Multicenter Evaluation of AI-generated DIR and PSIR for Cortical and Juxtacortical Multiple Sclerosis Lesion Detection

Abstract

Background Cortical multiple sclerosis lesions are clinically relevant but inconspicuous at conventional clinical MRI. Double inversion recovery (DIR) and phase-sensitive inversion recovery (PSIR) are more sensitive but often unavailable. In the past 2 years, artificial intelligence (AI) was used to generate DIR and PSIR from standard clinical sequences (eg, T1-weighted, T2-weighted, and fluid-attenuated inversion-recovery sequences), but multicenter validation is crucial for further implementation. Purpose To evaluate cortical and juxtacortical multiple sclerosis lesion detection for diagnostic and disease monitoring purposes on AI-generated DIR and PSIR images compared with MRI-acquired DIR and PSIR images in a multicenter setting. Materials and Methods Generative adversarial networks were used to generate AI-based DIR ( n = 50) and PSIR ( n = 43) images. The number of detected lesions between AI-generated images and MRI-acquired (reference) images was compared by randomized blinded scoring by seven readers (all with >10 years of experience in lesion assessment). Reliability was expressed as the intraclass correlation coefficient (ICC). Differences in lesion subtype were determined using Wilcoxon signed-rank tests. Results MRI scans of 202 patients with multiple sclerosis (mean age, 46 years ± 11 [SD]; 127 women) were retrospectively collected from seven centers (February 2020 to January 2021). In total, 1154 lesions were detected on AI-generated DIR images versus 855 on MRI-acquired DIR images (mean difference per reader, 35.0% ± 22.8; P < .001). On AI-generated PSIR images, 803 lesions were detected versus 814 on MRI-acquired PSIR images (98.9% ± 19.4; P = .87). Reliability was good for both DIR (ICC, 0.81) and PSIR (ICC, 0.75) across centers. Regionally, more juxtacortical lesions were detected on AI-generated DIR images than on MRI-acquired DIR images (495 [42.9%] vs 338 [39.5%]; P < .001). On AI-generated PSIR images, fewer juxtacortical lesions were detected than on MRI-acquired PSIR images (232 [28.9%] vs 282 [34.6%]; P = .02). Conclusion Artificial intelligence–generated double inversion-recovery and phase-sensitive inversion-recovery images performed well compared with their MRI-acquired counterparts and can be considered reliable in a multicenter setting, with good between-reader and between-center interpretative agreement. Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Zivadinov and Dwyer in this issue.

How far MS lesion detection and segmentation are integrated into the clinical workflow? A systematic review

“… lesion load, SI, Jaccard index, FPF, TPF 1st, 2nd ( Cerasa et al, 2011 ) 11 MS FLAIR cellular neural network skull stripping DSC, total lesion load 1st, 2nd ( Kuwazuru et al, 2011 ) 3 MS T1-w, T2-w, FLAIR SVM and artificial neural network enhancement by subtraction of background accuracy, SI, sensitivity, number of FP 1st, 2nd ( Bilello et al, 2013 ) 88 MS FLAIR image subtraction skull stripping, b.c. with and without CAD: lesion count and location, PPV, NPV, sensitivity, specificity, efficiency, AUC, lesion-wise sensitivity, FPR and PPV, time spent, clinical reporting 1st, 2nd, 3rd, 4th, 5th ( Hindsholm et al, 2021 ) 93 MS FLAIR 2D CNN normalisation, cropping or zero padding DSC, recall, F1, precision, qualitative assessment of output lesion masks 1st, 2nd ( Roy et al, 2015 ) 10 MS MPRAGE and FLAIR patch based with temporal information from timepoints normalisation DSC, LTPR, LFPR, AVD 1st, 2nd ( Bouman et al, 2023 ) 198 MS DIR or PSIR generated artificially from T1-w, T2-w, PD-w or FLAIR U-Net-like MNI reg., skull stripping, b.c. lesions count, precision 1st, 2nd ( Sitter et al, 201...…”

“…In six cases, the only input provided to the network were either T2-w images ( Abhale et al, 2022 , Yildirim and Dandil, 2021a ), MPRAGE (Magnetisation-prepared rapid gradient echo) ( Galimzianova et al, 2015 , Spies et al, 2013 ), MP2RAGE (Magnetisation-prepared 2 rapid gradient echo) ( Fartaria et al, 2019 ) or MR fingerprinting EPI (Echo-planar imaging) ( Hermann et al, 2021 ). Less common contrasts, such as diffusion basis spectrum imaging ( Ye et al, 2020 ), DIR ( Fartaria et al, 2015 , Schläger et al, 2022 , Bouman et al, 2023 ) and PSIR ( Bouman et al, 2023 ) were also adopted.…”

Advanced MRI Techniques: Diagnosis and Follow-Up of Multiple Sclerosis

“…Moreover, CLs are primarily visible on advanced MRI sequences at high (3T) and ultra-high (7T) magnetic fields [27,28], which are not routinely available in clinics due to their high cost and their use is often limited to the research practice. Pilot studies leveraging artificial intelligence (AI) have recently facilitated the generation of synthetic DIR images from standard clinical MRI protocols (i.e., combinations of T1and proton density-/T2-weighted sequences) [38]. CL volume may be an additional outcome measure relative to WM lesion burden in clinical trials, given its potential for correlating with EDSS.…”

“…The number of CLs changes according to the phase of the disease, as they are more prevalent in SPMS compared to CIS or RRMS [ 37 ], with an impact on disability and cognitive impairment, thereby being vital indicators of disease progression in MS patients [ 27 , 28 ]. However, due to their small size and low contrast relative to normal grey matter, cortical lesions are inconspicuous on images acquired using conventional clinical MRI sequences (e.g., T1w, T2w, or FLAIR) [ 38 ]. Advanced MRI techniques, including DIR, MP2RAGE, and phase-sensitive inversion–recovery (PSIR) sequences, have significantly improved the visualization of cortical lesions, the latter appearing as a hyperintense signal (in DIR) or as a hypointense lesions (MP2RAGE and PSIR represented in Figure 2 ); however, such techniques find limited use in routine diagnostic and clinical trial evaluation protocols due to their prolonged acquisition times (i.e., 10–15 min) [ 31 , 34 , 38 ].…”

“…However, due to their small size and low contrast relative to normal grey matter, cortical lesions are inconspicuous on images acquired using conventional clinical MRI sequences (e.g., T1w, T2w, or FLAIR) [ 38 ]. Advanced MRI techniques, including DIR, MP2RAGE, and phase-sensitive inversion–recovery (PSIR) sequences, have significantly improved the visualization of cortical lesions, the latter appearing as a hyperintense signal (in DIR) or as a hypointense lesions (MP2RAGE and PSIR represented in Figure 2 ); however, such techniques find limited use in routine diagnostic and clinical trial evaluation protocols due to their prolonged acquisition times (i.e., 10–15 min) [ 31 , 34 , 38 ]. Moreover, CLs are primarily visible on advanced MRI sequences at high (3T) and ultra-high (7T) magnetic fields [ 27 , 28 ], which are not routinely available in clinics due to their high cost and their use is often limited to the research practice.…”