Developing an AI-empowered head-only ultra-high-performance gradient MRI system for high spatiotemporal neuroimaging.

Recent advances in neuroscience requires high-resolution MRI to decipher the structural and functional details of the brain. Developing a high-performance gradient system is an ongoing effort in the field to facilitate high spatial and temporal encoding. Here, we proposed a head-only gradient system NeuroFrontier, dedicated for neuroimaging with an ultra-high gradient strength of 650 mT/m and 600 T/m/s. The proposed system features in 1) ultra-high power of 7MW achieved by running two gradient power amplifiers using a novel paralleling method; 2) a force/torque balanced gradient coil design with a two-step mechanical structure that allows high-efficiency and flexible optimization of the peripheral nerve stimulation; 3) a high-density integrated RF system that is miniaturized and customized for the head-only system; 4) an AI-empowered compressed sensing technique that enables ultra-fast acquisition of high-resolution images and AI-based acceleration in q-t space for diffusion MRI (dMRI); and 5) a prospective head motion correction technique that effectively corrects motion artifacts in real-time with 3D optical tracking. We demonstrated the potential advantages of the proposed system in imaging resolution, speed, and signal-to-noise ratio for 3D structural MRI (sMRI), functional MRI (fMRI) and dMRI in neuroscience applications of submillimeter layer-specific fMRI and dMRI. We also illustrated the unique strength of this system for dMRI-based microstructural mapping, e.g., enhanced lesion contrast at short diffusion-times or high b-values, and improved estimation accuracy for cellular microstructures using diffusion-time-dependent dMRI or for neurite microstructures using q-space approaches.

Validation of deep learning techniques for quality augmentation in diffusion MRI for clinical studies.

The objective of this study is to evaluate the efficacy of deep learning (DL) techniques in improving the quality of diffusion MRI (dMRI) data in clinical applications. The study aims to determine whether the use of artificial intelligence (AI) methods in medical images may result in the loss of critical clinical information and/or the appearance of false information. To assess this, the focus was on the angular resolution of dMRI and a clinical trial was conducted on migraine, specifically between episodic and chronic migraine patients. The number of gradient directions had an impact on white matter analysis results, with statistically significant differences between groups being drastically reduced when using 21 gradient directions instead of the original 61. Fourteen teams from different institutions were tasked to use DL to enhance three diffusion metrics (FA, AD and MD) calculated from data acquired with 21 gradient directions and a b-value of 1000 s/mm2. The goal was to produce results that were comparable to those calculated from 61 gradient directions. The results were evaluated using both standard image quality metrics and Tract-Based Spatial Statistics (TBSS) to compare episodic and chronic migraine patients. The study results suggest that while most DL techniques improved the ability to detect statistical differences between groups, they also led to an increase in false positive. The results showed that there was a constant growth rate of false positives linearly proportional to the new true positives, which highlights the risk of generalization of AI-based tasks when assessing diverse clinical cohorts and training using data from a single group. The methods also showed divergent performance when replicating the original distribution of the data and some exhibited significant bias. In conclusion, extreme caution should be exercised when using AI methods for harmonization or synthesis in clinical studies when processing heterogeneous data in clinical studies, as important information may be altered, even when global metrics such as structural similarity or peak signal-to-noise ratio appear to suggest otherwise.

Multi-modal multi-resolution atlas of the human neonatal cerebral cortex based on microstructural similarity.

The neonatal period is a critical window for the development of the human brain and may hold implications for the long-term development of cognition and disorders. Multi-modal connectome studies have revealed many important findings underlying the adult brain but related studies were rare in the early human brain. One potential challenge is the lack of an appropriate and unbiased parcellation that combines structural and functional information in this population. Using 348 multi-modal MRI datasets from the developing human connectome project, we found that the information fused from the structural, diffusion, and functional MRI was relatively stable across MRI features and showed high reproducibility at the group level. Therefore, we generated automated multi-resolution parcellations (300 - 500 parcels) based on the similarity across multi-modal features using a gradient-based parcellation algorithm. In addition, to acquire a parcellation with high interpretability, we provided a manually delineated parcellation (210 parcels), which was approximately symmetric, and the adjacent areas around each boundary were statistically different in terms of the integrated similarity metric and at least one kind of original features. Overall, the present study provided multi-resolution and neonate-specific parcellations of the cerebral cortex based on multi-modal MRI properties, which may facilitate future studies of the human connectome in the early development period.

