Siemens Healthcare AG
Advanced Clinical Imaging Technology

Author Of 5 Presentations

Machine Learning/Network Science Poster Presentation

P0011 - Lesion disconnectomics using atlas-based tractography (ID 1293)

Speakers
Presentation Number
P0011
Presentation Topic
Machine Learning/Network Science

Abstract

Background

Recent studies have described Multiple Sclerosis (MS) as a disconnection syndrome (Rocca et al. 2015). Modelling disconnectomes using brain networks enables to quantify connectivity loss using graph analysis. To build structural connectomes, high-quality diffusion Magnetic Resonance Imaging (dMRI) and robust tractography algorithms are typically required. However, high-quality dMRI is rarely acquired in clinical workups due to time constraints.

Objectives

We propose to use a tractography atlas to extract brain connectivity loss in response to lesions without requiring dMRI, and to model structural disconnectomes with brain graphs. Topological graph features are proposed as new radiological biomarkers and their relation with Total Lesion Volume (TLV) and Expanded Disability Status Scale (EDSS) are studied.

Methods

589 MS patients (159 males, age 28±8yo, EDSS 2.40±1.22, TLV 13.0±14.6mL) underwent MRI at 3T (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany). Acquisition protocols included T1-weighted magnetization-prepared rapid gradient echo (MPRAGE) and fluid-attenuated inversion recovery (FLAIR).

Lesions were segmented using LeMan-PV, a prototype lesion segmentation algorithm (Fartaria et al. 2016). The lesion masks were registered to standard MNI space and overlapped with the HCP842 tractography atlas (Yeh et al. 2018). Streamlines passing through lesions were isolated to define the affected connectivity.

The disconnectome graph was built using brain regions from the Brainnetome atlas (Fan et al. 2016) as nodes, whilst edges were weighted by the percent of unaffected streamlines connecting two nodes relative to the atlas connectivity. Topological features were extracted from the disconnectome graph and their Spearman’s correlations with TLV and EDSS were computed.

Results

Transitivity (T) and global efficiency (GE) decreased for larger TLV (R=-0.42 and R=-0.78), whereas the average shortest path length (PL) increased (R=0.78). When looking at correlations with EDSS, T (R=-0.17), GE (R=-0.24) and PL (R=0.23) showed stronger associations than lesion count (R=0.14) but were comparable to TLV (R=0.23). All correlations were significant (p<0.001).

Conclusions

We proposed an atlas-based disconnectome model which allowed to study connectivity loss in MS patients without requiring dMRI. Overall, patients showed a lower small-worldness and efficiency for larger TLV and worse disability. These observations were consistent with previous studies on diffusion-based connectomes and open new avenues of research for routine clinical data.

Collapse
Imaging Poster Presentation

P0533 - A Promising Biomarker Based on T1 Relaxation Time Mapping for Early MS (ID 863)

Abstract

Background

Regional brain atrophy is a sensitive disability marker for MS patients. A previous study has shown that atrophy of the corpus callosum is an early marker for disease progression. However, the relationship between diffuse pathology in specific brain regions and the course of regional atrophy development remains poorly understood.

Objectives

To investigate quantitative T1 maps and entropy (amount of T1 inhomogeneity) in regional brain structures from diagnostic MRI (performed at disease onset) of MS patients and compare these findings with healthy controls (HC).

Methods

Fifty MS patients and 102 HC were examined on a 3T MRI scanner (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany). The MRI protocol comprised 3D MP2RAGE, 3D MPRAGE, 3D FLAIR and 3D DIR. The calculation of T1 maps, brain structure segmentations and brain volume measurements were obtained from a single MP2RAGE scan. Lesion segmentation masks were obtained using the LeManPV prototype software (Siemens Healthcare, Erlangen, Germany). We evaluated T1 maps from normal-appearing white matter (excluding lesions) in the corpus callosum, the brain lobes, brainstem and cerebellum, as well as from normal-appearing gray matter (excluding lesions) in the thalami, basal ganglia, and cortical gray matter. We calculated median regional T1 relaxation times, T1 entropy and volume for the above-mentioned structures for the early-MS group and 50 age- and sex-matched HC subjects. Statistical comparison was performed using t-tests.

