?v=NStMzLcj1WE) for installing the plugin and running a segmentation model for the first time is available.
 
 <details>
@@ -38,12 +39,12 @@ A [step-by-step video](https://www.youtube.com/watch?v=NStMzLcj1WE) for installi
 * Download and install Docker (see [below](https://github.com/raidionics/Raidionics-Slicer#docker-setup--)).  
   The necessary _Raidionics_ Docker image should be collected automatically when downloading a model for the first time.
   Please do the following if it did not happen correctly:
-     > docker pull dbouget/raidionics-rads:v1.1-py38-cpu
+     > docker pull dbouget/raidionics-rads:v1.3-py39-cpu
 
 * Load the plugin into 3D Slicer:  
 	∘ All Modules > Extension Wizard.  
 	∘ Developer Tools > Extension Wizard.  
-	∘ Select Extension > point to the first Raidionics subfolder (inside Raidionics-Slicer) and add it to the path (tick the small box at the bottom).
+	∘ Select Extension > select your local Raidionics-Slicer folder and add it to the path (tick the small box at the bottom).
 
 * :warning: Restarting 3D Slicer to setup Python environment might be necessary on some occasions.
 
@@ -81,16 +82,17 @@ If you are using Raidionics-Slicer in your research, please use the following ci
 More information about the different models provided and architectures used can be accessed from the below-listed publications.  
 
 ### Neuro  
-* AGU-Net neural network architecture => [Meningioma Segmentation in T1-Weighted MRI Leveraging Global Context and Attention Mechanisms](https://www.frontiersin.org/articles/10.3389/fradi.2021.711514/full)
+* Raidionics => [Raidionics: an open software for pre-and postoperative central nervous system tumor segmentation and standardized reporting](https://www.nature.com/articles/s41598-023-42048-7)
+* Preoperative CNS segmentation performance => [Preoperative brain tumor imaging: models and software for segmentation and standardized reporting](https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2022.932219/full)
 * Standardized reporting and Data System => [Glioblastoma Surgery Imaging—Reporting and Data System: Standardized Reporting of Tumor Volume, Location, and Resectability Based on Automated Segmentations ](https://www.mdpi.com/2072-6694/13/12/2854)
 * Preoperative GBM segmentation performance => [Glioblastoma Surgery Imaging–Reporting and Data System: Validation and Performance of the Automated Segmentation Task ](https://www.mdpi.com/2072-6694/13/18/4674)
 * Postoperative GBM segmentation performance => [Segmentation of glioblastomas in early post-operative multi-modal MRI with deep neural networks](https://www.nature.com/articles/s41598-023-45456-x)
-* Preoperative CNS segmentation performance => [Preoperative brain tumor imaging: models and software for segmentation and standardized reporting](https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2022.932219/full)
+* AGU-Net neural network architecture => [Meningioma Segmentation in T1-Weighted MRI Leveraging Global Context and Attention Mechanisms](https://www.frontiersin.org/articles/10.3389/fradi.2021.711514/full)
 
 ### Mediastinum
 * Mediastinum organs segmentation => [Semantic segmentation and detection of mediastinal lymph nodes and anatomical structures in CT data for lung cancer staging](https://link.springer.com/article/10.1007/s11548-019-01948-8)  
 * Lymph nodes segmentation => [Mediastinal lymph nodes segmentation using 3D convolutional neural network ensembles and anatomical priors guiding](https://www.tandfonline.com/doi/pdf/10.1080/21681163.2022.2043778)
-* Airways segmentation => [AeroPath: An airway segmentation benchmark dataset with challenging pathology](https://arxiv.org/abs/2311.01138)
+* Airways segmentation => [AeroPath: An airway segmentation benchmark dataset with challenging pathology](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0311416)
 
 </details>
 
@@ -124,7 +126,7 @@ All images will be automatically downloaded upon model selection, which might
 take some minutes while the 3D Slicer interface won't be responding.  
 
