Multiple Sclerosis)
Conditions
Keywords
cortical lesions, leuco-cortical lesions, central vein sign
Brief summary
This study investigates the specificity/sensitivity of the combined presence of cortical lesions (CLs)/leuco-cortical lesions (LCLs) and central veins sign (CVs) for multiple sclerosis (MS) diagnosis and differential diagnosis.
Detailed description
CLs and LCLs may be detected at 3 Tesla (T) MRI by using dedicated sequences. 3D Double inversion recovery (DIR), Phase-Sensitive Inversion Recovery (PSIR), Magnetization Prepared - RApid Gradient Echo (MPRAGE) and Magnetization Prepared - 2- RApid Gradient Echo (MP2RAGE) have all shown variable sensitivity to CLs and LCLs. The central vein sign (CVS, i.e. the detection of a central vessel in a focal lesion) has been recently proposed as a biomarker for distinguishing between MS and not and not MS. Both the presence of CL and CVs brings high sensitivity and specificity in distinguishing MS from not MS patients; whether their combination achieves higher sensitivity and specificity to MS diagnosis and differential diagnosis is to date not know. This study investigates the specificity/sensitivity of the combined presence of cortical lesions (CLs)/leuco-cortical lesions (LCLs) and central veins sign (CVs) for multiple sclerosis (MS) diagnosis and differential diagnosis.
Interventions
OTHERImage analysis/ Central vein detection
Image analysis/ Central vein detection will be performed using a fully automated approach based on an ensemble of 3D convolutional neural Networks. The method is applied to T1w, T2w and T2\*w data . Results will be manually reviewed by 2 experts in CVs detection. MRI data must have been collected at 3 Tesla (3T) between 1990 and 2020 and must include T1 w, T2 w, T2-star based sequence (CVs detection).
OTHERImage analysis/ Automatic cortical lesion detection
Image analysis/ Automatic cortical lesion detection will be performed by using a convolutional neural network-based method. MRI data must have been collected at 3T between 1990 and 2020 and must include Double inversion recovery (DIR)/Phase-Sensitive Inversion Recovery (PSIR)/Magnetization Prepared - 2- RApid Gradient Echo Gradient Echo (MP2RAGE) (eventually also high spatial resolution Magnetization Prepared - RApid Gradient Echo (MPRAGE)) for cortical lesion detection.
Sponsors
University Hospital, Basel, Switzerland
Study design
Observational model
COHORT
Time perspective
RETROSPECTIVE
Eligibility
Sex/Gender
ALL
Healthy volunteers
No
Inclusion criteria
* Clinical and MRI data of patients with MS or MS mimics (migraine, vasculitis, diabetes, neuromyelitis optica spectrum disorder) * MRI data must have been collected at 3T between 1990 and 2020 and must include T1 w, T2 w, T2-star based sequence (CVs detection) and DIR/PSIR/MP2RAGE (eventually also high spatial resolution MPRAGE) for cortical lesion detection
Exclusion criteria
* Unavailability of institutional informed consent * Unclear diagnosis * Other neurological or psychiatric diseases.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| differentiation between patients with MS and clinically isolated syndrome and patients without MS ((specificity) | at Baseline | differentiation between patients with MS and clinically isolated syndrome and patients without MS by analyzing the combination of the presence of CL/LCLs and CVs as compared to the sensitivity/specificity achieved by either CL/LCLs or CVs (sensitivity) |
| specificity of the combination of CLs/ LCLs compared to the one of the current diagnostic criteria for MS diagnosis | at Baseline | specificity of the combination of CLs/ LCLs compared to the one of the current diagnostic criteria for MS diagnosis |
| differentiation between patients with MS and clinically isolated syndrome and patients without MS (specificity) | at Baseline | differentiation between patients with MS and clinically isolated syndrome and patients without MS by analyzing the combination of the presence of CL/LCLs and CVs as compared to the sensitivity/specificity achieved by either CL/LCLs or CVs (sensitivity) |
| sensitivity of the combination of CLs/ LCLs compared to the one of the current diagnostic criteria for MS diagnosis | at Baseline | sensitivity of the combination of CLs/ LCLs compared to the one of the current diagnostic criteria for MS diagnosis |
Countries
Switzerland
Outcome results
None listed