Skip to content

Latest commit

 

History

History
executable file
·
44 lines (31 loc) · 1.56 KB

README.md

File metadata and controls

executable file
·
44 lines (31 loc) · 1.56 KB

Mandora: Machine Learning Reveals Multilevel Interconnections of Joint Damages in Rheumatoid Arthritis

Mandora is a machine learning approach for quantifying joint damages in rheumatoid arthritis (RA) based on radiographic images. It ranked first in predicting joint space narrowing in the RA2 DREAM Challenge with high accuracy on held-out testing data. Beyond high predictive performance, it offers a new way to investigate the cross regulatory relationships among joints and damage types by extracting the characteristic symmetrical patterns in RA. Please contact ([email protected] or [email protected]) if you have any questions or suggestions.

Figure1

Installation

Git clone a copy of code:

git clone https://github.com/GuanLab/Mandora.git

Required dependencies

  • python (3.6.5)
  • numpy (1.13.3). It comes pre-packaged in Anaconda.
  • tensorflow (1.14.0) A popular deep learning package.
conda install tensorflow-gpu
  • keras (2.2.4) A popular deep learning package using tensorflow backend.
conda install keras

Dataset

Code for localizing joints

  • code_location

Code for segmentation of joint space regions and prediction of damage scores

  • code_segmentation

Code for learning symmetry and SHAP analysis

  • code_symmetry