Results

The median T1 of the corpus callosum in the early MS group was 838 ms (SD 38.5), with entropy 8.42 (SD 0.24); compared to 810 ms (SD 25.2) and 8.23 (SD 0.13) in the HC group. Statistically significant differences were found in T1 times and entropy between the groups (p<0.001); volumes were, however, not statistically different. Smaller but also statistically significant differences in T1 maps and entropy were found for white matter of the brain lobes (p<0.001). Thalami volumes showed statistically significant differences between groups, but not median T1 times (MS group 1055 ms, SD 32.6 vs. HC 1049 ms, SD 21.2).

Conclusions

Pathology of the normal-appearing white matter in T1 relaxometry can already be detected at MS disease onset. In particular, corpus callosum T1 times were considerably higher at clinical onset of MS compared to HC. We hypothesize that early microstructural changes detected at disease onset lead to evolution of regional brain atrophy.

Collapse
Imaging Poster Presentation

P0627 - Quantitative T1 changes relate to infratentorial pathology in early multiple sclerosis. (ID 1844)

Abstract

Background

The presence of infratentorial lesions early in the disease has been shown to have prognostic value for future disability in multiple sclerosis (MS). Quantitative imaging metrics such as T1 relaxometry might contribute to understanding the relationship between supratentorial (ST), infratentorial (IT), and spinal cord (SC) pathology.

Objectives

Our aim was to explore the association between ST, IT and SC pathology and microstructural tissue alterations assessed with T1 relaxometry in T2-hyperintense lesions as well as cerebral and cerebellar normal-appearing white matter (NAWM) in patients with recently diagnosed MS with- and without IT lesions.

Methods

Microstructural tissue alterations were assessed in 42 patients (mean age 33.6±8.0 years, median MS duration 0.2 years (0-2.3)) as deviations from normative T1 times, both obtained from the MP2RAGE sequence at 3T (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany). The normative T1 values were voxel-wise modelled via a study-specific atlas based on spatially normalized data from 102 healthy individuals (21-59 years). Relationship between normalized IT volumes (mesencephalon, pons, medulla oblongata, cerebellum), SC volume, ST and IT lesion loads estimated by the Morphobox prototype, Scanview and LemanPV prototype, respectively and the deviations from normative T1 times expressed as z-score-derived metrics (volumes and means of voxels with z-scores above z-score 2 and below z-score 2) in lesions, cerebral and cerebellar NAWM were studied by partial correlations adjusted for age and brain lesion volume.

Results

Patients with IT lesions (n=23, 33.0±8.5 years) had larger lesion load, higher volumes of voxels with positive z-scores (> 2), higher mean of z-scores above 2 in lesions, and larger thalami than patients without IT lesions (n=19, 34.3±7.7 years). The remaining volumes and z-scores derived metrics did not differ between groups. Cerebellar volume correlated negatively with volume of voxels with negative z-scores (< 2) in cerebellar NAWM (partial correlation coefficient r=-.437, p=.005) only in patients with IT lesions. In patients without IT lesions, SC and pons volumes correlated negatively with volume of voxels with positive z-scores corresponding to areas of supratentorial T2 lesions (SC: r=-.669, p=.003, pons: r=-0.606, p=0.01).

Conclusions

Microstructural alterations identified as T1 z-scores relate differently to IT and SC volumes in MS patients with and without IT lesions. In the presence of IT lesions, changes in cerebellar NAWM (T1 shortening relative to healthy controls) are associated with lower cerebellar volume. In the absence of IT lesions, the association of cerebellar NAWM and cerebellar volume is not present. In patients without IT lesions, microstructural alterations in ST lesions (T1 prolongation) that might indicate the extent of tissue damage in lesions, are associated with lower pontine and SC volumes regardless of the T2 lesion load.

Collapse
Imaging Poster Presentation

P0628 - Quantitative T1 deviations in brain lesions and NAWM improve the clinico-radiological correlation in early MS (ID 763)

Abstract

Background

Although conventional MRI acquisitions are of essence in the monitoring of MS, they show low specificity towards the microstructural nature of tissue alterations and exhibit rather low correlations with clinical metrics (“clinico-radiological paradox”). Conversely, recent advances in brain relaxometry allow characterizing microstructural alterations on a single-subject basis; the question yet remains whether such quantitative measurements can help bridging the gap between radiological and clinical findings.

