Resting state functional connectivity MRI (rsfc-MRI) reveals a wealth of information about the functional organization of the brain, but poses unique challenges for quantitative image analysis, mostly related to the large number of voxels with low signal-to-noise ratios. In this study, we tested the idea of using a prior spatial parcellation of the entire brain into various structural units, to perform an analysis on a structure-by-structure, rather than voxel-by-voxel, basis. This analysis, based upon atlas parcels, potentially offers enhanced SNR and reproducibility, and can be used as a common anatomical framework for cross-modality and cross-subject quantitative analysis. We used Large Deformation Diffeomorphic Metric Mapping (LDDMM) and a deformable brain atlas to parcel each brain into 185 regions. To investigate the precision of the cross-subject analysis, we computed inter-parcel correlations in 20 participants, each of whom were scanned twice, as well as the consistency of the connectivity patterns inter- and intra-subject, and the intersession reproducibility. We report significant inter-parcel correlations consistent with previous findings, and high test-retest reliability, an important consideration when the goal is to compare clinical populations. As an example of the cross-modality analysis, correlation with anatomical connectivity is also examined.

Atlas-based image analysis (ABA), in which an anatomical “parcellation map” is used for parcel-by-parcel image quantification, is widely used to analyze anatomical and functional changes related to brain development, aging, and various diseases. The parcellation maps are often created based on common MRI templates, which allow users to transform the template to target images, or vice versa, to perform parcel-by-parcel statistics, and report the scientific findings based on common anatomical parcels. The use of a study-specific template, which represents the anatomical features of the study population better than common templates, is preferable for accurate anatomical labeling; however, the creation of a parcellation map for a study-specific template is extremely labor intensive, and the definitions of anatomical boundaries are not necessarily compatible with those of the common template. In this study, we employed a Volume-based Template Estimation (VTE) method to create a neonatal brain template customized to a study population, while keeping the anatomical parcellation identical to that of a common MRI atlas. The VTE was used to morph the standardized parcellation map of the JHU-neonate-SS atlas to capture the anatomical features of a study population. The resultant “study-customized” T1-weighted and diffusion tensor imaging (DTI) template, with three-dimensional anatomical parcellation that defined 122 brain regions, was compared with the JHU-neonate-SS atlas, in terms of the registration accuracy. A pronounced increase in the accuracy of cortical parcellation and superior tensor alignment were observed when the customized template was used. With the customized atlas-based analysis, the fractional anisotropy (FA) detected closely approximated the manual measurements. This tool provides a solution for achieving normalization-based measurements with increased accuracy, while reporting scientific findings in a consistent framework.

Purpose: White matter hyperintense (WMHI) lesions on MR images are an important indication of various types of brain diseases that involve inflammation and blood vessel abnormalities.Automated quantification of the WMHI could be valuable for clinical management of the patients, but existing automated software is often developed for a single type of disease and may not be applicable for clinical scans with thick slices and different scanning protocols. Methods:We developed and evaluated "DeepWML", a U-net method for fully automated white matter lesion (WML) segmentation of multi-center FLAIR images. We used MRI from 507 patients, including in three distinct WM diseases, obtained in 9 centers, with a wide range of scanners and acquisition protocols. The automated delineation tool was evaluated through quantitative parameters of dice similarity, sensitivity and precision as compared to manual delineation (gold standard). Results:The overall median dice similarity coefficient was 0.78 (range 0.64~0.86) across the three disease types and multiple centers. The median sensitivity and precision was 0.84 (range 0.67~0.94) and 0.81 (range 0.64~0.92), respectively. The tool's performance increased with larger lesion volumes. Conclusion:DeepWML was successfully applied to a wide spectrum of MRI data in the three white matter disease types, which is potential to improve practical workflow of white matter lesion delineation.

Modern MRI systems usually load the predesigned RFs and the accompanying gradients during clinical scans, with minimal adaption to the specific requirements of each scan. Here, we describe a neural network-based method for real-time design of excitation RF pulses and the accompanying gradients' waveforms to achieve spatially two-dimensional selectivity. Nine thousand sets of radio frequency (RF) and gradient waveforms with two-dimensional spatial selectivity were generated as the training dataset using the Shinnar-Le Roux (SLR) method. Neural networks were created and trained with five strategies (TS-1 to TS-5). The neural network-designed RF and gradients were compared with their SLR-designed counterparts and underwent Bloch simulation and phantom imaging to investigate their performances in spin manipulations. We demonstrate a convolutional neural network (TS-5) with multi-task learning to yield both the RF pulses and the accompanying two channels of gradient waveforms that comply with the SLR design, and these design results also provide excitation spatial profiles comparable with SLR pulses in both simulation (normalized root mean square error [NRMSE] of 0.0075 ± 0.0038 over the 400 sets of testing data between TS-5 and SLR) and phantom imaging. The output RF and gradient waveforms between the neural network and SLR methods were also compared, and the joint NRMSE, with both RF and the two channels of gradient waveforms considered, was 0.0098 ± 0.0024 between TS-5 and SLR. The RF and gradients were generated on a commercially available workstation, which took 130 ms for TS-5. In conclusion, we present a convolutional neural network with multi-task learning, trained with SLR transformation pairs, that is capable of simultaneously generating RF and two channels of gradient waveforms, given the desired spatially two-dimensional excitation profiles.

To reduce the exposure of Gadolinium-based Contrast Agents (GBCAs) in brainstem glioma detection and provide high-resolution contrast information, we propose a novel multi-task generative network for contrast-enhanced T1-weight MR synthesis on brainstem glioma images. The proposed network can simultaneously synthesize the high-resolution contrast-enhanced image and the segmentation mask of brainstem glioma lesions.