Fusion of biomedical imaging studies for increased sample size and diversity: a case study of brain MRI.

Data collection, curation, and cleaning constitute a crucial phase in Machine Learning (ML) projects. In biomedical ML, it is often desirable to leverage multiple datasets to increase sample size and diversity, but this poses unique challenges, which arise from heterogeneity in study design, data descriptors, file system organization, and metadata. In this study, we present an approach to the integration of multiple brain MRI datasets with a focus on homogenization of their organization and preprocessing for ML. We use our own fusion example (approximately 84,000 images from 54,000 subjects, 12 studies, and 88 individual scanners) to illustrate and discuss the issues faced by study fusion efforts, and we examine key decisions necessary during dataset homogenization, presenting in detail a database structure flexible enough to accommodate multiple observational MRI datasets. We believe our approach can provide a basis for future similarly-minded biomedical ML projects.

Multi-scale V-net architecture with deep feature CRF layers for brain extraction.

BACKGROUND: Brain extraction is a computational necessity for researchers using brain imaging data. However, the complex structure of the interfaces between the brain, meninges and human skull have not allowed a highly robust solution to emerge. While previous methods have used machine learning with structural and geometric priors in mind, with the development of Deep Learning (DL), there has been an increase in Neural Network based methods. Most proposed DL models focus on improving the training data despite the clear gap between groups in the amount and quality of accessible training data between. METHODS: We propose an architecture we call Efficient V-net with Additional Conditional Random Field Layers (EVAC+). EVAC+ has 3 major characteristics: (1) a smart augmentation strategy that improves training efficiency, (2) a unique way of using a Conditional Random Fields Recurrent Layer that improves accuracy and (3) an additional loss function that fine-tunes the segmentation output. We compare our model to state-of-the-art non-DL and DL methods. RESULTS: Results show that even with limited training resources, EVAC+ outperforms in most cases, achieving a high and stable Dice Coefficient and Jaccard Index along with a desirable lower Surface (Hausdorff) Distance. More importantly, our approach accurately segmented clinical and pediatric data, despite the fact that the training dataset only contains healthy adults. CONCLUSIONS: Ultimately, our model provides a reliable way of accurately reducing segmentation errors in complex multi-tissue interfacing areas of the brain. We expect our method, which is publicly available and open-source, to be beneficial to a wide range of researchers.

Computational processing of brain images can enable better understanding and diagnosis of diseases that affect the brain. Brain Extraction is a computational method that can be used to remove areas of the head that are not the brain from images of the head. We compared various different computational methods that are available and used them to develop a better method. The method we describe in the paper is more accurate at imaging the brain of both healthy individuals and those known to have diseases that affect the brain than the other methods we evaluated. Our method might enable better understanding and diagnosis of diseases that affect the brain in the future.

Denoising of diffusion MRI in the cervical spinal cord - effects of denoising strategy and acquisition on intra-cord contrast, signal modeling, and feature conspicuity.

Quantitative diffusion MRI (dMRI) is a promising technique for evaluating the spinal cord in health and disease. However, low signal-to-noise ratio (SNR) can impede interpretation and quantification of these images. The purpose of this study is to evaluate several dMRI denoising approaches on their ability to improve the quality, reliability, and accuracy of quantitative diffusion MRI of the spinal cord. We evaluate three denoising approaches (Non-Local Means, Marchenko-Pastur PCA, and a newly proposed Patch2Self algorithm) and conduct five experiments to validate the denoising performance on clinical-quality and commonly-acquired dMRI acquisitions: 1) a phantom experiment to assess denoising error and bias; 2) a multi-vendor, multi-acquisition open experiment for both qualitative and quantitative evaluation of noise residuals; 3) a bootstrapping experiment to estimate uncertainty of parametric maps; 4) an assessment of spinal cord lesion conspicuity in a multiple sclerosis group; and 5) an evaluation of denoising for advanced parametric multi-compartment modeling. We find that all methods improve signal-to-noise ratio and conspicuity of MS lesions in individual diffusion weighted images (DWIs), but MPPCA and Patch2Self excel at improving the quality and intra-cord contrast of diffusion weighted images - removing signal fluctuations due to thermal noise while improving precision of estimation of diffusion parameters even with very few DWIs (i.e., 16-32) typical of clinical acquisitions. These denoising approaches hold promise for facilitating reliable diffusion observations and measurements in the spinal cord to investigate biological and pathological processes.

On the generalizability of diffusion MRI signal representations across acquisition parameters, sequences and tissue types: Chronicles of the MEMENTO challenge.

