Merge branch 'master' of https://github.com/CBICA/BrainMaGe

BrainMaGe/tester/test_ma.py

@@ -166,7 +166,7 @@ def infer_ma(cfg, device, save_brain, weights):
             for i in range(to_save_final.shape[2]):
                 if np.any(to_save_final[:, :, i]):
                     to_save_final[:, :, i] = binary_fill_holes(to_save_final[:, :, i])
-            to_save_final = getLargestCC(to_save_final)
+            to_save_final = getLargestCC(to_save_final).astype(np.uint8)
             to_save_final_nib = nib.Nifti1Image(to_save_final,
                                                 current_patient_dict['old_affine'])
 

BrainMaGe/tester/test_multi_4.py

@@ -115,7 +115,7 @@ def infer_multi_4(cfg, device, save_brain, weights):
             for i in range(to_save.shape[2]):
                 if np.any(to_save[:, :, i]):
                     to_save[:, :, i] = binary_fill_holes(to_save[:, :, i])
-            to_save = getLargestCC(to_save)
+            to_save = getLargestCC(to_save).astype(np.uint8)
             to_save_mask = nib.Nifti1Image(to_save, patient_nib.affine)
             nib.save(to_save_mask, os.path.join(params['results_dir'], patient[0],
                                                 patient[0]+'_mask.nii.gz'))

README.md

@@ -140,13 +140,12 @@ python setup.py install # install dependencies and BrainMaGe
 - We provide CPU (untested as of 2020/05/31) as well as GPU support. 
   - Running on GPU is a lot faster though and should always be preferred. 
   - You need an GPU memory of ~5-6GB for testing and ~8GB for training.
-
+- Added support for hole filling and largest CCA post processing
 ## TO-DO
 
 - Windows support (this currently works but needs a few work-arounds)
 - Give example of skull stripping dataset 
 - In inference, rename ```model_dir``` to ```results_dir``` for clarity in the configuration and script(s)
-- Add CCA for post-processing
 - Test on CPU
 - Move all dependencies to ```setup.py``` for consistency 
 - Put option to write logs to specific files in output directory