In this work, we introduce a method to classify Multiple Sclerosis (MS) patients into four clinical profiles using structural connectivity information. For the first time, we try to solve this ...question in a fully automated way using a computer-based method. The main goal is to show how the combination of graph-derived metrics with machine learning techniques constitutes a powerful tool for a better characterization and classification of MS clinical profiles.
Sixty-four MS patients 12 Clinical Isolated Syndrome (CIS), 24 Relapsing Remitting (RR), 24 Secondary Progressive (SP), and 17 Primary Progressive (PP) along with 26 healthy controls (HC) underwent MR examination. T1 and diffusion tensor imaging (DTI) were used to obtain structural connectivity matrices for each subject. Global graph metrics, such as density and modularity, were estimated and compared between subjects' groups. These metrics were further used to classify patients using tuned Support Vector Machine (SVM) combined with Radial Basic Function (RBF) kernel.
When comparing MS patients to HC subjects, a greater assortativity, transitivity, and characteristic path length as well as a lower global efficiency were found. Using all graph metrics, the best
-Measures (91.8, 91.8, 75.6, and 70.6%) were obtained for binary (HC-CIS, CIS-RR, RR-PP) and multi-class (CIS-RR-SP) classification tasks, respectively. When using only one graph metric, the best
-Measures (83.6, 88.9, and 70.7%) were achieved for modularity with previous binary classification tasks.
Based on a simple DTI acquisition associated with structural brain connectivity analysis, this automatic method allowed an accurate classification of different MS patients' clinical profiles.
Recent advances in image acquisition and processing techniques, along with the success of novel deep learning architectures, have given the opportunity to develop innovative algorithms capable to ...provide a better characterization of neurological related diseases. In this work, we introduce a neural network based approach to classify Multiple Sclerosis (MS) patients into four clinical profiles. Starting from their structural connectivity information, obtained by diffusion tensor imaging and represented as a graph, we evaluate the classification performances using unweighted and weighted connectivity matrices. Furthermore, we investigate the role of graph-based features for a better characterization and classification of the pathology. Ninety MS patients (12 clinically isolated syndrome, 30 relapsing-remitting, 28 secondary-progressive, and 20 primary-progressive) along with 24 healthy controls, were considered in this study. This work shows the great performances achieved by neural networks methods in the classification of the clinical profiles. Furthermore, it shows local graph metrics do not improve the classification results suggesting that the latent features created by the neural network in its layers have a much important informative content. Finally, we observe that graph weights representation of brain connections preserve important information to discriminate between clinical forms.
The neural substrate of high intelligence performances remains not well understood. Based on diffusion tensor imaging (DTI) which provides microstructural information of white matter fibers, we ...proposed in this work to investigate the relationship between structural brain connectivity and intelligence quotient (IQ) scores. Fifty-seven children (8-12 y.o.) underwent a MRI examination, including conventional T1-weighted and DTI sequences, and neuropsychological testing using the fourth edition of Wechsler Intelligence Scale for Children (WISC-IV), providing an estimation of the Full-Scale Intelligence Quotient (FSIQ) based on four subscales: verbal comprehension index (VCI), perceptual reasoning index (PRI), working memory index (WMI), and processing speed index (PSI). Correlations between the IQ scores and both graphs and diffusivity metrics were explored. First, we found significant correlations between the increased integrity of WM fiber-bundles and high intelligence scores. Second, the graph theory analysis showed that integration and segregation graph metrics were positively and negatively correlated with WISC-IV scores, respectively. These results were mainly driven by significant correlations between FSIQ, VCI, and PRI and graph metrics in the temporal and parietal lobes. In conclusion, these findings demonstrated that intelligence performances are related to the integrity of WM fiber-bundles as well as the density and homogeneity of WM brain networks.
The purpose of this study is classifying multiple sclerosis (MS) patients in the four clinical forms as defined by the McDonald criteria using machine learning algorithms trained on clinical data ...combined with lesion loads and magnetic resonance metabolic features.
