MIMoSA: An Approach to Automatically Segment T2 Hyperintense and T1 Hypointense Lesions in Multiple Sclerosis

  • Alessandra M. ValcarcelEmail author
  • Kristin A. Linn
  • Fariha Khalid
  • Simon N. Vandekar
  • Shahamat Tauhid
  • Theodore D. Satterthwaite
  • John Muschelli
  • Rohit Bakshi
  • Russell T. Shinohara
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11383)


Magnetic resonance imaging (MRI) is crucial for in vivo detection and characterization of white matter lesions (WML) in multiple sclerosis (MS). The most widely established MRI outcome measure is the volume of hyperintense lesions on T2-weighted images (T2L). Unfortunately, T2L are non-specific for the level of tissue destruction and show a weak relationship to clinical status. Interest in lesions appearing hypointense on T1-weighted images (T1L) (“black holes”), which provide more specificity for axonal loss and a closer link to neurologic disability, has thus grown. The technical difficulty of T1L segmentation has led investigators to rely on time-consuming manual assessments prone to inter- and intra-rater variability. We implement MIMoSA, a current T2L automatic segmentation approach, to delineate T1L. Using cross-validation, MIMoSA proved robust for segmenting both T2L and T1L. For T2L, a Sørensen-Dice coefficient (DSC) of 0.6 and partial AUC (pAUC) up to 1% false positive rate of 0.69 were achieved. For T1L, 0.48 DSC and 0.63 pAUC were achieved. The correlation between EDSS and manual versus automatic volumes were similar for T1L (0.32 manual vs. 0.34 MIMoSA) and T2L (0.34 vs. 0.34).


Logistic regression Inter-modal coupling Multiple sclerosis 



This project was supported in part by a pilot grant from the Center for Biomedical Computing and Analytics at the University of Pennsylvania as well as R01NS085211, R21NS093349, R01NS060910, and R01MH112847 from the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.