Microstructural Alterations in Projection and Association Fibers in Neonatal Hypoxia-Ischemia.

BACKGROUND: Diffusion MRI (dMRI) is known to be sensitive to hypoxic-ischemic encephalopathy (HIE). However, existing dMRI studies used simple diffusion tensor metrics and focused only on a few selected cerebral regions, which cannot provide a comprehensive picture of microstructural injury. PURPOSE: To systematically characterize the microstructural alterations in mild, moderate, and severe HIE neonates compared to healthy neonates with advanced dMRI using region of interest (ROI), tract, and fixel-based analyses. STUDY TYPE: Prospective. POPULATION: A total of 42 neonates (24 males and 18 females). FIELD STRENGTH/SEQUENCE: 3-T, diffusion-weighted echo-planar imaging. ASSESSMENT: Fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD), axial diffusivity (AD), fiber density (FD), fiber cross-section (FC), and fiber density and cross-section (FDC) were calculated in 40 ROIs and 6 tracts. Fixel-based analysis was performed to assess group differences in individual fiber components within a voxel (fixel). STATISTICAL TESTS: One-way analysis of covariance (ANCOVA) to compare dMRI metrics among severe/moderate/mild HIE and control groups and general linear model for fixel-wise group differences (age, sex, and body weight as covariates). Adjusted P value < 0.05 was considered statistically significant. RESULTS: For severe HIE, ROI-based analysis revealed widespread regions, including the deep nuclei and white matter with reduced FA, while in moderate injury, only FC was decreased around the posterior watershed zones. Tract-based analysis demonstrated significantly reduced FA, FD, and FC in the right inferior fronto-occipital fasciculus (IFOF), right inferior longitudinal fasciculus (ILF), and splenium of corpus callosum (SCC) in moderate HIE, and in right IFOF and left anterior thalamic radiation (ATR) in mild HIE. Correspondingly, we found altered fixels in the right middle-posterior IFOF and ILF, and in the central-to-right part of SCC in moderate HIE. DATA CONCLUSION: For severe HIE, extensive microstructural injury was identified. For moderate-mild HIE, association fiber injury in posterior watershed area with a rightward lateralization was found. EVIDENCE LEVEL: 1 TECHNICAL EFFICACY: Stage 3.

Layer-specific microstructural patterns of anterior hippocampus in Alzheimer's disease with ex vivo diffusion MRI at 14.1 T.

High-resolution ex vivo diffusion MRI (dMRI) can provide exquisite mesoscopic details and microstructural information of the human brain. Microstructural pattern of the anterior part of human hippocampus, however, has not been well elucidated with ex vivo dMRI, either in normal or disease conditions. The present study collected high-resolution (0.1 mm isotropic) dMRI of post-mortem anterior hippocampal tissues from four Alzheimer's diseases (AD), three primary age-related tauopathy (PART), and three healthy control (HC) brains on a 14.1 T spectrometer. We evaluated how AD affected dMRI-based microstructural features in different layers and subfields of anterior hippocampus. In the HC samples, we found higher anisotropy, lower diffusivity, and more streamlines in the layers within cornu ammonis (CA) than those within dentate gyrus (DG). Comparisons between disease groups showed that (1) anisotropy measurements in the CA layers of AD, especially stratum lacunosum (SL) and stratum radiatum (SR), had higher regional variability than the other two groups; (2) streamline density in the DG layers showed a gradually increased variance from HC to PART to AD; (3) AD also showed the higher variability in terms of inter-layer connectivity than HC or PART. Moreover, voxelwise correlation analysis between the coregistered dMRI and histopathology images revealed significant correlations between dMRI measurements and the contents of amyloid beta (Aß)/tau protein in specific layers of AD samples. These findings may reflect layer-specific microstructural characteristics in different hippocampal subfields at the mesoscopic resolution, which were associated with protein deposition in the anterior hippocampus of AD patients.

A diffusion MRI-based spatiotemporal continuum of the embryonic mouse brain for probing gene-neuroanatomy connections.

