A diffusion-matched principal component analysis (DM-PCA) based two-channel denoising procedure for high-resolution diffusion-weighted MRI

Artificial intelligence for diffusion MRI-based tissue microstructure estimation in the human brain: an overview

Artificial intelligence (AI) has made significant advances in the field of diffusion magnetic resonance imaging (dMRI) and other neuroimaging modalities. These techniques have been applied to various areas such as image reconstruction, denoising, detecting and removing artifacts, segmentation, tissue microstructure modeling, brain connectivity analysis, and diagnosis support. State-of-the-art AI algorithms have the potential to leverage optimization techniques in dMRI to advance sensitivity and inference through biophysical models. While the use of AI in brain microstructures has the potential to revolutionize the way we study the brain and understand brain disorders, we need to be aware of the pitfalls and emerging best practices that can further advance this field. Additionally, since dMRI scans rely on sampling of the q-space geometry, it leaves room for creativity in data engineering in such a way that it maximizes the prior inference. Utilization of the inherent geometry has been shown to improve general inference quality and might be more reliable in identifying pathological differences. We acknowledge and classify AI-based approaches for dMRI using these unifying characteristics. This article also highlighted and reviewed general practices and pitfalls involving tissue microstructure estimation through data-driven techniques and provided directions for building on them.

Principal component analysis denoising improves sensitivity of MR diffusion to detect white matter injury in neuroHIV

BACKGROUND AND PURPOSE

Diffusion-weighted imaging is able to capture important information about cerebral white matter (WM) structure. However, diffusion data can suffer from MRI and biological noise that degrades the quality of the images and makes finding important features difficult. We investigated how effectively local and nonlocal denoising increased the sensitivity to detect differences in cerebral WM in neuroHIV.

METHODS

We utilized principal component analysis (PCA) denoising to detect WM differences using fractional anisotropy. Local and nonlocal PCA denoising paradigms were implemented that varied in search area and number of components. We examined different-sized WM tracts that consistently show differences between people living with Human Immunodeficiency Virus (HIV) (PWH) and HIV-negative individuals (corpus callosum, forceps minor, and right uncinate fasciculus), and size-matched tracts not typically associated with HIV-related differences (spinothalamic, right medial lemniscus, and left occipitopontine). We first conducted a full sample comparison of WM differences between groups, and then randomly reduced the sample to the point where we still found differences in WM.

RESULTS

Nonlocal PCA denoising allowed us to detect differences after a sample reduction of 35% in the forceps minor, 17% in the right uncinate fasciculus, and 6% in the corpus callosum.

CONCLUSIONS

PCA denoising had a beneficial effect on detecting significant differences in PWH after sample size reduction. The smaller forceps minor tract and right uncinate fasciculus showed greater sensitivity to PCA denoising than the larger corpus callosum. These results show the importance of identifying the most effective PCA denoising strategy when investigating WM in PWH.

Improved Multi-Echo Gradient-Echo-Based Myelin Water Fraction Mapping Using Dimensionality Reduction

Multi-echo gradient-echo (mGRE)-based myelin water fraction (MWF) mapping is a promising myelin water imaging (MWI) modality but is vulnerable to noise and artifact corruption. The linear dimensionality reduction (LDR) method has recently shown improvements with regard to these challenges. However, the magnitude value based low rank operators have been shown to misestimate the MWF for regions with [Formula: see text] anisotropy. This paper presents a nonlinear dimensionality reduction (NLDR) method to estimate the MWF map better by encouraging nonlinear low dimensionality of mGRE signal sources. Specifically, we implemented a fully connected deep autoencoder to extract the low-dimensional features of complex-valued signals and incorporated a sparse regularization to separate the anomaly sources that do not reside in the low-dimensional manifold. Simulations and in vivo experiments were performed to evaluate the accuracy of the MWF map under various situations. The proposed NLDR-based MWF improves the accuracy of the MWF map over the conventional nonlinear least-squares method and the LDR-based MWF and maintains robustness against noise and artifact corruption.

Human immunodeficiency virus-related decreases in corpus callosal integrity and corresponding increases in functional connectivity

People living with human immunodeficiency virus (PLWH) often have neurocognitive impairment. However, findings on HIV-related differences in brain network function underlying these impairments are inconsistent. One principle frequently absent from these reports is that brain function is largely emergent from brain structure. PLWH commonly have degraded white matter; we hypothesized that functional communities connected by degraded white matter tracts would show abnormal functional connectivity. We measured white matter integrity in 69 PLWH and 67 controls using fractional anisotropy (FA) in 24 intracerebral white matter tracts. Then, among tracts with degraded FA, we identified gray matter regions connected to these tracts and measured their functional connectivity during rest. Finally, we identified cognitive impairment related to these structural and functional connectivity systems. We found HIV-related decreased FA in the corpus callosum body (CCb), which coordinates activity between the left and right hemispheres, and corresponding increases in functional connectivity. Finally, we found that individuals with impaired cognitive functioning have lower CCb FA and higher CCb functional connectivity. This result clarifies the functional relevance of the corpus callosum in HIV and provides a framework in which abnormal brain function can be understood in the context of abnormal brain structure, which may both contribute to cognitive impairment.

