Abstract
This study compares two different methods for the task of brain segmentation in rodent MR-images, a convolutional neural network (CNN) and majority voting of a registration based atlas (RBA) , and how limited training data affect their performance. The CNN was implemented in Tensorflow.
The RBA performs better on average when using a training set with fewer than 20 images but the CNN achieves a higher median dice-score with a training set of 19 images.
The RBA performs better on average when using a training set with fewer than 20 images but the CNN achieves a higher median dice-score with a training set of 19 images.
Original language | English |
---|---|
Publication date | 2018 |
Publication status | Published - 2018 |
Event | Joint Annual Meeting ISMRM-ESMRMB 2018 - Paris, France Duration: 16 Jun 2018 → 21 Oct 2018 |
Conference
Conference | Joint Annual Meeting ISMRM-ESMRMB 2018 |
---|---|
Country/Territory | France |
City | Paris |
Period | 16/06/2018 → 21/10/2018 |