A MS‐lesion pattern discrimination plot based on geostatistics

Abstract

IntroductionA geostatistical approach to characterize MS‐lesion patterns based on their geometrical properties is presented.MethodsA dataset of 259 binary MS‐lesion masks in MNI space was subjected to directional variography. A model function was fit to express the observed spatial variability in x, y, z directions by the geostatistical parameters Range and Sill.ResultsParameters Range and Sill correlate with MS‐lesion pattern surface complexity and total lesion volume. A scatter plot of ln(Range) versus ln(Sill), classified by pattern anisotropy, enables a consistent and clearly arranged presentation of MS‐lesion patterns based on geometry: the so‐called MS‐Lesion Pattern Discrimination Plot.ConclusionsThe geostatistical approach and the graphical representation of results are considered efficient exploratory data analysis tools for cross‐sectional, follow‐up, and medication impact analysis.

Geostatistical Analysis of White Matter Lesions in Multiple Sclerosis Identifies Gender Differences in Lesion Evolution

“…Closer inspection of Figure 2 reveals that F points markedly overbalance M points along the lower fringe of the point cloud. In the LDP, MS-WML patterns that plot near the lower fringe of the cloud typically are dominated by relatively few, extended lesions while patterns that plot near the upper fringe of the point cloud are characterized by many small lesions or complexly shaped MS-WML aggregates (Marschallinger et al, 2016; also check pattern 238 against pattern 518 in Figure 1A).…”

“…The surface complexness of biological structures (like MS-WML patterns) is conveniently expressed as the ratio of surface area to enclosed volume (Schmidt-Nielson, 1984). For a binary MS-WML pattern, the variogram model range a is proxy of MS-WML pattern surface smoothness and the variogram model sill c is substitute of total lesion volume (TLV) (Marschallinger et al, 2016): the higher a , the higher is the overall spatial correlation and the smoother (i.e., less geometrically complex) is the MS-WML pattern’s surface; the higher c , the higher is TLV. Compare Figure 1A for examples – pattern 518 is “complex,” pattern 238 is “smooth.” The MS-Lesion Discrimination Plot (LDP) aims at mapping above 3D variography data to a dimension-reduced, well-arranged 2D space spanned by A on the abscissa, and C on the ordinate.…”

“…A combined usage of novel imaging techniques and spatial-statistics models may provide further insights to the pathophysiological background of gender differences in disease course and lesion evolution. Based on geostatistical metrics (Marschallinger et al, 2016), we established a methodology to study MS-WML pattern characteristics including lesion pattern morphology. Here, we tested our geostatistical methodology in a longitudinal group consisting of men and women in order to evaluate gender-related changes and differences in MS-WML evolution.…”

A R-Script for Generating Multiple Sclerosis Lesion Pattern Discrimination Plots

“…Below, the rationale behind and the making of the MS-Lesion Pattern Discrimination Plot (MS-LDP) is reviewed in compact form. For a more in-depth discussion of the clinical background, especially the application of the MS-LDP to real-world data sets, see [ 5 , 6 , 7 ]. As an example of MS-LDP graphics depicting a larger cohort of patients with MS that was processed with LDPgenerator.r , see LDP_Supplement.jpg in Appendix A (from [ 7 ], where also an interpretation of this MS-LDP can be found).…”

“…Variography, a core method of classical geostatistics, proved suitable for explorative data analysis (EDA) of MS-White Matter Lesion patterns (MS-WML) and for extracting quantitative spatial-statistics metrics on MS-WML. The MS-Lesion Pattern Discrimination Plot (MS-LDP) summarizes these metrics in a clear and standardized form, to aid in follow-up, cross-sectional and medication impact analysis [ 5 , 6 ]. With the aid of the MS-LDP, significantly different evolution of MS-lesion patterns could be disclosed between male and female early MS cohorts [ 7 ].…”