Structural and Functional Brain Abnormalities in Human Immunodeficiency Virus Disease Revealed by Multimodal Magnetic Resonance Imaging Fusion: Association With Cognitive Function

BACKGROUND

Human immunodeficiency virus (HIV)-associated neurocognitive impairment remains a prevalent comorbidity that impacts daily functioning and increases morbidity. While HIV infection is known to cause widespread disruptions in the brain, different magnetic resonance imaging (MRI) modalities have not been effectively integrated. In this study, we applied 3-way supervised fusion to investigate how structural and functional coalterations affect cognitive function.

METHODS

Participants (59 people living with HIV and 58 without HIV) completed comprehensive neuropsychological testing and multimodal MRI scanning to acquire high-resolution anatomical, diffusion-weighted, and resting-state functional images. Preprocessed data were reduced using voxel-based morphometry, probabilistic tractography, and regional homogeneity, respectively. We applied multimodal canonical correlation analysis with reference plus joint independent component analysis using global cognitive functioning as the reference.

RESULTS

Compared with controls, participants living with HIV had lower global cognitive functioning. One joint component was both group discriminating and correlated with cognitive function. This component included the following covarying regions: fractional anisotropy in the corpus callosum, short and long association fiber tracts, and corticopontine fibers; gray matter volume in the thalamus, prefrontal cortex, precuneus, posterior parietal regions, and occipital lobe; and functional connectivity in frontoparietal and visual processing regions. Component loadings for fractional anisotropy also correlated with immunosuppression.

CONCLUSIONS

These results suggest that coalterations in brain structure and function can distinguish people with and without HIV and may drive cognitive impairment. As MRI becomes more commonplace in HIV care, multimodal fusion may provide neural biomarkers to support diagnosis and treatment of cognitive impairment.

Brain multimodal co-alterations related to delay discounting: a multimodal MRI fusion analysis in persons with and without cocaine use disorder

BACKGROUND

Delay discounting has been proposed as a behavioral marker of substance use disorders. Innovative analytic approaches that integrate information from multiple neuroimaging modalities can provide new insights into the complex effects of drug use on the brain. This study implemented a supervised multimodal fusion approach to reveal neural networks associated with delay discounting that distinguish persons with and without cocaine use disorder (CUD).

METHODS

Adults with (n = 35) and without (n = 37) CUD completed a magnetic resonance imaging (MRI) scan to acquire high-resolution anatomical, resting-state functional, and diffusion-weighted images. Pre-computed features from each data modality included whole-brain voxel-wise maps for gray matter volume, fractional anisotropy, and regional homogeneity, respectively. With delay discounting as the reference, multimodal canonical component analysis plus joint independent component analysis was used to identify co-alterations in brain structure and function.

RESULTS

The sample was 58% male and 78% African-American. As expected, participants with CUD had higher delay discounting compared to those without CUD. One joint component was identified that correlated with delay discounting across all modalities, involving regions in the thalamus, dorsal striatum, frontopolar cortex, occipital lobe, and corpus callosum. The components were negatively correlated with delay discounting, such that weaker loadings were associated with higher discounting. The component loadings were lower in persons with CUD, meaning the component was expressed less strongly.

CONCLUSIONS

Our findings reveal structural and functional co-alterations linked to delay discounting, particularly in brain regions involved in reward salience, executive control, and visual attention and connecting white matter tracts. Importantly, these multimodal networks were weaker in persons with CUD, indicating less cognitive control that may contribute to impulsive behaviors.

