The Latest Articles on Neuroimaging and Neurodegenerative Disease

Our staff editors continue to share exciting, interesting, and thought-provoking reading material in the recommended articles series.

This week, we would like to share several latest articles on Neuroimaging and Neurodegenerative Disease

Title: Assessment of synaptic loss in mouse models of β-amyloid and tau pathology using [18F]UCB-H PET imaging
Authors: Letizia Vogler, Anna Ballweg, Bernd Bohr, Nils Briel, Karin Wind, Melissa Antons, Lea Kunze, Johannes Gnörich, Simon Lindner, Franz-Josef Gildehaus, Karlheinz Baumann, Peter Bartenstein, Guido Boening, Sibylle I. Ziegler, Johannes Levin, Andreas Zwergal, Günter U. Höglinger, Jochen Herms, Matthias Brendel
Type: Research Article
Abstract:

Objective

In preclinical research, the use of [18F]Fluorodesoxyglucose (FDG) as a biomarker for neurodegeneration may induce bias due to enhanced glucose uptake by immune cells. In this study, we sought to investigate synaptic vesicle glycoprotein 2A (SV2A) PET with [18F]UCB-H as an alternative preclinical biomarker for neurodegenerative processes in two mouse models representing the pathological hallmarks of Alzheimer’s disease (AD).

Methods

A total of 29 PS2APP, 20 P301S and 12 wild-type mice aged 4.4 to 19.8 months received a dynamic [18F]UCB-H SV2A-PET scan (14.7 ± 1.5 MBq) 0-60 minutes post injection. Quantification of tracer uptake in cortical, cerebellar and brainstem target regions was implemented by calculating relative volumes of distribution (VT) from an image-derived-input-function (IDIF). [18F]UCB-H binding was compared across all target regions between transgenic and wild-type mice. Additional static scans were performed in a subset of mice to compare [18F]FDG and [18F]GE180 (18 kDa translocator protein tracer as a surrogate for microglial activation) standardized uptake values (SUV) with [18F]UCB-H binding at different ages. Following the final scan, a subset of mouse brains was immunohistochemically stained with synaptic markers for gold standard validation of the PET results.

Results

[18F]UCB-H binding in all target regions was significantly reduced in 8-months old P301S transgenic mice when compared to wild-type controls (temporal lobe: p = 0.014; cerebellum: p = 0.0018 ; brainstem: p = 0.0014). Significantly lower SV2A tracer uptake was also observed in 13-months (temporal lobe: p = 0.0080; cerebellum: p = 0.006) and 19-months old (temporal lobe: p = 0.0042; cerebellum: p = 0.011) PS2APP transgenic versus wild-type mice, whereas the brainstem revealed no significantly altered [18F]UCB-H binding. Immunohistochemical analyses of post-mortem mouse brain tissue confirmed the SV2A PET findings. Correlational analyses of [18F]UCB-H and [18F]FDG using Pearson’s correlation coefficient revealed a significant negative association in the PS2APP mouse model (R = -0.26, p = 0.018). Exploratory analyses further stressed microglial activation as a potential reason for this inverse relationship, since [18F]FDG and [18F]GE180 quantification were positively correlated in this cohort (R = 0.36, p = 0.0076)

Conclusion

[18F]UCB-H reliably depicts progressive synaptic loss in PS2APP and P301S transgenic mice, potentially qualifying as a more reliable alternative to [18F]FDG as a biomarker for assessment of neurodegeneration in preclinical research.

Access this article: https://doi.org/10.1016/j.nicl.2023.103484

Title: SF2Former: Amyotrophic lateral sclerosis identification from multi-center MRI data using spatial and frequency fusion transformer
Authors: Rafsanjany Kushol, Collin C. Luk, Avyarthana Dey, Michael Benatar, Hannah Briemberg, Annie Dionne, Nicolas Dupré, Richard Frayne, Angela Genge, Summer Gibson, Simon J. Graham, Lawrence Korngut, Peter Seres, Robert C. Welsh, Alan H. Wilman, Lorne Zinman, Sanjay Kalra, Yee-Hong Yang
Type: Research article
Abstract:
Amyotrophic Lateral Sclerosis (ALS) is a complex neurodegenerative disorder characterized by motor neuron degeneration. Significant research has begun to establish brain magnetic resonance imaging (MRI) as a potential biomarker to diagnose and monitor the state of the disease. Deep learning has emerged as a prominent class of machine learning algorithms in computer vision and has shown successful applications in various medical image analysis tasks. However, deep learning methods applied to neuroimaging have not achieved superior performance in classifying ALS patients from healthy controls due to insignificant structural changes correlated with pathological features. Thus, a critical challenge in deep models is to identify discriminative features from limited training data. To address this challenge, this study introduces a framework called SF² Former, which leverages the power of the vision transformer architecture to distinguish ALS subjects from the control group by exploiting the long-range relationships among image features. Additionally, spatial and frequency domain information is combined to enhance the network’s performance, as MRI scans are initially captured in the frequency domain and then converted to the spatial domain. The proposed framework is trained using a series of consecutive coronal slices and utilizes pre-trained weights from ImageNet through transfer learning. Finally, a majority voting scheme is employed on the coronal slices of each subject to generate the final classification decision. The proposed architecture is extensively evaluated with multi-modal neuroimaging data (i.e., T1-weighted, R2*, FLAIR) using two well-organized versions of the Canadian ALS Neuroimaging Consortium (CALSNIC) multi-center datasets. The experimental results demonstrate the superiority of the proposed strategy in terms of classification accuracy compared to several popular deep learning-based techniques.

