Fully Automated Detection of Paramagnetic Rims in Multiple Sclerosis Lesions on 3T Susceptibility-Based MR Imaging

Abstract

Background and PurposeThe presence of a paramagnetic rim around a white matter lesion has recently been shown to be a hallmark of a particular pathological type of multiple sclerosis (MS) lesion. Increased prevalence of these paramagnetic rim lesions (PRLs) is associated with a more severe disease course in MS. The identification of these lesions is time-consuming to perform manually. We present a method to automatically detect PRLs on 3T T2*-phase images.MethodsT1-weighted, T2-FLAIR, and T2*-phase MRI of the brain were collected at 3T for 19 subjects with MS. The images were then processed with lesion segmentation, lesion center detection, lesion labelling, and lesion-level radiomic feature extraction. A total of 877 lesions were identified, 118 (13%) of which contained a paramagnetic rim. We divided our data into a training set (15 patients, 673 lesions) and a testing set (4 patients, 204 lesions). We fit a random forest classification model on the training set and assessed our ability to classify lesions as PRL on the test set.ResultsThe number of PRLs per subject identified via our automated lesion labelling method was highly correlated with the gold standard count of PRLs per subject, r = 0.91 (95% CI [0.79, 0.97]). The classification algorithm using radiomic features can classify a lesion as PRL or not with an area under the curve of 0.80 (95% CI [0.67, 0.86]).ConclusionThis study develops a fully automated technique for the detection of paramagnetic rim lesions using standard T1 and FLAIR sequences and a T2*phase sequence obtained on 3T MR images.HighlightsA fully automated method for both the identification and classification of paramagnetic rim lesions is proposed.Radiomic features in conjunction with machine learning algorithms can accurately classify paramagnetic rim lesions.Challenges for classification are largely driven by heterogeneity between lesions, including equivocal rim signatures and lesion location.

QSMRim-Net: Imbalance-aware learning for identification of chronic active multiple sclerosis lesions on quantitative susceptibility maps

“…Our QSMRim-Net achieved better performance than two previously developed methods when applied to the QSM, APRL (RF) ( Lou et al, 2021 ) and Rim-Net ( Barquero et al, 2020 ). The results on a lesion-level were not found to be statistically significant, and this can be attributed to the small number of rim + lesions in the dataset.…”

“…QSMRim-Net outperformed the competitors in all metrics used for evaluation. With a slightly higher overall accuracy and specificity with other methods, QSMRim-Net resulted in a 9.8% and 23.3% improvement in F1 score, 3.5% and 14.3% improvement in sensitivity and 16.8% and 33.1% improvement in PPV compared to Rim-Net ( Barquero et al, 2020 ) and APRL (RF) ( Lou et al, 2021 ), respectively. The increase in accuracy and specificity were very moderate for QSMRim-Net compared to the other methods.…”

“…In this work, we used a binary classification of whether a lesion had a rim. As discussed in ( Lou et al, 2021 ) there are many factors that influence the strength of the rim + lesion signature on QSM and a more nuanced approach to classify these lesions may be beneficial.…”

“…RimNet ( Barquero et al, 2020 ) develops a multi-modal VGGNet ( Simonyan, 2014 ) to extract rim information from image patches of the phase and T2-FLAIR images. APRL ( Lou et al, 2021 ) applies a SMOTE and a random forest model to first-order radiomic features derived from individual lesions on the phase images. To evaluate QSMRim-Net, we compared the performance of the proposed algorithm with these two methods.…”

“…Two previous methods have been developed to identify chronic active MS lesions on phase imaging ( Barquero et al, 2020 , Lou et al, 2020 ). RimNet ( Barquero et al, 2020 ) uses convolutional features, while Automatic Paramagnetic Rim Lesions (APRL) ( Lou et al, 2021 ) uses radiomic features. To put the performance of QSMRim-Net into context, we compare it to these two methods applied to the QSM using both lesion-level and patient-level performance metrics.…”