  1. 1.
    Berer, K., Krishnamoorthy, G.: Microbial view of central nervous system autoimmunity. FEBS Lett. 588, 4207–4213 (2014)CrossRefGoogle Scholar
  2. 2.
    Feigin, V.L., et al.: Global, regional, and national burden of neurological disorders during 1990–2015: a systematic analysis for the global burden of disease study 2015. Lancet Neurol. 16, 877–897 (2017)CrossRefGoogle Scholar
  3. 3.
    Rovira, À., León, A.: MR in the diagnosis and monitoring of multiple sclerosis: an overview. Eur. J. Radiol. 67, 409–414 (2008)CrossRefGoogle Scholar
  4. 4.
    Katdare, A., Ursekar, M.: Systematic imaging review: multiple sclerosis. Ann. Indian Acad. Neurol. 18, S24–S29 (2015)CrossRefGoogle Scholar
  5. 5.
    Bakshi, R., Minagar, A., Jaisani, Z., Wolinsky, J.S.: Imaging of multiple sclerosis: role in neurotherapeutics. NeuroRx 2, 277–303 (2005)CrossRefGoogle Scholar
  6. 6.
    García-Lorenzo, D., Francis, S., Narayanan, S., Arnold, D.L., Collins, D.L.: Review of automatic segmentation methods of multiple sclerosis white matter lesions on conventional magnetic resonance imaging. Med. Image Anal. 17, 1–18 (2013)CrossRefGoogle Scholar
  7. 7.
    Meier, D.S., et al.: Dual-sensitivity multiple sclerosis lesion and CSF segmentation for multichannel 3T brain MRI. J. Neuroimaging Off. J. Am. Soc. Neuroimaging 28, 36–47 (2017)CrossRefGoogle Scholar
  8. 8.
    Sweeney, E.M., et al.: OASIS is automated statistical inference for segmentation, with applications to multiple sclerosis lesion segmentation in MRI. NeuroImage Clin. 2, 402–413 (2013)CrossRefGoogle Scholar
  9. 9.
    Shiee, N., Bazin, P.-L., Ozturk, A., Reich, D.S., Calabresi, P.A., Pham, D.L.: A topology-preserving approach to the segmentation of brain images with multiple sclerosis lesions. NeuroImage 49, 1524–1535 (2010)CrossRefGoogle Scholar
  10. 10.
    Valcarcel, A.M., et al.: MIMoSA: an automated method for intermodal segmentation analysis of multiple sclerosis brain lesions. J. Neuroimaging 28, 389–398 (2018)CrossRefGoogle Scholar
  11. 11.
    Ceccarelli, A., et al.: The impact of lesion in-painting and registration methods on voxel-based morphometry in detecting regional cerebral gray matter atrophy in multiple sclerosis. AJNR Am. J. Neuroradiol. 33, 1579–1585 (2012). PubMed – NCBI. Scholar
  12. 12.
    Filippi, M., Rovaris, M., Campi, A., Pereira, C., Comi, G.: Semi-automated thresholding technique for measuring lesion volumes in multiple sclerosis: effects of the change of the threshold on the computed lesion loads. Acta Neurol. Scand. 93, 30–34 (1996)CrossRefGoogle Scholar
  13. 13.
    Molyneux, P.D., et al.: The precision of T1 hypointense lesion volume quantification in multiple sclerosis treatment trials: a multicenter study. Mult. Scler. J. 6, 237–240 (2000)CrossRefGoogle Scholar
  14. 14.
    Datta, S., Sajja, B.R., He, R., Wolinsky, J.S., Gupta, R.K., Narayana, P.A.: Segmentation and quantification of black holes in multiple sclerosis. NeuroImage 29, 467–474 (2006)CrossRefGoogle Scholar
  15. 15.
    Wu, Y., et al.: Automated segmentation of multiple sclerosis lesion subtypes with multichannel MRI. NeuroImage 32, 1205–1215 (2006)CrossRefGoogle Scholar
  16. 16.
    Spies, L., et al.: Fully automatic detection of deep white matter T1 hypointense lesions in multiple sclerosis. Phys. Med. Biol. 58, 8323–8337 (2013)CrossRefGoogle Scholar
  17. 17.
    Muschelli, J.: extrantsr: extra functions to build on the ANTsR package (2014)Google Scholar
  18. 18.
    Shinohara, R.T., Muschelli, J.: WhiteStripe: white matter normalization for magnetic resonance images using WhiteStripe (2017)Google Scholar
  19. 19.
    Doshi, J., Erus, G., Ou, Y., Gaonkar, B., Davatzikos, C.: Multi-Atlas Skull-Stripping. Acad. Radiol. 20, 1566–1576 (2013)CrossRefGoogle Scholar
  20. 20.
    NITRC: CBICA: Multi Atlas Skull Stripping (MASS): Tool/Resource Info.
  21. 21.
    Tustison, N.J., et al.: N4ITK: improved N3 bias correction. IEEE Trans. Med. Imaging 29, 1310–1320 (2010)CrossRefGoogle Scholar
  22. 22.
    Vandekar, S.N.: Subject-level measurement of local cortical coupling. NeuroImage 133, 88–97 (2016)CrossRefGoogle Scholar
  23. 23.
    Hand, D.J., et al.: Classifier technology and the illusion of progress. Stat. Sci. 21, 1–14 (2006)MathSciNetCrossRefGoogle Scholar
  24. 24.
    Sing, T., Sander, O., Beerenwinkel, N., Lengauer, T.: ROCR: visualizing classifier performance in R. Bioinformatics 21, 3940–3941 (2005)CrossRefGoogle Scholar
  25. 25.
    Walter, S.D.: The partial area under the summary ROC curve. Stat. Med. 24, 2025–2040 (2005)MathSciNetCrossRefGoogle Scholar
  26. 26.
    Zivadinov, R., Bakshi, R.: Role of MRI in multiple sclerosis I: inflammation and lesions. Front. Biosci. J. Virtual Libr. 9, 665–683 (2004)CrossRefGoogle Scholar
  27. 27.
  28. 28.

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Alessandra M. Valcarcel
    • 1
    Email author
  • Kristin A. Linn
    • 1
  • Fariha Khalid
    • 2
  • Simon N. Vandekar
    • 1
  • Shahamat Tauhid
    • 2
  • Theodore D. Satterthwaite
    • 3
  • John Muschelli
    • 4
  • Rohit Bakshi
    • 2
  • Russell T. Shinohara
    • 1
  1. 1.Department of Biostatistics, Epidemiology, and InformaticsUniversity of PennsylvaniaPhiladelphiaUSA
  2. 2.Department of Neurology and RadiologyBrigham and Women’s HospitalBostonUSA
  3. 3.Department of PsychiatryUniversity of PennsylvaniaPhiladelphiaUSA
  4. 4.Department of BiostatisticsJohns Hopkins UniversityBaltimoreUSA

Personalised recommendations