Change image_utils.py to work with nnunetv2 predictions and ground truth labels with multi-class segmentations.

DIAGNijmegen / picai_eval

Evaluation of 3D detection and diagnosis performance —geared towards prostate cancer detection in MRI.

https://pi-cai.grand-challenge.org/

Apache License 2.0

20 stars 9 forks source link

Change image_utils.py to work with nnunetv2 predictions and ground truth labels with multi-class segmentations. #17

Closed NataliaAlves13 closed 7 months ago

NataliaAlves13 commented 7 months ago

The framework accepts .npz predictions with the "probabilities" key as produced by nnunetv2. The framework accepts ground truth labels with multi-class segmentations and assumes that the lesion segmentations correspond to label=1.