An Automated Statistical Technique for Counting Distinct Multiple Sclerosis Lesions

Abstract

Background Lesion load is a common biomarker in multiple sclerosis, yet it has historically shown modest associations with clinical outcomes. Lesion count, which encapsulates the natural history of lesion formation and is thought to provide complementary information, is difficult to assess in patients with confluent (i.e. spatially overlapping) lesions. We introduce a statistical technique for cross-sectionally counting pathologically distinct lesions. Methods MRI is used to assess the probability of lesion at each location. The texture of this map is quantified using a novel technique, and clusters resembling the center of a lesion are counted. Validity compared to a gold-standard count was demonstrated in 60 subjects observed longitudinally, and reliability was determined using 14 scans of a clinically stable subject acquired at 7 sites. Results The proposed count and the gold-standard count were highly correlated with (r=.97, p<.001), and not significantly different (t59=−.83, p=.41), and the proposed count’s variability across repeated scans was equivalent to that of lesion load. Accounting for lesion load and age, lesion count was negatively associated (t58=−2.73, p<.01) with the Expanded Disability Status Scale (EDSS). Average lesion size had a higher association with EDSS (r=.35, p<.01) than lesion load (r=.10, p=.44) or lesion count (r=−.12, p=.36) alone. Conclusion This study introduces a novel technique for counting pathologically distinct lesions using cross-sectional data, and demonstrates its ability to recover obscured longitudinal information. The proposed count allows for more accurate estimation of lesion size, which correlated more closely with disability scores than either lesion load or lesion count alone.

Do All Patients with Multiple Sclerosis Benefit from the Use of Contrast on Serial Follow-Up MR Imaging? A Retrospective Analysis

“…We do not routinely use pulse sequences that maximize visualization of contrast enhancement, such as magnetization transfer or fat-suppressed T1-weighted postcontrast imaging mainly due to our choice of imaging all these patients at 3T, because that allows generation of high-quality FLAIR volume images for the postprocessing software to improve detection of progression. We also recognize that this type of software is not available at most imaging centers, but there are other, similar products available, 42 and we were intent on maximizing detection of progression for this study. We believe that this approach improves detection of new white matter lesions and may have contributed to our overall finding that enhancement of lesions was only evident when there was progression.…”

Automated Integration of Multimodal MRI for the Probabilistic Detection of the Central Vein Sign in White Matter Lesions

“…For the results presented in this paper, a threshold of 0.30 is applied to this probability map in order to create a binary lesion mask. The threshold of 0.30 was chosen because previous work has found it to be a conservative cutoff that can limit the amount of false positive lesion tissue 23,24 . Following the definition for a CVS+ lesion given by the NAIMS cooperative 19 , lesions detected by the MIMoSA model that are smaller than 3 mm in any plane are removed from candidacy.…”

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

“…After lesion segmentation masks were obtained, we used the lesion probability maps as input to a center detection method (38) to identify distinct lesions based on the texture of the lesion tissue. We then used a nearest-neighbor approach to classify the remainder of the lesion segmentation map into those identified lesions (Figure 1).…”