The thalamus is a subcortical brain structure linked to the motor system. Since certain changes within this structure are related to diseases, such as multiple sclerosis and Parkinson’s, the characterization of the thalamus—e.g., shape assessment—is a crucial step in relevant studies and applications, including medical research and surgical planning. A robust and reliable thalamus-segmentation method is therefore, required to meet these demands. Despite presenting low contrast for this particular structure, T1-weighted imaging is still the most common MRI sequence for thalamus segmentation. However, diffusion MRI (dMRI) captures different micro-structural details of the biological tissue and reveals more contrast of the thalamic borders, thereby serving as a better candidate for thalamus-segmentation methods. Accordingly, we propose a baseline multimodality thalamus-segmentation pipeline that combines dMRI and T1-weighted images within a CNN approach, achieving state-of-the-art levels of Dice overlap. Furthermore, we are hosting an open benchmark with a large, preprocessed, publicly available dataset that includes co-registered, T1-weighted, dMRI, manual thalamic masks; masks generated by three distinct automated methods; and a STAPLE consensus of the masks. The dataset, code, environment, and instructions for the benchmark leaderboard can be found on our GitHub and CodaLab.
Corpus Callosum (CC) segmentation is required when the analysis from this structure is desirable. Many of these studies require the CC segmentation on diffusion tensor images (DTI). However, few methods perform segmentation directly in the DTI. Segmenting on DTI makes it possible to disregard the registration step after segmenting on T1 images. This work studies the possibility of improving automated segmentation of the CC using silver standard annotations. With incomplete silver standard annotations, limited to 5 or 7 central slices, experiments performed throughout this work were done to compare methods of pre-training and fine tuning in an attempt to translate silver standard knowledge to improved performance in 3D CC segmentation. Experiments include 3D and 2D U-Net as deep learning architectures. Results point to central limited silver standard annotations not being useful for improving the performance in gold standard 3D annotations. Our best method involved training a 3D U-Net with gold standards and post processing, achieving a 3D Dice of 83.33 Dice, surpassing 2D U-Net.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.