Julian de Wit

Thanks for the info.

Hello Thanks, Just my laziness. Usually images have a value range from 0-255 and I sometimes just use 0-255 as train target. But for Dice you need range 0-1. It's...

Hello, yes I think I did this. Perhaps it could be done in 1 step but I like taking small steps. So original label=>1 mm per voxel label->percentage label would...

Hello. Thanks. the ensembling happens in the "step 4....py" scripty. I gather the outputs of all networks en put them in a GBM model.

Hello Liu, There automatically generated labels (edges, false positives) There are manual nodule labels for LUNA / ndsb phase 1 trainset. There is also images + overlays for the U-net...

I did it myself with the help of the Kaggle trainings labels (bit still the results were worse than luna only.. I'm a bad radiologist :) ) I used a...

1st model was luna only. 2nd model was with manual labels and bit worse The combination was better than the two models separate.

MEAN PIXEL.. To get the average input to have a mean more like '0' which gives a more stable network calculation. Note that I did not adjust anymore after the...

Kaggle removed the data also to my dispair..

Hello. I did not write any paper. Perhaps the blog post :)