T2 Relaxometry MRI Predicts Cerebral Palsy in Preterm Infants

Fully automated quality control of rigid and affine registrations of T1w and T2w MRI in big data using machine learning

BACKGROUND

Magnetic resonance imaging (MRI)-based morphometry and relaxometry are proven methods for the structural assessment of the human brain in several neurological disorders. These procedures are generally based on T1-weighted (T1w) and/or T2-weighted (T2w) MRI scans, and rigid and affine registrations to a standard template(s) are essential steps in such studies. Therefore, a fully automatic quality control (QC) of these registrations is necessary in big data scenarios to ensure that they are suitable for subsequent processing.

METHOD

A supervised machine learning (ML) framework is proposed by computing similarity metrics such as normalized cross-correlation, normalized mutual information, and correlation ratio locally. We have used these as candidate features for cross-validation and testing of different ML classifiers. For 5-fold repeated stratified grid search cross-validation, 400 correctly aligned, 2000 randomly generated misaligned images were used from the human connectome project young adult (HCP-YA) dataset. To test the cross-validated models, the datasets from autism brain imaging data exchange (ABIDE I) and information eXtraction from images (IXI) were used.

RESULTS

The ensemble classifiers, random forest, and AdaBoost yielded best performance with F1-scores, balanced accuracies, and Matthews correlation coefficients in the range of 0.95-1.00 during cross-validation. The predictive accuracies reached 0.99 on the Test set #1 (ABIDE I), 0.99 without and 0.96 with noise on Test set #2 (IXI, stratified w.r.t scanner vendor and field strength).

CONCLUSIONS

The cross-validated and tested ML models could be used for QC of both T1w and T2w rigid and affine registrations in large-scale MRI studies.

Ultra-high field imaging in Major Depressive Disorder: a review of structural and functional studies

BACKGROUND

Major depressive disorder (MDD) is a severe and pervasive psychiatric condition with a lifetime prevalence of 15-25%. Numerous Magnetic Resonance Imaging (MRI) studies employing scans at field strengths of 1.5T or 3T have been carried out in the last decades, providing an unprecedented insight into the neural correlates of MDD. However, in recent years, MRI technology has largely progressed and the use of scans at ultra-high field (≥ 7T) has improved the sensitivity and the resolution of MR images. In this context, with this review we aim to summarize evidence of structural and functional brain mechanisms underlying MDD obtained with ultra-high field MRI.

METHODS

We conducted a search on PubMed, Scopus and Web of Science of neuroimaging studies on MDD patients, which employed ultra-high field MRI. We detected six structural MRI studies, two Diffusion Tensor Imaging (DTI) studies and five functional MRI (fMRI) studies.

RESULTS

Overall, the MRI and DTI studies showed volumetric and structural connectivity alterations in the hippocampus and, to a lesser extent, in the amygdala. In contrast, more heterogeneous results were reported by fMRI studies, which, though, described functional abnormalities in the cingulate cortex, thalamus and several other brain areas.

LIMITATIONS

The small sample size and the heterogeneity in patients' samples, processing and study design limit the conclusion of the present review.

CONCLUSIONS

Studies employing scans at ultra-high magnetic field may provide a useful contribution to the mixed body of literature on MDD. This preliminary but promising evidence confirms the importance of performing ultra-high field MRI investigations in order to detect and better characterize subtle brain abnormalities in MDD.

Associations between brain volumetry and relaxometry signatures and the Edmonton Frail Scale in frailty

Background

Frailty is a geriatric condition characterized by a decreased reserve. The Edmonton frailty scale (EFS) has been widely used as an assessment tool in clinical practice. However, the brain's underlying pathophysiological changes in frailty and their associations with the EFS remain unclear. This study aimed to explore the associations between brain volumetry and relaxometry signatures and the EFS (and each domain score of the EFS) in frailty.

Methods

A total of 40 non-demented subjects were enrolled in this prospective study. Frailty assessment was performed for each subject according to the EFS. All subjects underwent synthetic magnetic resonance imaging (MRI) (MAGnetic resonance image Compilation, MAGiC) and three-dimensional fast spoiled gradient-recalled echo (3D-FSPGR) T1-weighted structural image acquisitions on a 3.0 T MR scanner. Brain segmentation was performed based on quantitative values obtained from the MAGiC and 3D-FSPGR images. Volumetry and relaxometry of the global brain and regional gray matter (GM) were also obtained. The associations between the total EFS score (and the score of each domain) and the brain's volumetry and relaxometry were investigated by partial correlation while eliminating the effects of age. Multiple comparisons of regional GM volumetry and relaxometry analyses were controlled by false discovery rate (FDR) correction. All data were analyzed using the SPSS 13.0 statistical package (IBM, Armonk, NY, USA) and MATLAB (MathWorks, Natick, MA, USA).

Results

For global volumetry, significant correlations were found between multiple global volumetry parameters and the EFS, as well as the cognition score, functional independence score, nutrition score, and functional performance score (P<0.05). For global relaxometry, notable positive correlations were found between the T2 values of gray and white matter (WM) and the EFS (r=0.357, P=0.026; r=0.357, P=0.026, respectively). Significant correlations were also identified between the T2 value of GM, the T1, T2, and PD values of WM, and the cognition score (r=0.426, P=0.007; r=0.456, P=0.003; r=0.377, P=0.018; r=0.424, P=0.007, respectively), functional independence score (r=-0.392, P=0.014; r=-0.611, P<0.001; r=-0.367, P=0.022; r=-0.569, P<0.001, respectively), and functional performance score (r=0.337, P=0.036; r=0.472, P=0.002; r=0.354, P=0.027; r=0.376, P=0.018, respectively). For regional GM volumetry, multiple regions showed significant negative correlations with the EFS (P<0.05). Notable negative correlations were found between multiple regional GM volume and the functional independence score (P<0.05). For regional GM relaxometry, the T1 and T2 values of several regions showed significant negative correlations with the functional independence score (T1 value of caudate, r=-0.617, P<0.001; T2 value of insula, r=-0.510, P=0.015; T2 value of caudate, r=-0.633, P<0.001, respectively). No significant correlation was found between the domain scores of the EFS and regional GM PD values (P>0.05).

Conclusions

In conclusion, brain volumetry and relaxometry signatures showed strong associations with the EFS and some EFS domain scores in frailty. These associations may reveal the possible underlying pathophysiology of the EFS and different domains of the EFS.