Machine learning prediction of neurocognitive impairment among people with HIV using clinical and multimodal magnetic resonance imaging data

Diagnosis of HIV-associated neurocognitive impairment (NCI) continues to be a clinical challenge. The purpose of this study was to develop a prediction model for NCI among people with HIV using clinical- and magnetic resonance imaging (MRI)-derived features. The sample included 101 adults with chronic HIV disease. NCI was determined using a standardized neuropsychological testing battery comprised of seven domains. MRI features included gray matter volume from high-resolution anatomical scans and white matter integrity from diffusion-weighted imaging. Clinical features included demographics, substance use, and routine laboratory tests. Least Absolute Shrinkage and Selection Operator Logistic regression was used to perform variable selection on MRI features. These features were subsequently used to train a support vector machine (SVM) to predict NCI. Three different classification tasks were performed: one used only clinical features; a second used only selected MRI features; a third used both clinical and selected MRI features. Model performance was evaluated by area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity with a tenfold cross-validation. The SVM classifier that combined selected MRI with clinical features outperformed the model using clinical features or MRI features alone (AUC: 0.83 vs. 0.62 vs. 0.79; accuracy: 0.80 vs. 0.65 vs. 0.72; sensitivity: 0.86 vs. 0.85 vs. 0.86; specificity: 0.71 vs. 0.37 vs. 0.52). Our results provide preliminary evidence that combining clinical and MRI features can increase accuracy in predicting NCI and could be developed as a potential tool for NCI diagnosis in HIV clinical practice.

Evidence in support of chromosomal sex influencing plasma based metabolome vs APOE genotype influencing brain metabolome profile in humanized APOE male and female mice

Late onset Alzheimer’s disease (LOAD) is a progressive neurodegenerative disease with four well-established risk factors: age, APOE4 genotype, female chromosomal sex, and maternal history of AD. Each risk factor impacts multiple systems, making LOAD a complex systems biology challenge. To investigate interactions between LOAD risk factors, we performed multiple scale analyses, including metabolomics, transcriptomics, brain magnetic resonance imaging (MRI), and beta-amyloid assessment, in 16 months old male and female mice with humanized human APOE3 (hAPOE3) or APOE4 (hAPOE4) genes. Metabolomic analyses indicated a sex difference in plasma profile whereas APOE genotype determined brain metabolic profile. Consistent with the brain metabolome, gene and pathway-based RNA-Seq analyses of the hippocampus indicated increased expression of fatty acid/lipid metabolism related genes and pathways in both hAPOE4 males and females. Further, female transcription of fatty acid and amino acids pathways were significantly different from males. MRI based imaging analyses indicated that in multiple white matter tracts, hAPOE4 males and females exhibited lower fractional anisotropy than their hAPOE3 counterparts, suggesting a lower level of white matter integrity in hAPOE4 mice. Consistent with the brain metabolomic and transcriptomic profile of hAPOE4 carriers, beta-amyloid generation was detectable in 16-month-old male and female brains. These data provide therapeutic targets based on chromosomal sex and APOE genotype. Collectively, these data provide a framework for developing precision medicine interventions during the prodromal phase of LOAD, when the potential to reverse, prevent and delay LOAD progression is greatest.

Improved accuracy of apparent diffusion coefficient quantification using a fully automatic noise bias compensation method: Preliminary evaluation in prostate diffusion weighted imaging

Noise in diffusion magnetic resonance imaging can introduce bias in apparent diffusion coefficient (ADC) quantification. Previous studies proposed methods that are site-specific techniques as research tools with limited availability and typically require manual intervention, not completely ready to use in the clinical environment. The purpose of this study was to develop a fully automatic computational method to correct noise bias in ADC quantification and perform a preliminary evaluation in the clinical prostate diffusion weighted imaging (DWI). Using a pseudo replica approach for the noise map calculation as well as a direct mapping and a stepwise Chebychev polynomial modelling approach for the ADC fitting, a fully automatic noise-bias-compensated ADC calculation method was proposed and implemented both on the scanner and offline. The proposed method was validated in a computer simulation and a standardized diffusion phantom with ground-truth values. Two in vivo studies were performed to evaluate the proposed method in the clinical environment. The first in vivo study performed acquisitions using a clinically routine prostate DWI protocol on 29 subjects to evaluate the consistency between simulated and empirical results. In the second in vivo study, prostate ADC values of 14 subjects were compared between data acquired with external coils only and reconstructed with the proposed method vs. acquired with external combined with endorectal coils and reconstructed with the conventional method. In statistical analyses, p < 0.05 was regarded as significantly different. In the computer simulation, the proposed method showed smaller error percentage than the other methods and was significantly different (p < 2.2 × 10-16). With low signal-to-noise ratio (SNR), the conventional method underestimated ADC values compared to the ground truth values of the diffusion phantom, while the results of the proposed method were more consistent with the ground truth values. Statistical analyses showed no significant differences between measured and simulated results in the first in vivo study (p = 0.5618). Data from the second in vivo study showed that agreement between ADC measured with external coils only and combined coils was improved for the proposed method (mean bias: 0.04 × 10-3 mm2/s, 95% confidence interval (CI) = [-0.01, 0.09] × 10-3 mm2/s, p = 0.187), compared to the conventional method (mean bias: -0.12 × 10-3 mm2/s, 95% CI = [-0.17, -0.06] × 10-3 mm2/s, p < 0.0001). The proposed method compensates noise bias in low-SNR diffusion-weighted acquisitions and results show improved ADC quantification accuracy in the prostate. This method may be suitable for both clinical imaging and research utilizing ADC quantification.

