An official website of the United States government
7T magnetic resonance imaging (MRI) has the potential to drive our understanding of human brain function through new contrast and enhanced resolution. Whole brain segmentation is a key neuroimaging technique that allows for region-by-region analysis of the brain. Segmentation is also an important preliminary step that provides spatial and volumetric information for running other neuroimaging pipelines. Spatially localized atlas network tiles (SLANT) is a popular 3D convolutional neural network (CNN) tool that breaks the whole brain segmentation task into localized sub-tasks. Each sub-task involves a specific spatial location handled by an independent 3D convolutional network to provide high resolution whole brain segmentation results. SLANT has been widely used to generate whole brain segmentations from structural scans acquired on 3T MRI. However, the use of SLANT for whole brain segmentation from structural 7T MRI scans has not been successful due to the inhomogeneous image contrast usually seen across the brain in 7T MRI. For instance, we demonstrate the mean percent difference of SLANT label volumes between a 3T scan-rescan is approximately 1.73%, whereas its 3T-7T scan-rescan counterpart has higher differences around 15.13%. Our approach to address this problem is to register the whole brain segmentation performed on 3T MRI to 7T MRI and use this information to finetune SLANT for structural 7T MRI. With the finetuned SLANT pipeline, we observe a lower mean relative difference in the label volumes of ~8.43% acquired from structural 7T MRI data. Dice similarity coefficient between SLANT segmentation on the 3T MRI scan and the after finetuning SLANT segmentation on the 7T MRI increased from 0.79 to 0.83 with p<0.01. These results suggest finetuning of SLANT is a viable solution for improving whole brain segmentation on high resolution 7T structural imaging.
more »
« less
Precise and rapid whole-head segmentation from magnetic resonance images of older adults using deep learning
Abstract Whole-head segmentation from Magnetic Resonance Images (MRI) establishes the foundation for individualized computational models using finite element method (FEM). This foundation paves the path for computer-aided solutions in fields such as non-invasive brain stimulation. Most current automatic head segmentation tools are developed using healthy young adults. Thus, they may neglect the older population that is more prone to age-related structural decline such as brain atrophy. In this work, we present a new deep learning method called GRACE, which stands for General, Rapid, And Comprehensive whole-hEad tissue segmentation. GRACE is trained and validated on a novel dataset that consists of 177 manually corrected MR-derived reference segmentations that have undergone meticulous manual review. Each T1-weighted MRI volume is segmented into 11 tissue types, including white matter, grey matter, eyes, cerebrospinal fluid, air, blood vessel, cancellous bone, cortical bone, skin, fat, and muscle. To the best of our knowledge, this work contains the largest manually corrected dataset to date in terms of number of MRIs and segmented tissues. GRACE outperforms five freely available software tools and a traditional 3D U-Net on a five-tissue segmentation task. On this task, GRACE achieves an average Hausdorff Distance of 0.21, which exceeds the runner-up at an average Hausdorff Distance of 0.36. GRACE can segment a whole-head MRI in about 3 seconds, while the fastest software tool takes about 3 minutes. In summary, GRACE segments a spectrum of tissue types from older adults’ T1-MRI scans at favorable accuracy and speed. The trained GRACE model is optimized on older adult heads to enable high-precision modeling in age-related brain disorders. To support open science, the GRACE code and trained weights are made available online and open to the research community at https://github.com/lab-smile/GRACE.
