Aims. This project is designed to identify crucial clinical and conventional MRI features that can precisely forecast distinct multiple sclerosis disability progression profiles over time. Our primary goal is to characterise disability profiles through a data-driven methodology and identify the baseline clinical and MRI features with the most robust predictive potential for these disability profiles. Additionally, as a secondary objective, we aim to evaluate the impact of treatment strategies on these disability profiles. The patients to include, have to be provided with clinical evaluation (Expanded Disability Status Scale (EDSS) score rating, number of relapses, ongoing disease modifying treatment and MRI assessment (number of T2 hyperintense or gadolinium-enhancing lesions) at each available visit. Multivariate latent class mixed-effect models will be used to identify the disability profiles, while machine learning approces will be used to identify the predictors for these profiles. To our secondary aim, we will test and compare the effect of the different treatment strategies on the likelihood of belonging to a specific disability progression phenotypes.
Fondazione Italiana Sclerosi Multipla – FISM – Ente del Terzo Settore/ETS e, in forma abbreviata, FISM ETS. Iscrizione al RUNTS Rep. N° 89695 - Fondazione con Riconoscimento di Personalità Giuridica - C.F. 95051730109
Phenotyping progression in multiple sclerosis
Aims. This project is designed to identify crucial clinical and conventional MRI features that can precisely forecast distinct multiple sclerosis disability progression profiles over time. Our primary goal is to characterise disability profiles through a data-driven methodology and identify the baseline clinical and MRI features with the most robust predictive potential for these disability profiles. Additionally, as a secondary objective, we aim to evaluate the impact of treatment strategies on these disability profiles. The patients to include, have to be provided with clinical evaluation (Expanded Disability Status Scale (EDSS) score rating, number of relapses, ongoing disease modifying treatment and MRI assessment (number of T2 hyperintense or gadolinium-enhancing lesions) at each available visit. Multivariate latent class mixed-effect models will be used to identify the disability profiles, while machine learning approces will be used to identify the predictors for these profiles. To our secondary aim, we will test and compare the effect of the different treatment strategies on the likelihood of belonging to a specific disability progression phenotypes.
---
---
---