Eighty-seven MS patients 12 Clinically Isolated Syndrome (CIS), 30 Relapse Remitting (RR), 17 Primary Progressive (PP), and 28 Secondary Progressive (SP) and 18 healthy controls were included in this study. Longitudinal data available for each MS patient included clinical (e.g., age, disease duration, Expanded Disability Status Scale), conventional magnetic resonance imaging and spectroscopic imaging. We extract
-acetyl-aspartate (NAA), Choline (Cho), and Creatine (Cre) concentrations, and we compute three features for each spectroscopic grid by averaging metabolite ratios (NAA/Cho, NAA/Cre, Cho/Cre) over good quality voxels. We built linear mixed-effects models to test for statistically significant differences between MS forms. We test nine binary classification tasks on clinical data, lesion loads, and metabolic features, using a leave-one-patient-out cross-validation method based on 100 random patient-based bootstrap selections. We compute F1-scores and BAR values after tuning Linear Discriminant Analysis (LDA), Support Vector Machines with gaussian kernel (SVM-rbf), and Random Forests.
Statistically significant differences were found between the disease starting points of each MS form using four different response variables: Lesion Load, NAA/Cre, NAA/Cho, and Cho/Cre ratios. Training SVM-rbf on clinical and lesion loads yields F1-scores of 71-72% for CIS vs. RR and CIS vs. RR+SP, respectively. For RR vs. PP we obtained good classification results (maximum F1-score of 85%) after training LDA on clinical and metabolic features, while for RR vs. SP we obtained slightly higher classification results (maximum F1-score of 87%) after training LDA and SVM-rbf on clinical, lesion loads and metabolic features.
Our results suggest that metabolic features are better at differentiating between relapsing-remitting and primary progressive forms, while lesion loads are better at differentiating between relapsing-remitting and secondary progressive forms. Therefore, combining clinical data with magnetic resonance lesion loads and metabolic features can improve the discrimination between relapsing-remitting and progressive forms.
Most of motor recovery usually occurs within the first 3 months after stroke. Herein is reported a remarkable late recovery of the right upper-limb motor function after a left middle cerebral artery ...stroke. This recovery happened progressively, from two to 12 years post-stroke onset, and along a proximo-distal gradient, including dissociated finger movements after 5 years. Standardized clinical assessment and quantified analysis of the reach-to-grasp movement were repeated over time to characterize the recovery. Twelve years after stroke onset, diffusion tensor imaging (DTI), functional magnetic resonance imaging (fMRI), and transcranial magnetic stimulation (TMS) analyses of the corticospinal tracts were carried out to investigate the plasticity mechanisms and efferent pathways underlying motor control of the paretic hand. Clinical evaluations and quantified movement analysis argue for a true neurological recovery rather than a compensation mechanism. DTI showed a significant decrease of fractional anisotropy, associated with a severe atrophy, only in the upper part of the left corticospinal tract (CST), suggesting an alteration of the CST at the level of the infarction that is not propagated downstream. The finger opposition movement of the right paretic hand was associated with fMRI activations of a broad network including predominantly the contralateral sensorimotor areas. Motor evoked potentials were normal and the selective stimulation of the right hemisphere did not elicit any response of the ipsilateral upper limb. These findings support the idea that the motor control of the paretic hand is mediated mainly by the contralateral sensorimotor cortex and the corresponding CST, but also by a plasticity of motor-related areas in both hemispheres. To our knowledge, this is the first report of a high quality upper-limb recovery occurring more than 2 years after stroke with a genuine insight of brain plasticity mechanisms.
The idea that intelligence is embedded not only in a single brain network, but instead in a complex, well-optimized system of complementary networks, has led to the development of whole brain network ...analysis. Using graph theory to analyze resting-state functional MRI data, we investigated the brain graph networks (or brain networks) of high intelligence quotient (HIQ) children. To this end, we computed the "hub disruption index κ," an index sensitive to graph network modifications. We found significant topological differences in the integration and segregation properties of brain networks in HIQ compared to standard IQ children, not only for the whole brain graph, but also for each hemispheric graph, and for the homotopic connectivity. Moreover, two profiles of HIQ children, homogenous and heterogeneous, based on the differences between the two main IQ subscales verbal comprehension index (VCI) and perceptual reasoning index (PRI), were compared. Brain network changes were more pronounced in the heterogeneous than in the homogeneous HIQ subgroups. Finally, we found significant correlations between the graph networks' changes and the full-scale IQ (FSIQ), as well as the subscales VCI and PRI. Specifically, the higher the FSIQ the greater was the brain organization modification in the whole brain, the left hemisphere, and the homotopic connectivity. These results shed new light on the relation between functional connectivity topology and high intelligence, as well as on different intelligence profiles.
