Mouse brain MR super-resolution using a deep learning network trained with optical imaging data.

Introduction: The resolution of magnetic resonance imaging is often limited at the millimeter level due to its inherent signal-to-noise disadvantage compared to other imaging modalities. Super-resolution (SR) of MRI data aims to enhance its resolution and diagnostic value. While deep learning-based SR has shown potential, its applications in MRI remain limited, especially for preclinical MRI, where large high-resolution MRI datasets for training are often lacking. Methods: In this study, we first used high-resolution mouse brain auto-fluorescence (AF) data acquired using serial two-photon tomography (STPT) to examine the performance of deep learning-based SR for mouse brain images. Results: We found that the best SR performance was obtained when the resolutions of training and target data were matched. We then applied the network trained using AF data to MRI data of the mouse brain, and found that the performance of the SR network depended on the tissue contrast presented in the MRI data. Using transfer learning and a limited set of high-resolution mouse brain MRI data, we were able to fine-tune the initial network trained using AF to enhance the resolution of MRI data. Discussion: Our results suggest that deep learning SR networks trained using high-resolution data of a different modality can be applied to MRI data after transfer learning.

Towards reliable reconstruction of the mouse brain corticothalamic connectivity using diffusion MRI.

Diffusion magnetic resonance imaging (dMRI) tractography has yielded intriguing insights into brain circuits and their relationship to behavior in response to gene mutations or neurological diseases across a number of species. Still, existing tractography approaches suffer from limited sensitivity and specificity, leading to uncertain interpretation of the reconstructed connections. Hence, in this study, we aimed to optimize the imaging and computational pipeline to achieve the best possible spatial overlaps between the tractography and tracer-based axonal projection maps within the mouse brain corticothalamic network. We developed a dMRI-based atlas of the mouse forebrain with structural labels imported from the Allen Mouse Brain Atlas (AMBA). Using the atlas and dMRI tractography, we first reconstructed detailed node-to-node mouse brain corticothalamic structural connectivity matrices using different imaging and tractography parameters. We then investigated the effects of each condition for accurate reconstruction of the corticothalamic projections by quantifying the similarities between the tractography and the tracer data from the Allen Mouse Brain Connectivity Atlas (AMBCA). Our results suggest that these parameters significantly affect tractography outcomes and our atlas can be used to investigate macroscopic structural connectivity in the mouse brain. Furthermore, tractography in mouse brain gray matter still face challenges and need improved imaging and tractography methods.

Using mesoscopic tract-tracing data to guide the estimation of fiber orientation distributions in the mouse brain from diffusion MRI.

Diffusion MRI (dMRI) tractography is the only tool for non-invasive mapping of macroscopic structural connectivity over the entire brain. Although it has been successfully used to reconstruct large white matter tracts in the human and animal brains, the sensitivity and specificity of dMRI tractography remained limited. In particular, the fiber orientation distributions (FODs) estimated from dMRI signals, key to tractography, may deviate from histologically measured fiber orientation in crossing fibers and gray matter regions. In this study, we demonstrated that a deep learning network, trained using mesoscopic tract-tracing data from the Allen Mouse Brain Connectivity Atlas, was able to improve the estimation of FODs from mouse brain dMRI data. Tractography results based on the network generated FODs showed improved specificity while maintaining sensitivity comparable to results based on FOD estimated using a conventional spherical deconvolution method. Our result is a proof-of-concept of how mesoscale tract-tracing data can guide dMRI tractography and enhance our ability to characterize brain connectivity.

The effect of long-term methadone maintenance treatment on coupling among three large-scale brain networks in male heroin-dependent individuals: A resting-state fMRI study.

