Transfer learning by feature-space transformation: A method for Hippocampus segmentation across scanners

TY - JOUR

T1 - Transfer learning by feature-space transformation

T2 - A method for Hippocampus segmentation across scanners

AU - van Opbroek, Annegreet

AU - Achterberg, Hakim C

AU - Vernooij, Meike W

AU - Arfan Ikram, M

AU - de Bruijne, Marleen

PY - 2018

Y1 - 2018

N2 - Many successful approaches in MR brain segmentation use supervised voxel classification, which requires manually labeled training images that are representative of the test images to segment. However, the performance of such methods often deteriorates if training and test images are acquired with different scanners or scanning parameters, since this leads to differences in feature representations between training and test data. In this paper we propose a feature-space transformation (FST) to overcome such differences in feature representations. The proposed FST is derived from unlabeled images of a subject that was scanned with both the source and the target scan protocol. After an affine registration, these images give a mapping between source and target voxels in the feature space. This mapping is then used to map all training samples to the feature representation of the test samples. We evaluated the benefit of the proposed FST on hippocampus segmentation. Experiments were performed on two datasets: one with relatively small differences between training and test images and one with large differences. In both cases, the FST significantly improved the performance compared to using only image normalization. Additionally, we showed that our FST can be used to improve the performance of a state-of-the-art patch-based-atlas-fusion technique in case of large differences between scanners.

AB - Many successful approaches in MR brain segmentation use supervised voxel classification, which requires manually labeled training images that are representative of the test images to segment. However, the performance of such methods often deteriorates if training and test images are acquired with different scanners or scanning parameters, since this leads to differences in feature representations between training and test data. In this paper we propose a feature-space transformation (FST) to overcome such differences in feature representations. The proposed FST is derived from unlabeled images of a subject that was scanned with both the source and the target scan protocol. After an affine registration, these images give a mapping between source and target voxels in the feature space. This mapping is then used to map all training samples to the feature representation of the test samples. We evaluated the benefit of the proposed FST on hippocampus segmentation. Experiments were performed on two datasets: one with relatively small differences between training and test images and one with large differences. In both cases, the FST significantly improved the performance compared to using only image normalization. Additionally, we showed that our FST can be used to improve the performance of a state-of-the-art patch-based-atlas-fusion technique in case of large differences between scanners.

KW - Classification

KW - Domain adaptation

KW - Hippocampus

KW - MRI

KW - Segmentation

KW - Transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85051644174&partnerID=8YFLogxK

U2 - 10.1016/j.nicl.2018.08.005

DO - 10.1016/j.nicl.2018.08.005

M3 - Journal article

C2 - 30128285

SN - 2213-1582

VL - 20

SP - 466

EP - 475

JO - NeuroImage: Clinical

JF - NeuroImage: Clinical

ER -