Page MenuHomePhabricator | MITK

isensee (Fabian Isensee)
User

Projects (7)

User Details

User Since
Aug 1 2016, 12:10 PM (115 w, 2 d)

Recent Activity

Jul 2 2018

isensee added a comment to T25020: Fabians Experiments.

Dice reweighting:
Oversampling foreground makes the dice believe the foreground is more frequent than it actually is. Therefore, it does not punish false positives hard enough. I therefore tried to balance the dice by doing the following:

Jul 2 2018, 3:23 PM · Restricted Project
isensee added a comment to T25020: Fabians Experiments.

Multiview:
Similar to Jakobs segmentation method. 3 UNets, each trained with a different perspective. Little improvement on Hippocampus, no difference on BraTS, but worse performance on other datasets -> Max be due to anisotropy

Jul 2 2018, 3:20 PM · Restricted Project
isensee created T25020: Fabians Experiments.
Jul 2 2018, 3:18 PM · Restricted Project