segmentation_metrics

segmentation_metrics copied to clipboard

Computing values on ground truth does not give perfect scores

5

Following one of the examples in the `README` file, a quick test using a ground truth image as the predicted image, ``` labels = [0, 1, 2] gdth_img = np.array([[0,0,1],...

jhlegarreta

Adding .dcm contour data

2

Hi, would it be possible to add the rt-struct formatted labels to the list?

lionmsk11

Add .png suffix

5

Hi, Is it possible to add the .png suffix please ? Thanks a lot

Noemie-B

Check msd computation

2

@Jingnan-Jia If `msd` is the mean (Average) surface distance, shouldn't it provide the exact same result as the `assd` value in `medpy`? ``` metrics = sg.write_metrics( labels=labels, gdth_img=gdth_np, pred_img=pred_np, spacing=spacing,...

jhlegarreta

Evaluation metric 3D or 2D based segmentation

1

Hi, If I used 2D configuration for a volumetric image, is the evaluation metric computed for the 3D volume or per 2D slices and then aggregating? Thanks

Merom99

Hi, Jingnan, Thank you for sharing this repository. I do have a question regarding to `spacing` in the input. Is it the same spacing between pixels? That leads to a...

wangtongy