Limits...
MRI-based diagnostic biomarkers for early onset pediatric multiple sclerosis.

Weygandt M, Hummel HM, Schregel K, Ritter K, Allefeld C, Dommes E, Huppke P, Haynes JD, Wuerfel J, Gärtner J - Neuroimage Clin (2014)

Bottom Line: As expected, maximal diagnostic information for distinguishing PMS patients and HC was found in a periventricular WM area containing lesions (87.1% accuracy, p < 2.2 × 10(-5)).Taken together, we were able to identify biomarkers reflecting pathognomonic processes specific for MS patients with very early onset.Especially GM involvement in the separation between PMS subgroups suggests that conventional MRI contains a richer set of diagnostically informative features than previously assumed.

View Article: PubMed Central - PubMed

Affiliation: Bernstein Center for Computational Neuroscience Berlin, Charité - Universitätsmedizin, Berlin, Germany ; NeuroCure Clinical Research Center, Charité - Universitätsmedizin Berlin, Germany.

ABSTRACT
Currently, it is unclear whether pediatric multiple sclerosis (PMS) is a pathoetiologically homogeneous disease phenotype due to clinical and epidemiological differences between early and late onset PMS (EOPMS and LOPMS). Consequently, the question was raised whether diagnostic guidelines need to be complemented by specific EOPMS markers. To search for such markers, we analyzed cerebral MRI images acquired with standard protocols using computer-based classification techniques. Specifically, we applied classification algorithms to gray (GM) and white matter (WM) tissue probability parameters of small brain regions derived from T2-weighted MRI images of EOPMS patients (onset <12 years), LOPMS patients (onset ≥12 years), and healthy controls (HC). This was done for PMS subgroups matched for disease duration and participant age independently. As expected, maximal diagnostic information for distinguishing PMS patients and HC was found in a periventricular WM area containing lesions (87.1% accuracy, p < 2.2 × 10(-5)). MRI-based biomarkers specific for EOPMS were identified in prefrontal cortex. Specifically, a coordinate in middle frontal gyrus contained maximal diagnostic information (77.3%, p = 1.8 × 10(-4)). Taken together, we were able to identify biomarkers reflecting pathognomonic processes specific for MS patients with very early onset. Especially GM involvement in the separation between PMS subgroups suggests that conventional MRI contains a richer set of diagnostically informative features than previously assumed.

Show MeSH

Related in: MedlinePlus

Brain areas with diagnostic information. Gray matter (GM) areas with significant diagnostic information for the separation of a) early onset pediatric MS (EOPMS) and late onset pediatric MS (LOPMS) patients. c) EOPMS and healthy control (HC) subjects, e) LOPMS and HC. White matter (WM) areas with diagnostic information for the separation of b) EOPMS and LOPMS patients, d) EOPMS and HC, and f) LOPMS and HC. Significant coordinates are shown superimposed on subjects’ mean T2-weighted MRI image co-registered to the standard space of the Montreal Neurological Institute (MNI) brain template (Tzourio-Mazoyer et al., 2002). In panels a, c, and e, brain area abbreviations refer to the GM region a given coordinate is located in as determined by the MNI brain atlas (Tzourio-Mazoyer et al., 2002). In panels b, d, and f, brain area abbreviations in brackets refer to the GM region with the closest Euclidean distance to a given WM coordinate as determined by the MNI brain atlas. As accuracy measure, we computed the root mean square error (RMSE) between class labels and predicted class membership probability and denoted it beneath area abbreviations. For illustrative purposes, we also report the mean of sensitivity and specificity (MSS) as alternative accuracy measure beneath RMSE scores. Indices beneath axial brain slices report the z-coordinate in MNI space. Slices are displayed in neurological orientation. Brain area abbreviations: ACC, anterior cingulate cortex; ANG, angular gyrus; CAL, calcarine fissure, CDN, caudate nucleus; CER, cerebellum; CUN, cuneus; HIP, hippocampus; INS, insular cortex; IPG, inferior parietal gyrus; MCC, middle cingulated gyrus; MSF, medial superior frontal gyrus; MTG, middle temporal gyrus; OIF, orbital inferior frontal gyrus; OMF, orbital middle frontal gyrus; OSF, orbital superior frontal gyrus; PAL, pallidum; PCC, posterior cingulated gyrus; PCG, postcentral gyrus; PRC, precuneus; PUT, putamen; ROP, rolandic operculum; SFG, superior frontal gyrus; THA, thalamus.
© Copyright Policy - CC BY-NC-ND
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4310929&req=5

