We are in the process of consolidating the code and moving them to a new repository.

For technical details, please reference the following paper:

• The Active Atlas: Combining 3D Anatomical Models with Texture Detectors, Yuncong Chen et al., MICCAI 2017 [pdf]

User Guides

Explanations for how to install and run the code, intended for the non-programmer, are in the User Guide.

Code Structure

Main Pipeline

• annotation: code related to processing human annotations
• gui: code for two GUIs
  • brain labeling GUI (used to annotate landmarks in brain)
  • preprocessing GUI: processing raw images, remoing background, aligning stack.
• learning: code for extracting patches, generating features, training classifiers and processing score maps
• preprocess: code for preprocessing
• reconstruct: code for reconstructing volumes
• registration: code for 3D registration
• utilities: utilities code
• visualization: notebook for testing visualization code
• web_service: code for the web service that accept request for preprocessing and other tasks from GUI.
• intensity: codes for running intensity-based registration methods (via third-party program elastix).

Utilities

• utilities: utility modules
• 3d: code for rendering and displaying 3D models. Implementation uses VTK.
• aws: setup scripts for cfncluster.

Side Projects

• spm: experiments about represent textures using SPM (Spatial Pyramid Matching)
• snake: experiments about active contour methods
• adaboost_m2
• cells: classification using cell-based texture descriptors.
• dictionary: experiments for dictionary learning methods.
• new_region: