BootGraph: Probabilistic fiber tractography using bootstrap algorithms and graph theory

Construction of brain structural connectivity network using a novel integrated algorithm based on ensemble average propagator

An important task for neuroscience is to accurately construct structural connectivity network of human brain. Tractography constructed based on high angular resolution diffusion imaging (HARDI) provides valuable information of human brain structural connections. Existing algorithms, mainly categorized as deterministic or probabilistic, come with inherent limitations (e.g., fiber direction uncertainty induced by noise, or anatomically unreasonable connections and heavy computational cost). In this study, a novel integrated algorithm was proposed to construct brain structural connectivity network by incorporating the deterministic path planning and probabilistic connection strength estimation, based on ensemble average propagator (EAP). We first estimated EAPs from multi-shell samples using the spherical polar Fourier imaging (SPFI), and then extracted diffusion orientations coinciding with neural fiber tracts. Only under angular constraints, the deterministic path planning algorithm was subsequently used to find all reasonable pathways between pairwise white matter (WM) voxels in different regions of interest (ROIs). Consequently, a train of consecutive WM voxels along each of the identified pathways was determined, and the connection strength of these pathways was computed by integrating their EAP alignment over a solid angle. The connection strength of a pair of WM voxels was assigned as the connection strength with the largest connection possibility. Finally, the connection strength between two ROIs was calculated as the sum of all the connection probabilities of each pair of WM voxels in the ROIs. A comparison against voxel-graph based probabilistic tractography method was performed on Fibercup phantom dataset, and the results demonstrated that the proposed method can produce better structural connection and is more computationally economical. Lastly, three datasets from Human Connectome Project (HCP) S1200 group were tested and their structural connectivity networks were constructed for topological analysis. The results showed great consistency in network metrics with previous WM network studies in healthy adults.

Tracking and validation techniques for topographically organized tractography

Topographic regularity of axonal connections is commonly understood as the preservation of spatial relationships between nearby neurons and is a fundamental structural property of the brain. In particular the retinotopic mapping of the visual pathway can even be quantitatively computed. Inspired from this previously untapped anatomical knowledge, we propose a novel tractography method that preserves both topographic and geometric regularity. We make use of parameterized curves with Frenet-Serret frame and introduce a highly flexible mechanism for controlling geometric regularity. At the same time, we incorporate a novel local data support term in order to account for topographic organization. Unifying geometry with topographic regularity, we develop a Bayesian framework for generating highly organized streamlines that accurately follow neuroanatomy. We additionally propose two novel validation techniques to quantify topographic regularity. In our experiments, we studied the results of our approach with respect to connectivity, reproducibility and topographic regularity aspects. We present both qualitative and quantitative comparisons of our technique against three algorithms from MRtrix3. We show that our method successfully generates highly organized fiber tracks while capturing bundle anatomy that are geometrically challenging for other approaches.

Fiber tractography using machine learning

We present a fiber tractography approach based on a random forest classification and voting process, guiding each step of the streamline progression by directly processing raw diffusion-weighted signal intensities. For comparison to the state-of-the-art, i.e. tractography pipelines that rely on mathematical modeling, we performed a quantitative and qualitative evaluation with multiple phantom and in vivo experiments, including a comparison to the 96 submissions of the ISMRM tractography challenge 2015. The results demonstrate the vast potential of machine learning for fiber tractography.

DTI Image Registration under Probabilistic Fiber Bundles Tractography Learning

Diffusion Tensor Imaging (DTI) image registration is an essential step for diffusion tensor image analysis. Most of the fiber bundle based registration algorithms use deterministic fiber tracking technique to get the white matter fiber bundles, which will be affected by the noise and volume. In order to overcome the above problem, we proposed a Diffusion Tensor Imaging image registration method under probabilistic fiber bundles tractography learning. Probabilistic tractography technique can more reasonably trace to the structure of the nerve fibers. The residual error estimation step in active sample selection learning is improved by modifying the residual error model using finite sample set. The calculated deformation field is then registered on the DTI images. The results of our proposed registration method are compared with 6 state-of-the-art DTI image registration methods under visualization and 3 quantitative evaluation standards. The experimental results show that our proposed method has a good comprehensive performance.

