Artificial Intelligence in Predicting Progression in Multiple Sclerosis Study

University of Ljubljana654 enrolled

Overview

The study proposal focuses on multiple sclerosis (MS), a chronic incurable disease of the central nervous system (CNS). The MS disease is characterised by recurrent transient disability progression, quantified by increase in the extended disability status score (EDSS), and subsequent remission (disappearance of symptoms and reduced EDSS score) or, alternatively, a gradual EDSS disability progression and exacerbation of associated symptoms. At the same time, the MS is characterised by multifocal inflammatory lesions disseminated throughout the white and grey matter of the CNS, which can be observed and quantified in the magnetic resonance (MR) scans. The proposed study will address the critical unmet need of computer-assisted extraction and assessment of prognostic factors based from an individual patient's brain MR scan, such as lesion count, volume, whole-brain and regional brain atrophy, and atrophied lesion volume, in order to evaluate the capability for personalized future disability progression prediction.

Study Type

OBSERVATIONAL

Enrollment

654

Conditions

Multiple Sclerosis Multiple Sclerosis Lesion Multiple Sclerosis Brain Lesion

Eligibility

Sex: ALLMin age: 18 YearsMax age: 65 YearsHealthy volunteers:

Medical Language ↔ Plain English

Inclusion Criteria: * persons diagnosed with MS (any phenotype; according to the 2010 McDonald criteria) and CIS patients * availability of at least two MRI exams with both FLAIR and T1-weighted scans of the same participant over a period of at least 6 months at the most recent examination * availability of demographic, clinical data and treatment information for the same participant over a period of at least 6 months at the most recent examination * availability of EDSS score and at least one previous EDSS scores for the same participant over a period of at least 6 months at the most recent examination Exclusion Criteria: * other clinically relevant systemic diseases if the researcher considers them to be significant

Locations (4)

University medical center Ljubljana

Ljubljana, Osrednjeslovenska, Slovenia

General and teaching hospital Celje

Celje, Slovenia

General hospital Izola

Izola, Slovenia

University medical center Maribor

Maribor, Slovenia

Outcomes

Primary Outcomes

Atrophied lesion volume derived from MRI predicts confirmed EDSS disability progression

Patients will be divided into two groups based on the presence or absence of EDSS disability progression (DP) during the observation period. The DP converters will be classified as patients with an EDSS change of at least 1.5 if the baseline EDSS is less than 1.0, those with an EDSS change of at least 1.0 if the baseline EDSS is 1.0-5.5, and those with an EDSS change of at least 0.5 if the baseline EDSS is 5.5 or higher \[15\]. DP converters should have confirmed progression of EDSS impairment over a period of at least 6 months. DP non-converters include individuals who do not meet the criteria for conversion. Atrophied lesion volume will be quantified from MR scans taken \>6 months prior to the observed EDSS increase. Advanced artificial intelligence based image analysis tools will be applied to assess the atrophied lesion volume.

Time frame: Atrophied lesion volume quantified from two or more MR scans across the span of at least one and up to five years

Secondary Outcomes

Atrophied lesion volume derived from MRI predicts conversion to secondary progressive multiple sclerosis

Patients will be divided into two groups, i.e. those who transitioned from clinically isolate syndrome (CIS) or relapsing-remitting (RR) to secondary progressive (SP) form of MS and those who were diagnosed with CIS/RRMS during the observation period. A consilium for patients with MS will confirm the SPMS diagnosis by consensus. Atrophied lesion volume will be quantified from MR scans taken \>6 months prior to the observed conversion to the SPMS. Advanced artificial intelligence based image analysis tools will be applied to assess the atrophied lesion volume.

Time frame: Atrophied lesion volume quantified from two or more MR scans across the span of least one and up to five years