The embryonic mouse brain undergoes drastic changes in establishing basic anatomical compartments and laying out major axonal connections of the developing brain. Correlating anatomical changes with gene-expression patterns is an essential step toward understanding the mechanisms regulating brain development. Traditionally, this is done in a cross-sectional manner, but the dynamic nature of development calls for probing gene-neuroanatomy interactions in a combined spatiotemporal domain. Here, we present a four-dimensional (4D) spatiotemporal continuum of the embryonic mouse brain from E10.5 to E15.5 reconstructed from diffusion magnetic resonance microscopy (dMRM) data. This study achieved unprecedented high-definition dMRM at 30- to 35-µm isotropic resolution, and together with computational neuroanatomy techniques, we revealed both morphological and microscopic changes in the developing brain. We transformed selected gene-expression data to this continuum and correlated them with the dMRM-based neuroanatomical changes in embryonic brains. Within the continuum, we identified distinct developmental modes comprising regional clusters that shared developmental trajectories and similar gene-expression profiles. Our results demonstrate how this 4D continuum can be used to examine spatiotemporal gene-neuroanatomical interactions by connecting upstream genetic events with anatomical changes that emerge later in development. This approach would be useful for large-scale analysis of the cooperative roles of key genes in shaping the developing brain.

Extracting and Selecting Robust Radiomic Features from PET/MR Images in Nasopharyngeal Carcinoma.

OBJECTIVES: This work aims to study the variation, robustness, and feature redundancy of PET/MR radiomic features in the primary tumor of nasopharyngeal carcinoma (NPC). PROCEDURES: PET/MR scans of 21 NPC patients were used in this study. The primary tumor volumes were defined using PET, T2-weighted-MR (T2-MR), and diffusion-weighted MR (DW-MR) images. A random-dilation-erosion method was used to simulate 10 sets of tumor volumes for identifying features invariant with manual segmentation uncertainties. Feature robustness was evaluated against imaging modalities, pixel sizes, slice thickness, and grey-level bin sizes using intraclass correlation coefficient (ICC) and spearman correlation coefficient. Feature redundancy was analyzed using the hierarchical cluster analysis. RESULTS: Voxel size of 0.5 × 0.5 × 1.0 mm3 was found optimal for robust feature extraction from PET and MR. Normalized grey level of 64 and 128 was suggested for PET and MR, respectively. The features from wavelet-transformed images were less stable than those from the original images. The robustness analysis and volume correlation analysis identified 335 (62.04 %) PET features, 240 (44.44 %) T2-MR features, and 366 (67.78 %) DW-MR features. The cluster analysis grouped PET, T2-MR, and DW-MR features into 106, 83, and 133 representative features, respectively. CONCLUSIONS: The present study analyzed and identified robust features extracted from tumor volumes on PET/MR, which can provide guidance and promote standardization for PET/MR radiomic studies in NPC.

A multi-organ cancer study of the classification performance using 2D and 3D image features in radiomics analysis.

The purpose of this study was to investigate the predictive performance of 2D and 3D image features across multi-organ cancers using multi-modality images in radiomics studies. In this retrospective study, we included 619 patients with three different cancer types (intrahepatic cholangiocarcinoma (ICC), high-grade osteosarcoma (HOS), pancreatic neuroendocrine tumors (pNETs)) and four clinical end points (early recurrence (ER), lymph node metastasis (LNM), 5-year survival and histologic grade). The image features included fifty-eight 2D image features and fifty-eight 3D image features. The 3D image features were extracted based on the 3D tumor volumes. The 2D image features were extracted based on 2D tumor region, which was the layer with the maximum tumor diameter within the 3D tumor volume. The predictive performance of individual 2D and 3D image feature was measured using the area under the receiver operating characteristic curve (AUC) with univariate analysis. Radiomics signatures were further developed using multivariable analysis with 4-fold cross-validation method. Using univariate analysis, we found that more 3D image features showed the statistically predictive capabilities than 2D image features across all the included cancer types. By comparing the predictive performance of radiomics signatures developed by 2D and 3D image features, we observed better prediction performance in radiomics signatures based on 3D image features than those based on 2D image features for patients with ICC and HGO. Meanwhile, the signatures based on 2D and 3D image features performed closely in the pNETs dataset with the clinical end point of the histologic grade. The reason for this inconsistent result might be that the gross tumor volumes of pNETs were generally small. The tumor heterogeneity was mostly presented in the middle several layers within the tumor volume. Both 2D and 3D image features have certain predictive capacities. By contrast, the 3D image features show better or close predictive performance than 2D image features using both univariate analysis and multivariate analysis. In brief, 3D image features are recommended in radiomics studies.