Access this article: https://doi.org/10.1016/j.compmedimag.2023.102279

Title: Alzheimer’s diseases diagnosis using fusion of high informative BiLSTM and CNN features of EEG signal
Authors: Maryam Imani
Type: Research Article
Abstract:
Electroencephalography (EEG) signals are low cost and available data for diagnosis of mental disorders such as Alzheimer’s diseases (AD). Each EEG signal contains information about electrical brain activities during the time. In addition to temporal features, different locations of brain show regional information in each time instance. Both of the mentioned features are used for AD diagnosis in this work. Bidirectional long short-term memory (BiLSTM) networks are used for analyzing the time sequences while the convolutional neural network (CNN) is used for exploration of relationship among EEG signals recorded by different channels located in different parts of the brain. The temporal and regional features are then fused through a fully connected neural network. Moreover, channel selection using entropy measure and data augmentation using autoencoder networks are implemented in this work to improve the diagnosis accuracy. The proposed framework is assessed in different cases and compared with several state-of-the-art methods in various experiments done by different EEG recording channels. The experiments show that the proposed method can reach to 100% accuracy for AD diagnosis using the studied EEG database.
Access this article: https://doi.org/10.1016/j.bspc.2023.105298

Title: Conv-Swinformer: Integration of CNN and shift window attention for Alzheimer’s disease classification
Authors: Zhentao Hu, Yanyang Li, Zheng Wang, Shuo Zhang, Wei Hou, Alzheimer’s Disease Neuroimaging Initiative
Type: Research Article
Abstract:
Deep learning (DL) algorithms based on brain MRI images have achieved great success in the prediction of Alzheimer’s disease (AD), with classification accuracy exceeding even that of the most experienced clinical experts. As a novel feature fusion method, Transformer has achieved excellent performance in many computer vision tasks, which also greatly promotes the application of Transformer in medical images. However, when Transformer is used for 3D MRI image feature fusion, existing DL models treat the input local features equally, which is inconsistent with the fact that adjacent voxels have stronger semantic connections than spatially distant voxels. In addition, due to the relatively small size of the dataset for medical images, it is difficult to capture local lesion features in limited iterative training by treating all input features equally. This paper proposes a deep learning model Conv-Swinformer that focuses on extracting and integrating local fine-grained features. Conv-Swinformer consists of a CNN module and a Transformer encoder module. The CNN module summarizes the planar features of the MRI slices, and the Transformer module establishes semantic connections in 3D space for these planar features. By introducing the shift window attention mechanism in the Transformer encoder, the attention is focused on a small spatial area of the MRI image, which effectively reduces unnecessary background semantic information and enables the model to capture local features more accurately. In addition, the layer-by-layer enlarged attention window can further integrate local fine-grained features, thus enhancing the model’s attention ability. Compared with DL algorithms that indiscriminately fuse local features of MRI images, Conv-Swinformer can fine-grained extract local lesion features, thus achieving better classification results.
Access this article: https://doi.org/10.1016/j.compbiomed.2023.107304

Title: Diffusion tensor imaging techniques show that parkin gene S/N167 polymorphism is responsible for extensive brain white matter damage in patients with Parkinson's disease
Authors: Jinqiu Yu, Jinying Shi, Lina Chen, Yingqing Wang, Guoen Cai, Xiaochun Chen, Weiming Hong, Qinyong Ye
Type: Research Article

Abstract:

Objective
To explore the influence of disease and genetic factors on the white matter microstructure in patients with PD. The white matter microstructural changes in the substantia nigra-striatum system were detected by diffusion tensor imaging (DTI) using the region of interest (ROI) and diffusion tensor tracer (DTT) methods.    

Methods
Patients with primary Parkinson's disease (PD) without a family history of PD were selected and divided into PD-G/G and PD-G/A groups according to their parkin S/N167 polymorphism. Control groups matched for age, sex, and gene type (G/G and G/A) were also included. Three-dimensional brain volume imaging (3D-BRAVO) and DTI were performed. The microstructural changes in the substantia nigra-striatum system were evaluated by the ROI and DTT methods. The Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Hoehn-Yahr (H–Y) staging, and the third part of the Unified Parkinson's Disease Rating (UPDRS-III) scales evaluated the cognitive and motor function impairment in patients with PD. Independent samples t-test compared normally-distributed data, and the Wilcoxon rank sum test compared measurement or categorical non-normally distributed data. Multiple regression analysis was used to analyze the correlation between various DTI indicators and the MMSE, MoCA, UPDRS-III, and H–Y scores in the PD-G/G and PD-G/A groups. P < 0.05 was considered statistically significant.

Results
The white matter microstructural changes in the nigrostriatal pathway differed significantly between the PD or PD-G/A and the control group (P < 0.05) The ROI method showed that the left globus pallidus radial diffusivity (RD) value was negatively correlated with the MMSE score (r = −0.404, P = 0.040), and the left substantia nigra (LSN) fractional anisotropy (FA) value was positively correlated with the MoCA score (r = 0.405, P = 0.040) and negatively with the H–Y stage (r = −0.479, P = 0.013).
The DTT method showed that the MMSE score was positively correlated with the right substantia nigra (RSN) FA value (r = 0.592, P = 0.001) and negatively with its RD value (r = −0.439, P = 0.025). The H–Y grade was negatively correlated with the number of fibers in the RSN (r = −0.406, P = 0.040). The UPDRS-Ⅲ score was positively correlated with the mean diffusivity (r = 0.420, P = 0.033) and RD (r = 0.396, P = 0.045) values of the LSN, and the AD value of the RSN (r = 0.439, P = 0.025).

Conclusion
The DTI technique detected extensive white matter fiber damage in patients with PD, primarily in those with the G/A genotype, that led to motor and cognitivesymptoms.

Access this article: https://doi.org/10.1016/j.heliyon.2023.e18395