more »
« less
- PAR ID:
- 10539638
- Publisher / Repository:
- MIT Press Direct
- Date Published:
- Journal Name:
- Imaging Neuroscience
- Volume:
- 2
- ISSN:
- 2837-6056
- Page Range / eLocation ID:
- 1 to 21
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The hippocampus is a crucial brain structure involved in memory formation, spatial navigation, emotional regulation, and learning. An accurate MRI image segmentation of the human hippocampus plays an important role in multiple neuro-imaging research and clinical practice, such as diagnosing neurological diseases and guiding surgical interventions. While most hippocampus segmentation studies focus on using T1-weighted or T2-weighted MRI scans, we explore the use of diffusion-weighted MRI (dMRI), which offers unique insights into the microstructural properties of the hippocampus. Particularly, we utilize various anisotropy measures derived from diffusion MRI (dMRI), including fractional anisotropy, mean diffusivity, axial diffusivity, and radial diffusivity, for a multi-contrast deep learning approach to hippocampus segmentation. To exploit the unique benefits offered by various contrasts in dMRI images for accurate hippocampus segmentation, we introduce an innovative multimodal deep learning architecture integrating cross-attention mechanisms. Our proposed framework comprises a multi-head encoder designed to transform each contrast of dMRI images into distinct latent spaces, generating separate image feature maps. Subsequently, we employ a gated cross-attention unit following the encoder, which facilitates the creation of attention maps between every pair of image contrasts. These attention maps serve to enrich the feature maps, thereby enhancing their effectiveness for the segmentation task. In the final stage, a decoder is employed to produce segmentation predictions utilizing the attention-enhanced feature maps. The experimental outcomes demonstrate the efficacy of our framework in hippocampus segmentation and highlight the benefits of using multi-contrast images over single-contrast images in diffusion MRI image segmentation.more » « less
-
In the medical sector, three-dimensional (3D) images are commonly used like computed tomography (CT) and magnetic resonance imaging (MRI). The 3D MRI is a non-invasive method of studying the soft-tissue structures in a knee joint for osteoarthritis studies. It can greatly improve the accuracy of segmenting structures such as cartilage, bone marrow lesion, and meniscus by identifying the bone structure first. U-net is a convolutional neural network that was originally designed to segment the biological images with limited training data. The input of the original U-net is a single 2D image and the output is a binary 2D image. In this study, we modified the U-net model to identify the knee bone structures using 3D MRI, which is a sequence of 2D slices. A fully automatic model has been proposed to detect and segment knee bones. The proposed model was trained, tested, and validated using 99 knee MRI cases where each case consists of 160 2D slices for a single knee scan. To evaluate the model’s performance, the similarity, dice coefficient (DICE), and area error metrics were calculated. Separate models were trained using different knee bone components including tibia, femur, patella, as well as a combined model for segmenting all the knee bones. Using the whole MRI sequence (160 slices), the method was able to detect the beginning and ending bone slices first, and then segment the bone structures for all the slices in between. On the testing set, the detection model accomplished 98.79% accuracy and the segmentation model achieved DICE 96.94% and similarity 93.98%. The proposed method outperforms several state-of-the-art methods, i.e., it outperforms U-net by 3.68%, SegNet by 14.45%, and FCN-8 by 2.34%, in terms of DICE score using the same dataset.more » « less
-
Sex differences in brain structure significantly influence traumatic brain injury (TBI) onset and progression, yet this area is understudied. Herein, we developed sex-specific brain anatomical (macroscale) and axonal tract (mesoscale) templates and explored the sex variations at subject level using a set of T1-MRI (609 males, 721 females) and DTI images (506 males, 594 females). The FreeSurfer, ANTs, and DSI-Studio packages were used. We investigated overall/regional volumes, DTI metrics (including fractional anisotropy (FA), mean diffusivity, and radial diffusivity), and connectivity matrix across 23 brain regions. The brain connectome was derived by multiplying the fiber tract counts and the FA values within the connecting tracts, quantifying the connection strength within each pair of regions. Our subject-wise analysis revealed significant sex based differences (Mann-Whitney p-values < 0.05) across most studied regions for all parameters. The largest sex differences in brain connections were observed in five regions: corpus callosum and right/left cortex and cerebral white matter, all stronger in females. Brain regions were typically larger in males, yet females had higher fractional volumes in the majority of regions except for CSF and ventricles, known for their cushioning effect during head impacts. Additionally, the sex-specific templates better represented their targeted sex compared to opposite or mixed-sex populations as evaluated by root-mean-square-errors when comparing the DTI metrics and connectivity from the DTI templates against the median of subjects and deformation field in registering the subjects to the T1-MRI templates. Our findings highlight the necessity of sex-specific templates in accurate brain modeling and TBI research.more » « less
-
Creative divergent thinking involves the generation of unique ideas by pulling from semantic memory stores and exercising cognitive flexibility to shape these memories into something new. Although cognitive abilities such as episodic memory decline with age, semantic memory tends to remain intact. The current study aims to take advantage of older adults’ strength in semantic memory to investigate the effectiveness of a brief cognitive training to improve creative divergent thinking. Specifically, older adults were trained using a semantic retrieval strategy known as the disassembly strategy in order to improve creativity in the Alternate Uses Task (AUT), which involves generating original uses for objects. We also investigated whether this strategy would transfer to other creativity tasks, specifically, the Divergent Association Task (DAT). Participants were tested on the AUT and DAT across three time points in a single session: before the disassembly strategy was introduced (T0 and T1) and afterwards (T2). Results showed that the disassembly strategy enhances idea novelty in the AUT, though this enhancement did not transfer to DAT performance. Additionally, participants that initially scored lowest on the AUT at T0 showed the greatest increase in AUT performance at T2. This finding provides evidence that older adults can effectively use a semantic retrieval strategy to engage and enhance elements of creative divergent thinking.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1162/imag_a_00090