Objectives

This study investigates whether automatically assessed alterations of T1 relaxation times in brain lesions and normal-appearing white matter (NAWM) improve clinico–radiological correlations in early MS with respect to conventional measures.

Methods

102 healthy controls (65% female, [21-59] y/o) and 50 early-MS patients (76% female, [19-52] y/o) underwent MRI at 3T (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany). The employed 3D protocol comprised MPRAGE, FLAIR (both used for lesion segmentation as in [Fartaria et al., 2017, MICCAI]), and MP2RAGE for T1 mapping.

After the healthy controls’ data were spatially normalized into a study-specific template, reference T1 values in healthy tissues were established by linear, voxel-wise modelling of the T1 inter-subject variability [Piredda et al., MRM, 2020]. In the MS cohort, T1 deviations from the established references were calculated as z-score maps.

Correlations between the EDSS and conventional measures, i.e. lesion volume and count, were compared against correlations with z-score-derived metrics in lesions and NAWM, namely the volume of voxels exceeding a given z-score threshold.

Results

Correlations between EDSS and lesion volume and count were found to be 0.23 and 0.18, respectively. Higher correlations were found between EDSS and the volume of voxels exceeding an absolute z-score threshold of 2, both in lesions and NAWM, with ρ=0.3 and ρ=0.33, respectively. Correlation further improved when considering only negative z-scores, ρ=0.36 for lesions and ρ=0.39 for NAWM. The highest correlation was found when considering absolute z-scores in the occipital lobe NAWM, ρ=0.47.

Conclusions

Microstructural alterations identified as T1 z-scores were found to improve clinico–radiological correlation in comparison to conventional measures (lesion volume and count). Of notice, negative z-scores (i.e. abnormal T1 shortening), which may be due to an increase in iron content, appear to be a potential predictor for the clinical state of an early MS patient.

Collapse
Imaging Poster Presentation

P0645 - Spinal cord pathology in a large cohort of MS patients with different levels of disability and MS phenotypes (ID 865)

Abstract

Background

SC pathology occurs early in the course of MS. However, few studies have investigated the relationship between lesions, diffuse changes and mean upper cervical cord area (MUCCA) in MS patients with different levels of disability in detail.

Objectives

To explore spinal cord (SC) pathology in multiple sclerosis (MS) patients with different levels of disability and MS phenotypes.

Methods

638 MS patients with different degrees of disability and 102 healthy controls (HC) underwent MRI on a 3T (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany). The MRI protocol comprised transversal 3D-T2WI for MUCCA, sagittal T2WI-Fat-Sat and PDWI for SC pathology, and 3D-MPRAGE for regional brain volume (BV). MUCCA was measured automatically between the C3 and C4 vertebra (ScanView.cz). Global and regional BVs were estimated by the fully automated MorphoBox prototype (Siemens Healthcare, Erlangen, Germany). Diffuse changes, number and location of SC lesions were assessed manually. Patients and HC were matched by sex and age using propensity scores. MUCCA, regional BVs and SC pathology were compared among matched subgroups of: 54 patients with mild disability (EDSS=<1.5), 54 patients with mild-to-moderate disability (EDSS 2-3.5), 54 patients with severe disability (EDSS 4-4.5), 54 patients with very severe disability (EDSS>=5), 18 primary progressive (PP) patients, and 54 controls from the HC group. ANOVA test was used for between-group comparison.

Results

There was a trend of lower MUCCA with higher disability level. Mean MUCCA was 76.5±10.8 mm2 invery severe, 80.1±9.6 mm2 in severe, 85.7±8.0 mm2 in moderate, 85.6±8.5 mm2 in mild disability, and 90±7.7 mm2 in HC groups. There was a significant difference in MUCCA between HC and mild disability group (p<0.001). SC pathology was prominent in 64.1% of the patients with mild disability, compared to 90.4% patients with very severe disability. The percentage of diffuse changes varied greatly between the groups, with prevalence increasing almost four times between patients with mild and very severe disability.

Conclusions

SC pathology is present in all disability MS groups. MUCCA differentiated between patients with mild disability and healthy controls, suggesting that it may be promising for the implementation in diagnostic protocols. The evaluation of diffuse changes can help to predict disability. Low MUCCA together with prominent diffuse changes could help differentiate PP MS from other MS phenotypes.

Collapse