fig2: Brain areas with diagnostic information. Gray matter (GM) areas with significant diagnostic information for the separation of a) early onset pediatric MS (EOPMS) and late onset pediatric MS (LOPMS) patients. c) EOPMS and healthy control (HC) subjects, e) LOPMS and HC. White matter (WM) areas with diagnostic information for the separation of b) EOPMS and LOPMS patients, d) EOPMS and HC, and f) LOPMS and HC. Significant coordinates are shown superimposed on subjects’ mean T2-weighted MRI image co-registered to the standard space of the Montreal Neurological Institute (MNI) brain template (Tzourio-Mazoyer et al., 2002). In panels a, c, and e, brain area abbreviations refer to the GM region a given coordinate is located in as determined by the MNI brain atlas (Tzourio-Mazoyer et al., 2002). In panels b, d, and f, brain area abbreviations in brackets refer to the GM region with the closest Euclidean distance to a given WM coordinate as determined by the MNI brain atlas. As accuracy measure, we computed the root mean square error (RMSE) between class labels and predicted class membership probability and denoted it beneath area abbreviations. For illustrative purposes, we also report the mean of sensitivity and specificity (MSS) as alternative accuracy measure beneath RMSE scores. Indices beneath axial brain slices report the z-coordinate in MNI space. Slices are displayed in neurological orientation. Brain area abbreviations: ACC, anterior cingulate cortex; ANG, angular gyrus; CAL, calcarine fissure, CDN, caudate nucleus; CER, cerebellum; CUN, cuneus; HIP, hippocampus; INS, insular cortex; IPG, inferior parietal gyrus; MCC, middle cingulated gyrus; MSF, medial superior frontal gyrus; MTG, middle temporal gyrus; OIF, orbital inferior frontal gyrus; OMF, orbital middle frontal gyrus; OSF, orbital superior frontal gyrus; PAL, pallidum; PCC, posterior cingulated gyrus; PCG, postcentral gyrus; PRC, precuneus; PUT, putamen; ROP, rolandic operculum; SFG, superior frontal gyrus; THA, thalamus.

Mentions: In this analysis, we searched for differential diagnostic information for separation of EOPMS vs. LOPMS patients contained in modulated tissue probability parameters extracted from GM (WM) areas in classification Analysis 1a (1b). Analysis 1a primarily revealed frontal and parietal GM areas as being diagnostically informative. Consistently, WM Analysis 1b revealed diagnostic information in WM areas located in the vicinity of frontal gyri. Please see Table 2 and Fig. 2a and b for details. Please see Table S2 for results for this group comparison based on PMS subgroups matched for age, gender and lesion load.


MRI-based diagnostic biomarkers for early onset pediatric multiple sclerosis.

Weygandt M, Hummel HM, Schregel K, Ritter K, Allefeld C, Dommes E, Huppke P, Haynes JD, Wuerfel J, Gärtner J - Neuroimage Clin (2014)

Brain areas with diagnostic information. Gray matter (GM) areas with significant diagnostic information for the separation of a) early onset pediatric MS (EOPMS) and late onset pediatric MS (LOPMS) patients. c) EOPMS and healthy control (HC) subjects, e) LOPMS and HC. White matter (WM) areas with diagnostic information for the separation of b) EOPMS and LOPMS patients, d) EOPMS and HC, and f) LOPMS and HC. Significant coordinates are shown superimposed on subjects’ mean T2-weighted MRI image co-registered to the standard space of the Montreal Neurological Institute (MNI) brain template (Tzourio-Mazoyer et al., 2002). In panels a, c, and e, brain area abbreviations refer to the GM region a given coordinate is located in as determined by the MNI brain atlas (Tzourio-Mazoyer et al., 2002). In panels b, d, and f, brain area abbreviations in brackets refer to the GM region with the closest Euclidean distance to a given WM coordinate as determined by the MNI brain atlas. As accuracy measure, we computed the root mean square error (RMSE) between class labels and predicted class membership probability and denoted it beneath area abbreviations. For illustrative purposes, we also report the mean of sensitivity and specificity (MSS) as alternative accuracy measure beneath RMSE scores. Indices beneath axial brain slices report the z-coordinate in MNI space. Slices are displayed in neurological orientation. Brain area abbreviations: ACC, anterior cingulate cortex; ANG, angular gyrus; CAL, calcarine fissure, CDN, caudate nucleus; CER, cerebellum; CUN, cuneus; HIP, hippocampus; INS, insular cortex; IPG, inferior parietal gyrus; MCC, middle cingulated gyrus; MSF, medial superior frontal gyrus; MTG, middle temporal gyrus; OIF, orbital inferior frontal gyrus; OMF, orbital middle frontal gyrus; OSF, orbital superior frontal gyrus; PAL, pallidum; PCC, posterior cingulated gyrus; PCG, postcentral gyrus; PRC, precuneus; PUT, putamen; ROP, rolandic operculum; SFG, superior frontal gyrus; THA, thalamus.
© Copyright Policy - CC BY-NC-ND
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4310929&req=5