Alteration of Diffusion-Tensor Magnetic Resonance Imaging Measures in Brain Regions Involved in Early Stages of Parkinson's Disease

Many nonmotor symptoms (e.g., hyposmia) appear years before the cardinal motor features of Parkinson's disease (PD). It is thus desirable to be able to use noninvasive brain imaging methods, such as magnetic resonance imaging (MRI), to detect brain abnormalities in early PD stages. Among the MRI modalities, diffusion-tensor imaging (DTI) is suitable for detecting changes in brain tissue structure due to neurological diseases. The main purpose of this study was to investigate whether DTI signals measured from brain regions involved in early stages of PD differ from those of healthy controls. To answer this question, we analyzed whole-brain DTI data of 30 early-stage PD patients and 30 controls using improved region of interest-based analysis methods. Results showed that (i) the fractional anisotropy (FA) values in the olfactory tract (connected with the olfactory bulb: one of the first structures affected by PD) are lower in PD patients than healthy controls; (ii) FA values are higher in PD patients than healthy controls in the following brain regions: corticospinal tract, cingulum (near hippocampus), and superior longitudinal fasciculus (temporal part). Experimental results suggest that the tissue property, measured by FA, in olfactory regions is structurally modulated by PD with a mechanism that is different from other brain regions.

Principal component analysis fosr fast and model-free denoising of multi b-value diffusion-weighted MR images

Despite the utility of tumour characterisation using quantitative parameter maps from multi-b-value diffusion-weighted MRI (DWI), clinicians often prefer the use of the image with highest diffusion-weighting (b-value), for instance for defining regions of interest (ROIs). However, these images are typically degraded by noise, as they do not utilize the information from the full acquisition. We present a principal component analysis (PCA) approach for model-free denoising of DWI data. PCA-denoising was compared to synthetic MRI, where a diffusion model is fitted for each voxel and a denoised image at a given b-value is generated from the model fit. A quantitative comparison of systematic and random errors was performed on data simulated using several diffusion models (mono-exponential, bi-exponential, stretched-exponential and kurtosis). A qualitative visual comparison was also performed for in vivo images in six healthy volunteers and three pancreatic cancer patients. In simulations, the reduction in random errors from PCA-denoising was substantial (up to 55%) and similar to synthetic MRI (up to 53%). Model-based synthetic MRI denoising resulted in substantial (up to 29% of signal) systematic errors, whereas PCA-denoising was able to denoise without introducing systematic errors (less than 2%). In vivo, the signal-to-noise ratio (SNR) and sharpness of PCA-denoised images were superior to synthetic MRI, resulting in clearer tumour boundaries. In the presence of motion, PCA-denoising did not cause image blurring, unlike image averaging or synthetic MRI. Multi-b-value MRI can be denoised model-free with our PCA-denoising strategy that reduces noise to a level similar to synthetic MRI, but without introducing systematic errors associated with the synthetic MRI method.

Improving the Accuracy, Quality, and Signal-To-Noise Ratio of MRI Parametric Mapping Using Rician Bias Correction and Parametric-Contrast-Matched Principal Component Analysis (PCM-PCA)

MRI parametric mapping, including T2 mapping, can quantitatively characterize tissue properties and is an important MRI procedure in biomedical research and studies of diseases [1-3]. However, the accuracy, quality, and signal-to-noise ratio (SNR) of MRI parametric mapping may be negatively impacted by Rician noise in multi-contrast MRI data [4]. As such, it is important to develop a post-processing method to minimize the negative impact of Rician noise. In this study, we report a new parametric-contrast-matched principal component analysis (PCM-PCA) denoising method that involves 1) identifying voxels with similar T2 decay characteristics and 2) using the principal component analysis (PCA) to denoise multi-contrast MRI data along the echo time (TE) dimension. We additionally evaluated the effects of integrating Rician bias correction and the new PCM-PCA method. In this study, we mathematically added Rician noise at various levels to human brain MRI data and performed different combinations of denoising and Rician bias correction on the magnitude-valued images. We found that MRI denoising using the PCM-PCA method resulted in improved image quality, SNR, and accuracy of the measured T2 relaxation time constants. Additionally, we found that for data with low SNR (e.g., 1.5 or lower), Rician bias correction further improved image quality and T2 mapping accuracy. In summary, our experimental results demonstrated that the new PCM-PCA denoising method and Rician bias correction adequately improve multi-contrast MRI quality and T2 parametric mapping accuracy.