PURPOSE: Methadone maintenance treatment (MMT) is considered as an effective and mainstream therapy for heroin dependence. However, whether long-term MMT would improve the coupling among the three core large-scale brain networks (salience, default mode, and executive control) and its relationship with the craving for heroin is unknown. METHODS: Forty-four male heroin-dependent individuals during long-term MMT, 27 male heroin-dependent individuals after short-term detoxification/abstinence (SA), and 26 demographically matched healthy controls (HC) underwent resting-state functional magnetic resonance imaging. We analyzed the difference in coupling among the salience, default mode, and executive control networks among the three groups and examined how the coupling among these large-scale networks was associated with craving before and after drug-cue exposure. RESULTS: Compared with the SA group, the MMT group showed lower craving before and after cue exposure and stronger connectivity between the dorsal anterior cingulate cortex (a key node of the salience network) and key regions of the bilateral executive control network, including the bilateral dorsolateral prefrontal cortex, posterior parietal cortex, and dorsomedial prefrontal cortex. Among the heroin-dependent individuals, the functional connectivity was negatively correlated with the craving before and after heroin-cue exposure. CONCLUSION: Our findings suggest that long-term MMT could increase the coupling between the salience and bilateral executive control networks and decrease craving for heroin. These findings contribute to the understanding of the neural mechanism of MMT, from the perspective of large-scale brain networks.

Virtual mouse brain histology from multi-contrast MRI via deep learning.

1H MRI maps brain structure and function non-invasively through versatile contrasts that exploit inhomogeneity in tissue micro-environments. Inferring histopathological information from magnetic resonance imaging (MRI) findings, however, remains challenging due to absence of direct links between MRI signals and cellular structures. Here, we show that deep convolutional neural networks, developed using co-registered multi-contrast MRI and histological data of the mouse brain, can estimate histological staining intensity directly from MRI signals at each voxel. The results provide three-dimensional maps of axons and myelin with tissue contrasts that closely mimic target histology and enhanced sensitivity and specificity compared to conventional MRI markers. Furthermore, the relative contribution of each MRI contrast within the networks can be used to optimize multi-contrast MRI acquisition. We anticipate our method to be a starting point for translation of MRI results into easy-to-understand virtual histology for neurobiologists and provide resources for validating novel MRI techniques.

SuperDTI: Ultrafast DTI and fiber tractography with deep learning.

PURPOSE: To develop a deep learning-based reconstruction framework for ultrafast and robust diffusion tensor imaging and fiber tractography. METHODS: SuperDTI was developed to learn the nonlinear relationship between DWIs and the corresponding diffusion tensor parameter maps. It bypasses the tensor fitting procedure, which is highly susceptible to noises and motions in DWIs. The network was trained and tested using data sets from the Human Connectome Project and patients with ischemic stroke. Results from SuperDTI were compared against widely used methods for tensor parameter estimation and fiber tracking. RESULTS: Using training and testing data acquired using the same protocol and scanner, SuperDTI was shown to generate fractional anisotropy and mean diffusivity maps, as well as fiber tractography, from as few as six raw DWIs, with a quantification error of less than 5% in all white-matter and gray-matter regions of interest. It was robust to noises and motions in the testing data. Furthermore, the network trained using healthy volunteer data showed no apparent reduction in lesion detectability when directly applied to stroke patient data. CONCLUSIONS: Our results demonstrate the feasibility of superfast DTI and fiber tractography using deep learning with as few as six DWIs directly, bypassing tensor fitting. Such a significant reduction in scan time may allow the inclusion of DTI into the clinical routine for many potential applications.

Gray matter alteration in heroin-dependent men: An atlas-based magnetic resonance imaging study.