fig2: Brain areas with diagnostic information. Gray matter (GM) areas with significant diagnostic information for the separation of a) early onset pediatric MS (EOPMS) and late onset pediatric MS (LOPMS) patients. c) EOPMS and healthy control (HC) subjects, e) LOPMS and HC. White matter (WM) areas with diagnostic information for the separation of b) EOPMS and LOPMS patients, d) EOPMS and HC, and f) LOPMS and HC. Significant coordinates are shown superimposed on subjects’ mean T2-weighted MRI image co-registered to the standard space of the Montreal Neurological Institute (MNI) brain template (Tzourio-Mazoyer et al., 2002). In panels a, c, and e, brain area abbreviations refer to the GM region a given coordinate is located in as determined by the MNI brain atlas (Tzourio-Mazoyer et al., 2002). In panels b, d, and f, brain area abbreviations in brackets refer to the GM region with the closest Euclidean distance to a given WM coordinate as determined by the MNI brain atlas. As accuracy measure, we computed the root mean square error (RMSE) between class labels and predicted class membership probability and denoted it beneath area abbreviations. For illustrative purposes, we also report the mean of sensitivity and specificity (MSS) as alternative accuracy measure beneath RMSE scores. Indices beneath axial brain slices report the z-coordinate in MNI space. Slices are displayed in neurological orientation. Brain area abbreviations: ACC, anterior cingulate cortex; ANG, angular gyrus; CAL, calcarine fissure, CDN, caudate nucleus; CER, cerebellum; CUN, cuneus; HIP, hippocampus; INS, insular cortex; IPG, inferior parietal gyrus; MCC, middle cingulated gyrus; MSF, medial superior frontal gyrus; MTG, middle temporal gyrus; OIF, orbital inferior frontal gyrus; OMF, orbital middle frontal gyrus; OSF, orbital superior frontal gyrus; PAL, pallidum; PCC, posterior cingulated gyrus; PCG, postcentral gyrus; PRC, precuneus; PUT, putamen; ROP, rolandic operculum; SFG, superior frontal gyrus; THA, thalamus.
Mentions: In this analysis, we searched for differential diagnostic information for separation of EOPMS vs. LOPMS patients contained in modulated tissue probability parameters extracted from GM (WM) areas in classification Analysis 1a (1b). Analysis 1a primarily revealed frontal and parietal GM areas as being diagnostically informative. Consistently, WM Analysis 1b revealed diagnostic information in WM areas located in the vicinity of frontal gyri. Please see Table 2 and Fig. 2a and b for details. Please see Table S2 for results for this group comparison based on PMS subgroups matched for age, gender and lesion load.

Bottom Line: As expected, maximal diagnostic information for distinguishing PMS patients and HC was found in a periventricular WM area containing lesions (87.1% accuracy, p < 2.2 × 10(-5)).Taken together, we were able to identify biomarkers reflecting pathognomonic processes specific for MS patients with very early onset.Especially GM involvement in the separation between PMS subgroups suggests that conventional MRI contains a richer set of diagnostically informative features than previously assumed.

View Article: PubMed Central - PubMed

Affiliation: Bernstein Center for Computational Neuroscience Berlin, Charité - Universitätsmedizin, Berlin, Germany ; NeuroCure Clinical Research Center, Charité - Universitätsmedizin Berlin, Germany.

ABSTRACT
Currently, it is unclear whether pediatric multiple sclerosis (PMS) is a pathoetiologically homogeneous disease phenotype due to clinical and epidemiological differences between early and late onset PMS (EOPMS and LOPMS). Consequently, the question was raised whether diagnostic guidelines need to be complemented by specific EOPMS markers. To search for such markers, we analyzed cerebral MRI images acquired with standard protocols using computer-based classification techniques. Specifically, we applied classification algorithms to gray (GM) and white matter (WM) tissue probability parameters of small brain regions derived from T2-weighted MRI images of EOPMS patients (onset <12 years), LOPMS patients (onset ≥12 years), and healthy controls (HC). This was done for PMS subgroups matched for disease duration and participant age independently. As expected, maximal diagnostic information for distinguishing PMS patients and HC was found in a periventricular WM area containing lesions (87.1% accuracy, p < 2.2 × 10(-5)). MRI-based biomarkers specific for EOPMS were identified in prefrontal cortex. Specifically, a coordinate in middle frontal gyrus contained maximal diagnostic information (77.3%, p = 1.8 × 10(-4)). Taken together, we were able to identify biomarkers reflecting pathognomonic processes specific for MS patients with very early onset. Especially GM involvement in the separation between PMS subgroups suggests that conventional MRI contains a richer set of diagnostically informative features than previously assumed.

Show MeSH
Related in: MedlinePlus