Assessment of automatic decision-support systems for detecting active T2 lesions in multiple sclerosis patients.

BACKGROUND: Active (new/enlarging) T2 lesion counts are routinely used in the clinical management of multiple sclerosis. Thus, automated tools able to accurately identify active T2 lesions would be of high interest to neuroradiologists for assisting in their clinical activity. OBJECTIVE: To compare the accuracy in detecting active T2 lesions and of radiologically active patients based on different visual and automated methods. METHODS: One hundred multiple sclerosis patients underwent two magnetic resonance imaging examinations within 12 months. Four approaches were assessed for detecting active T2 lesions: (1) conventional neuroradiological reports; (2) prospective visual analyses performed by an expert; (3) automated unsupervised tool; and (4) supervised convolutional neural network. As a gold standard, a reference outcome was created by the consensus of two observers. RESULTS: The automated methods detected a higher number of active T2 lesions, and a higher number of active patients, but a higher number of false-positive active patients than visual methods. The convolutional neural network model was more sensitive in detecting active T2 lesions and active patients than the other automated method. CONCLUSION: Automated convolutional neural network models show potential as an aid to neuroradiological assessment in clinical practice, although visual supervision of the outcomes is still required.

Assessment of 2D conventional and synthetic MRI in multiple sclerosis.

PURPOSE: To qualitatively and quantitatively compare synthetic and conventional MRI sequences acquired on a 1.5-T system for patients with multiple sclerosis (MS). METHODS: Prospective study that involved twenty-seven consecutive relapsing-remitting MS patients scanned on a 1.5-T MRI scanner. The MRI protocol included 2D transverse conventional spin-echo sequences: proton density-weighted (PD), T2-weighted, T2-FLAIR, and T1-weighted. Synthetic images were generated using 2D transverse QRAPMASTER and SyMRI software with the same voxel size, repetition, echo, and inversion times as the conventional sequences. Four raters performed a crosstab qualitative analysis that involved evaluating global image quality, contrast, flow artefacts, and confidence in lesion assessment introducing the concepts of predominance, agreement, and disagreement. A quantitative analysis was also performed and included evaluating the number of lesions (periventricular, juxtacortical, brainstem, and cerebellum) and the contrast-to-noise ratio between regions (CSF, white matter, grey matter, lesions). RESULTS: The global image quality assessment showed predominance for better scores for conventional sequences over synthetic sequences, whereas contrast, confidence in lesion assessment, and flow artefacts showed predominance for agreement between sequences. There was predominance for disagreement between all pairs of raters in most of the evaluated qualitative parameters. Synthetic PD and T2-FLAIR images showed higher contrast-to-noise ratios than the corresponding conventional images for most comparison between regions. There were no significant differences in the number of lesions detected for most of the study regions between conventional and synthetic images. CONCLUSION: Synthetic MRI can be potentially used as an alternative to conventional brain MRI sequences in the assessment of MS.