Sensitive Detection of Infratentorial and Upper Cervical Cord Lesions in Multiple Sclerosis with Combined 3D FLAIR and T2-Weighted (FLAIR3) Imaging

Wernicke Encephalopathy: Typical and Atypical Findings in Alcoholics and Non-Alcoholics and Correlation with Clinical Symptoms

PURPOSE

Clinical diagnosis of Wernicke encephalopathy (WE) can be challenging due to incomplete presentation of the classical triad. The aim was to provide an update on the relevance of standard MRI and to put typical and atypical imaging findings into context with clinical features.

METHODS

In this two-center retrospective observational study, the local radiology information system was searched for consecutive patients with clinical or imaging suspicion of WE. Two independent raters evaluated T2-weighted imaging (WI), fluid-attenuation inversion recovery (FLAIR), diffusion WI (DWI), T2*WI and/or susceptibility WI (SWI), and contrast-enhanced (CE)-T1WI, and noted the involvement of typical (i.e., mammillary bodies (MB), periaqueductal grey (PAG), thalamus, hypothalamus, tectal plate) and atypical (all others) lesion sites. Unusual signal patterns like hemorrhages were also documented. Reported clinical features together with the diagnostic criteria of the latest guidelines of the European Federation of Neurological Societies (EFNS) were used to test for relationships with MRI biomarkers.

RESULTS

47 patients with clinically confirmed WE were included (Jan '99-Apr '23; mean age, 53 yrs; 70% males). Interrater reliability for imaging findings was substantial (κ = 0.71), with lowest agreements for T2WI (κ = 0.85) compared to all other sequences and for PAG (κ = 0.65) compared to all other typical regions. In consensus, 77% (n = 36/47) of WE cases were rated MRI positive, with FLAIR (n = 36/47, 77%) showing the strongest relation (χ2 = 47.0; P < 0.001) compared to all other sequences. Microbleeds in the MB were detected in four out of ten patients who received SWI, not visible on corresponding T2*WI. Atypical findings were observed in 23% (n = 11/47) of cases, always alongside typical findings, in both alcoholics (n = 9/44, 21%) and non-alcoholics (n = 2/3, 67%). Isolated involvement of structures, explicitly PAG (n = 4/36; 11%) or MB (n = 1/36; 3%), was present but observed less frequently than combined lesions (n = 31/36; 86%). A cut-off width of 2.5 mm for the PAG on 2D axial FLAIR was established between cases and age- and sex-matched controls. An independent association was demonstrated only between short-term memory loss and changes in the MB (OR = 2.2 [95% CI: 1.1-4.5]; P = 0.024). In retrospect, EFNS criteria were positive (≥ 2 out of 4) in every case, but its count (range, 2-4) showed no significant (P = 0.427) relationship with signal changes on standard MRI.

CONCLUSION

The proposed sequence protocol (FLAIR, DWI, SWI and T1WI + CE) yielded good detection rates for neuroradiological findings in WE, with SWI showing microbleeds in the MB with superior detectability. However, false negative results in about a quarter of cases underline the importance of neurological alertness for the diagnosis. Awareness of atypical MRI findings should be raised, not only in non-alcoholics. There is limited correlation between clinical signs and standard MRI biomarkers.

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

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.

Infratentorial lesions in multiple sclerosis patients: intra- and inter-rater variability in comparison to a fully automated segmentation using 3D convolutional neural networks

OBJECTIVE

Automated quantification of infratentorial multiple sclerosis lesions on magnetic resonance imaging is clinically relevant but challenging. To overcome some of these problems, we propose a fully automated lesion segmentation algorithm using 3D convolutional neural networks (CNNs).

METHODS

The CNN was trained on a FLAIR image alone or on FLAIR and T1-weighted images from 1809 patients acquired on 156 different scanners. An additional training using an extra class for infratentorial lesions was implemented. Three experienced raters manually annotated three datasets from 123 MS patients from different scanners.

RESULTS

The inter-rater sensitivity (SEN) was 80% for supratentorial lesions but only 62% for infratentorial lesions. There was no statistically significant difference between the inter-rater SEN and the SEN of the CNN with respect to the raters. For supratentorial lesions, the CNN featured an intra-rater intra-scanner SEN of 0.97 (R1 = 0.90, R2 = 0.84) and for infratentorial lesion a SEN of 0.93 (R1 = 0.61, R2 = 0.73).

CONCLUSION

The performance of the CNN improved significantly for infratentorial lesions when specifically trained on infratentorial lesions using a T1 image as an additional input and matches the detection performance of experienced raters. Furthermore, for infratentorial lesions the CNN was more robust against repeated scans than experienced raters.

KEY POINTS

• A 3D convolutional neural network was trained on MRI data from 1809 patients (156 different scanners) for the quantification of supratentorial and infratentorial multiple sclerosis lesions. • Inter-rater variability was higher for infratentorial lesions than for supratentorial lesions. The performance of the 3D convolutional neural network (CNN) improved significantly for infratentorial lesions when specifically trained on infratentorial lesions using a T1 image as an additional input. • The detection performance of the CNN matches the detection performance of experienced raters.