Test MS segmentation models in 3T 7T MP2RAGE from Marseille

[Nilser3]

Model to test: model_seg_ms_mp2rage

release: r20230210

Preprocessing based on script: #63 (comment)

• 3T MP2RAGE ( r = 0.9 x 0.9 x 0.9 )

Related in #63 (comment)

Example of automatic segmentation:

DICE Score values (N=40)

Continuous Dice Coefficient for "soft masks" based on this implementation and the Classic Dice between the GT and binarized masks (thresholding at 1/5 = 0.2) because we have 5 segmentation models.

Analysis of subjects with Dice = 0

7T MP2RAGE

• isotropic images ( r = 0.7 x 0.7 x 0.7 )

• isotropic images resampled to 1mm isotropic
  
  

• anisotropic images (Same subject) ( r = 0.4 x 0.4 x 4 )
  
  

• anisotropic images resampled to 1mm isotropic
  
  

subject with 3 lesions, 2 were segmented, one was not detected

[jcohenadad]

Thank you for the thorough report @Nilser3, I find it very encouraging!