Previous imaging studies on heroin addiction have reported brain morphological alterations. However, the effects of heroin exposure on gray matter volume varied among different studies due to different factors such as substitution treatment or mandatory abstinence. Meanwhile, the relationship between gray matter and heroin use history remains unknown. Thirty-three male heroin-dependent (HD) individuals who are not under any substitution treatment or mandatory abstinence and 40 male healthy controls (HC) were included in this structural magnetic resonance imaging study. With an atlas-based approach, gray matter structures up to individual functional area were delineated, and the differences in their volumes between the HD and HC groups were analyzed. In addition, the relationship between gray matter volume and duration of heroin use was explored. The HD group demonstrated significantly lower cortical volume mainly in the prefrontal cortex and mesolimbic dopaminergic regions across different parcellation levels, whereas several visual and somatosensory cortical regions in the HD group had greater volume relative to the HC group at a more detailed parcellation level. The duration of heroin use was negatively correlated with the gray matter volume of prefrontal cortex. These findings suggest that heroin addiction be related to gray matter alteration and might be related to damage/maladaption of the inhibitory control, reward, visual, and somatosensory functions of the brain, although cognitive correlates are warranted in future study. In addition, the atlas-based morphology analysis is a potential tool to help researchers search biomarkers of heroin addiction.

Brain MRI Pattern Recognition Translated to Clinical Scenarios.

We explored the performance of structure-based computational analysis in four neurodegenerative conditions [Ataxia (AT, n = 16), Huntington's Disease (HD, n = 52), Alzheimer's Disease (AD, n = 66), and Primary Progressive Aphasia (PPA, n = 50)], all characterized by brain atrophy. The independent variables were the volumes of 283 anatomical areas, derived from automated segmentation of T1-high resolution brain MRIs. The segmentation based volumetric quantification reduces image dimensionality from the voxel level [on the order of [Formula: see text](106)] to anatomical structures [[Formula: see text](102)] for subsequent statistical analysis. We evaluated the effectiveness of this approach on extracting anatomical features, already described by human experience and a priori biological knowledge, in specific scenarios: (1) when pathologies were relatively homogeneous, with evident image alterations (e.g., AT); (2) when the time course was highly correlated with the anatomical changes (e.g., HD), an analogy for prediction; (3) when the pathology embraced heterogeneous phenotypes (e.g., AD) so the classification was less efficient but, in compensation, anatomical and clinical information were less redundant; and (4) when the entity was composed of multiple subgroups that had some degree of anatomical representation (e.g., PPA), showing the potential of this method for the clustering of more homogeneous phenotypes that can be of clinical importance. Using the structure-based quantification and simple linear classifiers (partial least square), we achieve 87.5 and 73% of accuracy on differentiating AT and pre-symptomatic HD patents from controls, respectively. More importantly, the anatomical features automatically revealed by the classifiers agreed with the patterns previously described on these pathologies. The accuracy was lower (68%) on differentiating AD from controls, as AD does not display a clear anatomical phenotype. On the other hand, the method identified PPA clinical phenotypes and their respective anatomical signatures. Although most of the data are presented here as proof of concept in simulated clinical scenarios, structure-based analysis was potentially effective in characterizing phenotypes, retrieving relevant anatomical features, predicting prognosis, and aiding diagnosis, with the advantage of being easily translatable to clinics and understandable biologically.

Evaluation of Cross-Protocol Stability of a Fully Automated Brain Multi-Atlas Parcellation Tool.

Brain parcellation tools based on multiple-atlas algorithms have recently emerged as a promising method with which to accurately define brain structures. When dealing with data from various sources, it is crucial that these tools are robust for many different imaging protocols. In this study, we tested the robustness of a multiple-atlas, likelihood fusion algorithm using Alzheimer's Disease Neuroimaging Initiative (ADNI) data with six different protocols, comprising three manufacturers and two magnetic field strengths. The entire brain was parceled into five different levels of granularity. In each level, which defines a set of brain structures, ranging from eight to 286 regions, we evaluated the variability of brain volumes related to the protocol, age, and diagnosis (healthy or Alzheimer's disease). Our results indicated that, with proper pre-processing steps, the impact of different protocols is minor compared to biological effects, such as age and pathology. A precise knowledge of the sources of data variation enables sufficient statistical power and ensures the reliability of an anatomical analysis when using this automated brain parcellation tool on datasets from various imaging protocols, such as clinical databases.