Diffusion tensor imaging (DTI) is a sensitive tool for the assessment of microstructural alterations in brain white matter (WM). We propose a new processing technique to detect, local and global ...longitudinal changes of diffusivity metrics, in homologous regions along WM fiber-bundles. To this end, a reliable and automatic processing pipeline was developed in three steps: 1) co-registration and diffusion metrics computation, 2) tractography, bundle extraction and processing, and 3) longitudinal fiber-bundle analysis. The last step was based on an original Gaussian mixture model providing a fine analysis of fiber-bundle cross-sections, and allowing a sensitive detection of longitudinal changes along fibers. This method was tested on simulated and clinical data. High levels of F-Measure were obtained on simulated data. Experiments on cortico-spinal tract and inferior fronto-occipital fasciculi of five patients with Multiple Sclerosis (MS) included in a weekly follow-up protocol highlighted the greater sensitivity of this fiber scale approach to detect small longitudinal alterations.
The precise origin of phosphate that is removed during hemodialysis remains unclear; only a minority comes from the extracellular space. One possibility is that the remaining phosphate originates ...from the intracellular compartment, but there have been no available data from direct assessment of intracellular phosphate in patients undergoing hemodialysis.
We used phosphorus magnetic resonance spectroscopy to quantify intracellular inorganic phosphate (Pi), phosphocreatine (PCr), and
ATP. In our pilot, single-center, prospective study, 11 patients with ESKD underwent phosphorus (
P) magnetic resonance spectroscopy examination during a 4-hour hemodialysis treatment. Spectra were acquired every 152 seconds during the hemodialysis session. The primary outcome was a change in the PCr-Pi ratio during the session.
During the first hour of hemodialysis, mean phosphatemia decreased significantly (-41%;
<0.001); thereafter, it decreased more slowly until the end of the session. We found a significant increase in the PCr-Pi ratio (+23%;
=0.001) during dialysis, indicating a reduction in intracellular Pi concentration. The PCr-
ATP ratio increased significantly (+31%;
=0.001) over a similar time period, indicating a reduction in
ATP. The change of the PCr-
ATP ratio was significantly correlated to the change of depurated Pi.
Phosphorus magnetic resonance spectroscopy examination of patients with ESKD during hemodialysis treatment confirmed that depurated Pi originates from the intracellular compartment. This finding raises the possibility that excessive dialytic depuration of phosphate might adversely affect the intracellular availability of high-energy phosphates and ultimately, cellular metabolism. Further studies are needed to investigate the relationship between objective and subjective effects of hemodialysis and decreases of intracellular Pi and
ATP content.
Intracellular Phosphate Concentration Evolution During Hemodialysis by MR Spectroscopy (CIPHEMO), NCT03119818.
Abstract
BACKGROUND
Diffusion imaging tractography caught the attention of the scientific community by describing the white matter architecture in vivo and noninvasively, but its application to small ...structures such as cranial nerves remains difficult. The few attempts to track cranial nerves presented highly variable acquisition and tracking settings.
OBJECTIVE
To conduct and present a targeted review collecting all technical details and pointing out challenges and solutions in cranial nerve tractography.
METHODS
A “targeted” review of the scientific literature was carried out using the MEDLINE database. We selected studies that reported how to perform the tractography of cranial nerves, and extracted the following: clinical context; imaging acquisition settings; tractography parameters; regions of interest (ROIs) design; and filtering methods.
RESULTS
Twenty-one published articles were included. These studied the optic nerves in suprasellar tumors, the trigeminal nerve in neurovascular conflicts, the facial nerve position around vestibular schwannomas, or all cranial nerves. Over time, the number of MRI diffusion gradient directions increased from 6 to 101. Nine tracking software packages were used which offered various types of tridimensional display. Tracking parameters were disparately detailed except for fractional anisotropy, which ranged from 0.06 to 0.5, and curvature angle, which was set between 20° and 90°. ROI design has evolved towards a multi-ROI strategy. Furthermore, new algorithms are being developed to avoid spurious tracts and improve angular resolution.
CONCLUSION
This review highlights the variability in the settings used for cranial nerve tractography. It points out challenges that originate both from cranial nerve anatomy and the tractography technology, and allows a better understanding of cranial nerve tractography.