Diffusion MRI (dMRI) has become an invaluable tool to assess the microstructural organization of brain tissue. Depending on the specific acquisition settings, the dMRI signal encodes specific properties of the underlying diffusion process. In the last two decades, several signal representations have been proposed to fit the dMRI signal and decode such properties. Most methods, however, are tested and developed on a limited amount of data, and their applicability to other acquisition schemes remains unknown. With this work, we aimed to shed light on the generalizability of existing dMRI signal representations to different diffusion encoding parameters and brain tissue types. To this end, we organized a community challenge - named MEMENTO, making available the same datasets for fair comparisons across algorithms and techniques. We considered two state-of-the-art diffusion datasets, including single-diffusion-encoding (SDE) spin-echo data from a human brain with over 3820 unique diffusion weightings (the MASSIVE dataset), and double (oscillating) diffusion encoding data (DDE/DODE) of a mouse brain including over 2520 unique data points. A subset of the data sampled in 5 different voxels was openly distributed, and the challenge participants were asked to predict the remaining part of the data. After one year, eight participant teams submitted a total of 80 signal fits. For each submission, we evaluated the mean squared error, the variance of the prediction error and the Bayesian information criteria. The received submissions predicted either multi-shell SDE data (37%) or DODE data (22%), followed by cartesian SDE data (19%) and DDE (18%). Most submissions predicted the signals measured with SDE remarkably well, with the exception of low and very strong diffusion weightings. The prediction of DDE and DODE data seemed more challenging, likely because none of the submissions explicitly accounted for diffusion time and frequency. Next to the choice of the model, decisions on fit procedure and hyperparameters play a major role in the prediction performance, highlighting the importance of optimizing and reporting such choices. This work is a community effort to highlight strength and limitations of the field at representing dMRI acquired with trending encoding schemes, gaining insights into how different models generalize to different tissue types and fiber configurations over a large range of diffusion encodings.

QSIPrep: an integrative platform for preprocessing and reconstructing diffusion MRI data.

Diffusion-weighted magnetic resonance imaging (dMRI) is the primary method for noninvasively studying the organization of white matter in the human brain. Here we introduce QSIPrep, an integrative software platform for the processing of diffusion images that is compatible with nearly all dMRI sampling schemes. Drawing on a diverse set of software suites to capitalize on their complementary strengths, QSIPrep facilitates the implementation of best practices for processing of diffusion images.

Pandora: 4-D White Matter Bundle Population-Based Atlases Derived from Diffusion MRI Fiber Tractography.

Brain atlases have proven to be valuable neuroscience tools for localizing regions of interest and performing statistical inferences on populations. Although many human brain atlases exist, most do not contain information about white matter structures, often neglecting them completely or labelling all white matter as a single homogenous substrate. While few white matter atlases do exist based on diffusion MRI fiber tractography, they are often limited to descriptions of white matter as spatially separate "regions" rather than as white matter "bundles" or fascicles, which are well-known to overlap throughout the brain. Additional limitations include small sample sizes, few white matter pathways, and the use of outdated diffusion models and techniques. Here, we present a new population-based collection of white matter atlases represented in both volumetric and surface coordinates in a standard space. These atlases are based on 2443 subjects, and include 216 white matter bundles derived from 6 different automated state-of-the-art tractography techniques. This atlas is freely available and will be a useful resource for parcellation and segmentation.

Bifurcated Topological Optimization for IVIM.

In this work, we shed light on the issue of estimating Intravoxel Incoherent Motion (IVIM) for diffusion and perfusion estimation by characterizing the objective function using simplicial homology tools. We provide a robust solution via topological optimization of this model so that the estimates are more reliable and accurate. Estimating the tissue microstructure from diffusion MRI is in itself an ill-posed and a non-linear inverse problem. Using variable projection functional (VarPro) to fit the standard bi-exponential IVIM model we perform the optimization using simplicial homology based global optimization to better understand the topology of objective function surface. We theoretically show how the proposed methodology can recover the model parameters more accurately and consistently by casting it in a reduced subspace given by VarPro. Additionally we demonstrate that the IVIM model parameters cannot be accurately reconstructed using conventional numerical optimization methods due to the presence of infinite solutions in subspaces. The proposed method helps uncover multiple global minima by analyzing the local geometry of the model enabling the generation of reliable estimates of model parameters.

Segmentation of the brain using direction-averaged signal of DWI images.

Segmentation of brain tissue in diffusion MRI image space has some unique advantages. A novel segmentation method using the direction-averaged diffusion weighted imaging (DWI) signal is proposed. Two images can be obtained from the fitting of the direction-averaged DWI signal as a function of b-value: one with superior contrast between the gray matter and white matter; one with prominent CSF contrast. A pseudo T1 weighted image can be constructed and standard segmentation tools can be applied. The method was tested on the HCP dataset using SPM12, and showed good agreement with segmentation using the T1 weighted image with the same resolution. The Dice score was all greater than 0.88 for GM or WM with full DWI data and very stable against subsampling of the DWI data in number of diffusion directions, number of shells, and spatial resolution.