Study Design and Population

The study included 34 subjects (mean age 38 ± 9 years; 13 males), comprising 17 MS patients and 17 controls matched by age and sex, all recruited from August 1, 2022, to October 31, 2022. Subjects were included if they were: over 18 years old; diagnosed with MS according to the revised McDonald criteria ¹ and had an RRMS disease course (for MS patients); or volunteered for the study and had no history of known neurological disease or structural brain abnormality (for controls). Exclusion criteria comprised of: contraindications to MRI (such as metallic implants, claustrophobia, or pregnancy), other neurological disease, brain injury, substance abuse, major psychiatric illnesses, or datasets with significant artifacts. For patients, Expanded Disability Status Scale (EDSS) scores and disease duration were documented, as well as the use of Disease Modifying Therapies (DMT) within the previous year.

Image Processing and Analysis

MRI scanning was conducted using a 3.0T Achieva scanner (Philips Healthcare, Best, the Netherlands with an eight-channel head coil for signal reception and a body coil for signal transmission. The imaging protocol comprised of 3D T1-weighted MP-RAGE, 2D T2-weighted turbo spin echo (TSE), 3D T2-weighted FLAIR, 2D DWI with b-values 0, 500, and 1000, vendor-supplied 2D B1 mapping, and the proposed 3D fast MPF-SL acquisition. ADC maps were created using the scanner's software, based on a mono-exponential model. MPF maps were produced using custom MATLAB scripts with predefined parameters (T_2b=10μs, and k_ba=23s^-1 for white matter and 40 s^-1 for grey matter).

For the proposed approach of off-resonance spin-lock used for rapid MPF mapping, MPF mapping can be calculated using two magnitude images M⁽¹⁾ and M⁽²⁾ acquired at off-resonance spin-lock using (ω₁⁽¹⁾, Δω⁽¹⁾) and (ω₁⁽²⁾, Δω⁽²⁾), respectively:  MPF=-log(M⁽²⁾/ M⁽¹⁾)/TSL, where TSL represents time of spin-lock. The spin-lock pulse was performed with the following parameters: ω₁⁽¹⁾/2π=100Hz, ω₁⁽²⁾/2π=500Hz, Δω⁽¹⁾/2π=800Hz, Δω⁽²⁾/2π=4000Hz, and TSL = 50 ms.

Image Processing and Analysis

MRI scanning was conducted using a 3.0T Achieva scanner (Philips Healthcare, Best, the Netherlands with an eight-channel head coil for signal reception and a body coil for signal transmission. The imaging protocol comprised of 3D T1-weighted MP-RAGE, 2D T2-weighted turbo spin echo (TSE), 3D T2-weighted FLAIR, 2D DWI with b-values 0, 500, and 1000, vendor-supplied 2D B1 mapping, and the proposed 3D fast MPF-SL acquisition. ADC maps were created using the scanner's software, based on a mono-exponential model. MPF maps were produced using custom MATLAB scripts with predefined parameters (T_2b=10μs, and k_ba=23s^-1 for white matter and 40 s^-1 for grey matter).

For the proposed approach of off-resonance spin-lock used for rapid MPF mapping, MPF mapping can be calculated using two magnitude images M⁽¹⁾ and M⁽²⁾ acquired at off-resonance spin-lock using (ω₁⁽¹⁾, Δω⁽¹⁾) and (ω₁⁽²⁾, Δω⁽²⁾), respectively:  MPF=-log(M⁽²⁾/ M⁽¹⁾)/TSL, where TSL represents time of spin-lock. The spin-lock pulse was performed with the following parameters: ω₁⁽¹⁾/2π=100Hz, ω₁⁽²⁾/2π=500Hz, Δω⁽¹⁾/2π=800Hz, Δω⁽²⁾/2π=4000Hz, and TSL = 50 ms.

Statistical Methods

Patients' MPF values in NAWM, NAGM, and NADGM were compared to those of controls using Mann-Whitney U and t-tests. Percentage differences and effect sizes (Cohen’s d) were computed to assess changes in MPF. Regional MPF and ADC values were compared using t-tests, Mann-Whitney U tests, and Wilcoxon signed-rank tests. The relationship between MPF and ADC in NAWM was evaluated using Pearson’s correlation. Statistical significance was set at a p-value of less than 0.05. All analyses were conducted using MedCalc [14] (MedCalc Software V20.211, https://www.medcalc.org/).