Introduction: Volumetric and morphological changes in subcortical brain structures are present in persons with dementia, but it is unknown if these changes occur prior to diagnosis. Methods: Between 2005 and 2016, 5522 Rotterdam Study participants (mean age: 64.4) underwent cerebral magnetic resonance imaging (MRI) and were followed for development of dementia until 2018. Volume and shape measures were obtained for seven subcortical structures. Results: During 12 years of follow-up, 272 dementia cases occurred. Mean volumes of thalamus (hazard ratio [HR] per standard deviation [SD] decrease 1.94, 95% confidence interval [CI]: 1.55-2.43), amygdala (HR 1.66,, and hippocampus (HR 1.64, 95% CI: 1.43-1.88) were strongly associated with dementia risk. Associations for accumbens, pallidum, and caudate volumes were less pronounced. Shape analyses identified regional surface changes in the amygdala, limbic thalamus, and caudate.Discussion: Structure of the amygdala, thalamus, hippocampus, and caudate is associated with risk of dementia in a large population-based cohort of older adults.

No abstract

No abstract

Segmenting deep brain structures from magnetic resonance images is important for patient diagnosis, surgical planning, and research. Most current state-of-the-art solutions follow a segmentation-byregistration approach, where subject MRIs are mapped to a template with well-defined segmentations. However, registration-based pipelines are time-consuming, thus, limiting their clinical use. This paper uses deep learning to provide a robust and efficient deep brain segmentation solution. The method consists of a pre-processing step to conform all MRI images to the same orientation, followed by a convolutional neural network using the nnU-Net framework. We use a total of 14 datasets from both research and clinical collections. Of these, seven were used for training and validation and seven were retained for independent testing. We trained the network to segment 30 deep brain structures, as well as a brain mask, using labels generated from a registration-based approach. We evaluated the generalizability of the network by performing a leave-one-dataset-out cross-validation, and extensive testing on external datasets. Furthermore, we assessed cross-domain transportability by evaluating the results separately on different domains. We achieved an average DSC of 0.89 ± 0.04 on the independent testing datasets when compared to the registration-based gold standard. On our test system, the computation time decreased from 42 minutes for a reference registration-based pipeline to 1 minute. Our proposed method is fast, robust, and generalizes with high reliability. It can be extended to the segmentation of other brain structures. The method is publicly available on GitHub, as well as a pip package for convenient usage.

Segmenting deep brain structures from magnetic resonance images is important for patient diagnosis, surgical planning, and research. Most current state-of-the-art solutions follow a segmentation-by-registration approach, where subject magnetic resonance imaging (MRIs) are mapped to a template with well-defined segmentations.However, registration-based pipelines are time-consuming, thus, limiting their clinical use. This paper uses deep learning to provide a one-step, robust, and efficient deep brain segmentation solution directly in the native space. The method consists of a preprocessing step to conform all MRI images to the same orientation, followed by a convolutional neural network using the nnU-Net framework. We use a total of 14 datasets from both research and clinical collections. Of these, seven were used for training and validation and seven were retained for testing. We trained the network to segment 30 deep brain structures, as well as a brain mask, using labels generated from a registration-based approach. We evaluated the generalizability of the network by performing a leave-one-dataset-out cross-validation, and independent testing on unseen datasets. Furthermore, we assessed cross-domain transportability by evaluating the results separately on different domains. We achieved an average dice score similarity of 0.89 ± 0.04 on the test datasets when compared to the registration-based gold standard. On our test system, the computation time decreased from 43 min for a reference registration-based pipeline to 1.3 min. Our proposed method is fast, robust, and generalizes with high reliability. It can be extended to the segmentation of other brain structures. It is publicly available on GitHub, and as a pip package for convenient usage.