Training shortest-path tractography: Automatic learning of spatial priors

Tractography is the standard tool for automatic delineation of white matter tracts from diffusion weighted images. However, the output of tractography often requires post-processing to remove false positives and ensure a robust delineation of the studied tract, and this demands expert prior knowledge. Here we demonstrate how such prior knowledge, or indeed any prior spatial information, can be automatically incorporated into a shortest-path tractography approach to produce more robust results. We describe how such a prior can be automatically generated (learned) from a population, and we demonstrate that our framework also retains support for conventional interactive constraints such as waypoint regions. We apply our approach to the open access, high quality Human Connectome Project data, as well as a dataset acquired on a typical clinical scanner. Our results show that the use of a learned prior substantially increases the overlap of tractography output with a reference atlas on both populations, and this is confirmed by visual inspection. Furthermore, we demonstrate how a prior learned on the high quality dataset significantly increases the overlap with the reference for the more typical yet lower quality data acquired on a clinical scanner. We hope that such automatic incorporation of prior knowledge and the obviation of expert interactive tract delineation on every subject, will improve the feasibility of large clinical tractography studies.

Insight from uncertainty: bootstrap-derived diffusion metrics differentially predict memory function among older adults

Diffusion tensor imaging suffers from an intrinsic low signal-to-noise ratio. Bootstrap algorithms have been introduced to provide a non-parametric method to estimate the uncertainty of the measured diffusion parameters. To quantify the variability of the principal diffusion direction, bootstrap-derived metrics such as the cone of uncertainty have been proposed. However, bootstrap-derived metrics are not independent of the underlying diffusion profile. A higher mean diffusivity causes a smaller signal-to-noise ratio and, thus, increases the measurement uncertainty. Moreover, the goodness of the tensor model, which relies strongly on the complexity of the underlying diffusion profile, influences bootstrap-derived metrics as well. The presented simulations clearly depict the cone of uncertainty as a function of the underlying diffusion profile. Since the relationship of the cone of uncertainty and common diffusion parameters, such as the mean diffusivity and the fractional anisotropy, is not linear, the cone of uncertainty has a different sensitivity. In vivo analysis of the fornix reveals the cone of uncertainty to be a predictor of memory function among older adults. No significant correlation occurs with the common diffusion parameters. The present work not only demonstrates the cone of uncertainty as a function of the actual diffusion profile, but also discloses the cone of uncertainty as a sensitive predictor of memory function. Future studies should incorporate bootstrap-derived metrics to provide more comprehensive analysis.

Strengths and weaknesses of state of the art fiber tractography pipelines--A comprehensive in-vivo and phantom evaluation study using Tractometer

Many different tractography approaches and corresponding isolated evaluation attempts have been presented over the last years, but a comparative and quantitative evaluation of tractography algorithms still remains a challenge, particularly in-vivo. The recently presented evaluation framework Tractometer is the first attempt to approach this challenge in a quantitative, comparative, persistent and open-access way. Tractometer is currently based on the evaluation of several global connectivity and tract-overlap metrics on hardware phantom data. The work presented in this paper focuses on extending Tractometer with a metric that enables the assessment of the local consistency of tractograms with the underlying image data that is not only applicable to phantom dataset but allows the quantitative and purely data-driven evaluation of in-vivo tractography. We furthermore present an extensive reference-based evaluation study of 25,000 tractograms obtained on phantom and in-vivo datasets using the presented local metric as well as all the methods already established in Tractometer. The experiments showed that the presented local metric successfully reflects the behavior of in-vivo tractography under different conditions and that it is consistent with the results of previous studies. Additionally our experiments enabled a multitude of conclusions with implications for fiber tractography in general, including recommendations regarding optimal choice of a local modeling technique, tractography algorithm, and parameterization, confirming and complementing the results of earlier studies.