Based on this context, the NextMRI project will carry out a clinical evaluation study involving a multidisciplinary team of radiologists, rheumatologists, physiotherapists, data managers, and physicists. The study will focus on patients with suspected multiple sclerosis (MS). Its main goals are to:
1. Assess the diagnostic value of the low-field MRI prototype.
2. Evaluate usability for end users.
3. Train AI algorithms using deep learning to differentiate between healthy and damaged tissues, thereby aiding radiologists, reducing diagnostic time, and lightening their workload.
Radiologists will first generate structured reports based on standard high-resolution MRI scans. They will then evaluate the low-field MRI images from the NextMRI prototype. The initial high-field MRI reports will serve as the reference standard for assessing performance and training the AI system.
Each patient will undergo two MRI scans on the same day (after providing informed consent): one at the standard 3T MR scan and other at the low-field MRI scan (NextMRI). To prevent bias, both scans will be interpreted by a professional radiologist with an 8-week interval between readings.
A total of 50 patients aged between 18 and 65 years with suspected multiple sclerosis will be included. Patients with any contraindications for high-field MRI will be excluded from the study.
