Evaluating the impact of MR image harmonization on thalamus deep network segmentation.
JournalProceedings of SPIE
PublisherInternational Society for Optics and Photonics
MetadataShow full item record
AbstractMedical image segmentation is one of the core tasks of medical image analysis. Automatic segmentation of brain magnetic resonance images (MRIs) can be used to visualize and track changes of the brain's anatomical structures that may occur due to normal aging or disease. Machine learning techniques are widely used in automatic structure segmentation. However, the contrast variation between the training and testing data makes it difficult for segmentation algorithms to generate consistent results. To address this problem, an image-to-image translation technique called MR image harmonization can be used to match the contrast between different data sets. It is important for the harmonization to transform image intensity while maintaining the underlying anatomy. In this paper, we present a 3D U-Net algorithm to segment the thalamus from multiple MR image modalities and investigate the impact of harmonization on the segmentation algorithm. Manual delineations of thalamic nuclei on two data sets are available. However, we aim to analyze the thalamus in another large data set where ground truth labels are lacking. We trained two segmentation networks, one with unharmonized images and the other with harmonized images, on one data set with manual labels, and compared their performances on the other data set with manual labels. These two data groups were diagnosed with two brain disorders and were acquired with similar imaging protocols. The harmonization target is the large data set without manual labels, which also has a different imaging protocol. The networks trained on unharmonized and harmonized data showed no significant difference when evaluating on the other data set; demonstrating that image harmonization can maintain the anatomy and does not affect the segmentation task. The two networks were evaluated on the harmonization target data set and the network trained on harmonized data showed significant improvement over the network trained on unharmonized data. Therefore, the network trained on harmonized data provides the potential to process large amounts of data from other sites, even in the absence of site-specific training data.
Identifier to cite or link to this itemhttp://hdl.handle.net/10713/18795
- Automatic upper airway segmentation in static and dynamic MRI via anatomy-guided convolutional neural networks.
- Authors: Xie L, Udupa JK, Tong Y, Torigian DA, Huang Z, Kogan RM, Wootton D, Choy KR, Sin S, Wagshul ME, Arens R
- Issue date: 2022 Jan
- Deep learning of the sectional appearances of 3D CT images for anatomical structure segmentation based on an FCN voting method.
- Authors: Zhou X, Takayama R, Wang S, Hara T, Fujita H
- Issue date: 2017 Oct
- Automatic segmentation of prostate MRI using convolutional neural networks: Investigating the impact of network architecture on the accuracy of volume measurement and MRI-ultrasound registration.
- Authors: Ghavami N, Hu Y, Gibson E, Bonmati E, Emberton M, Moore CM, Barratt DC
- Issue date: 2019 Dec
- IAS-NET: Joint intraclassly adaptive GAN and segmentation network for unsupervised cross-domain in neonatal brain MRI segmentation.
- Authors: Li B, You X, Wang J, Peng Q, Yin S, Qi R, Ren Q, Hong Z
- Issue date: 2021 Nov
- A conventional-to-spectral CT image translation augmentation workflow for robust contrast injection-independent organ segmentation.
- Authors: Lartaud PJ, Dupont C, Hallé D, Schleef A, Dessouky R, Vlachomitrou AS, Rouet JM, Nempont O, Boussel L
- Issue date: 2022 Feb