 * The main Docker image can also be downloaded manually by:   
-    > docker pull dbouget/raidionics-rads:v1.1-py38-cpu
+    > docker pull dbouget/raidionics-rads:v1.3-py39-cpu
 
 * When you execute for the first time, you might get a pop-up from Docker asking to allow
 the sharing of a `.raidonics-slicer/` directory, accept!

Raidionics.s4ext

@@ -16,14 +16,12 @@
 depends ScriptedModule
 
 # homepage
-homepage  http://www.slicer.org
+homepage  https://raidionics.github.io/
 
 # match category in the xml description of the module (where it shows up in Modules menu)
 category  Machine Learning
 
-# give people an idea what to expect from this code 
-# - is it just a test or something you stand beind?
 status    Beta
 
 # One line description of what the module does
-description This is an extension module written in python for Deep Learning inference
+description Module for pre-/post-operative central nervous system tumor segmentation and standardized reporting

Raidionics/CMakeLists.txt

@@ -1,27 +1,30 @@
-cmake_minimum_required(VERSION 3.5)
-
-project(Raidionics)
-
 #-----------------------------------------------------------------------------
-# Extension meta-information
-set(EXTENSION_HOMEPAGE "http://slicer.org/slicerWiki/index.php/Documentation/Nightly/Extensions/Raidionics")
-set(EXTENSION_CATEGORY "AI")
-set(EXTENSION_CONTRIBUTORS ", David Bouget (Medical Technology, SINTEF Digital)")
-set(EXTENSION_DESCRIPTION "This is an example of a simple extension")
-set(EXTENSION_ICONURL "http://www.example.com/Slicer/Extensions/Raidionics.png")
-set(EXTENSION_SCREENSHOTURLS "http://www.example.com/Slicer/Extensions/Raidionics/Screenshots/1.png")
-set(EXTENSION_DEPENDS "NA") # Specified as a space separated string, a list or 'NA' if any
+set(MODULE_NAME Raidionics)
 
 #-----------------------------------------------------------------------------
-# Extension dependencies
-find_package(Slicer REQUIRED)
-include(${Slicer_USE_FILE})
+set(MODULE_PYTHON_SCRIPTS
+  ${MODULE_NAME}.py
+  )
+
+set(MODULE_PYTHON_RESOURCES
+  Resources/Icons/${MODULE_NAME}.png
+  )
 
 #-----------------------------------------------------------------------------
-# Extension modules
-add_subdirectory(Raidionics)
-## NEXT_MODULE
+slicerMacroBuildScriptedModule(
+  NAME ${MODULE_NAME}
+  SCRIPTS ${MODULE_PYTHON_SCRIPTS}
+  RESOURCES ${MODULE_PYTHON_RESOURCES}
+  WITH_GENERIC_TESTS
+  )
 
 #-----------------------------------------------------------------------------
-include(${Slicer_EXTENSION_GENERATE_CONFIG})
-include(${Slicer_EXTENSION_CPACK})
+if(BUILD_TESTING)
+
+  # Register the unittest subclass in the main script as a ctest.
+  # Note that the test will also be available at runtime.
+  slicer_add_python_unittest(SCRIPT ${MODULE_NAME}.py)
+
+  # Additional build-time testing
+  add_subdirectory(Testing)
+endif()

Raidionics/Raidionics/Raidionics.py Raidionics/Raidionics.pyRaidionics/Raidionics/Raidionics.py renamed to Raidionics/Raidionics.py

@@ -30,7 +30,7 @@ def __init__(self, parent):
     self.parent.title = "Raidionics"
     self.parent.categories = ["Machine Learning"]
     self.parent.dependencies = []
-    self.parent.contributors = ["David Bouget (Medical Technology, SINTEF Digital) david.bouget@sintef.no"]
+    self.parent.contributors = ["David Bouget (Medical Image Analysis group, Health Research Department, SINTEF Digital) david.bouget@sintef.no"]
     self.parent.helpText = """
     The Raidionics plugin for 3D Slicer allows users to run pre-trained models for segmentation and standardized
     reporting (RADS) over MRI volumes for patients diagnosed with brain cancer, and CT volumes for patients diagnosed