Skip to main content
SpringerLink
Log in

Joint radiomics and spatial distribution model for MRI-based discrimination of multiple sclerosis, neuromyelitis optica spectrum disorder, and myelin-oligodendrocyte-glycoprotein-IgG-associated disorder

  • Imaging Informatics and Artificial Intelligence
  • Published:
European Radiology Aims and scope Submit manuscript

Abstract

Objective

To develop a discrimination pipeline concerning both radiomics and spatial distribution features of brain lesions for discrimination of multiple sclerosis (MS), aquaporin-4-IgG-seropositive neuromyelitis optica spectrum disorder (NMOSD), and myelin-oligodendrocyte-glycoprotein-IgG-associated disorder (MOGAD).

Methods

Hyperintensity T2 lesions were delineated in 212 brain MRI scans of MS (n = 63), NMOSD (n = 87), and MOGAD (n = 45) patients. To avoid the effect of fixed training/test dataset sampling when developing machine learning models, patients were allocated into 4 sub-groups for cross-validation. For each scan, 351 radiomics and 27 spatial distribution features were extracted. Three models, i.e., multi-lesion radiomics, spatial distribution, and joint models, were constructed using random forest and logistic regression algorithms for differentiating: MS from the others (MS models) and MOGAD from NMOSD (MOG-NMO models), respectively. Then, the joint models were combined with demographic characteristics (i.e., age and sex) to create MS and MOG-NMO discriminators, respectively, based on which a three-disease discrimination pipeline was generated and compared with radiologists.

Results

For classification of both MS-others and MOG-NMO, the joint models performed better than radiomics or spatial distribution model solely. The MS discriminator achieved AUC = 0.909 ± 0.027 and bias-corrected C-index = 0.909 ± 0.027, and the MOG-NMO discriminator achieved AUC = 0.880 ± 0.064 and bias-corrected C-index = 0.883 ± 0.068. The three-disease discrimination pipeline differentiated MS, NMOSD, and MOGAD patients with 75.0% accuracy, prominently outperforming the three radiologists (47.6%, 56.6%, and 66.0%).

Conclusions

The proposed pipeline integrating multi-lesion radiomics and spatial distribution features could effectively differentiate MS, NMOSD, and MOGAD.

Clinical relevance statement

The discrimination pipeline merging both radiomics and spatial distribution features of brain lesions may facilitate the differential diagnoses of multiple sclerosis, neuromyelitis optica spectrum disorder, and myelin-oligodendrocyte-glycoprotein-IgG-associated disorder.

Key Points

• Our study introduces an approach by combining radiomics and spatial distribution models.

• The joint model exhibited superior performance in distinguishing multiple sclerosis from aquaporin-4-IgG-seropositive neuromyelitis optica spectrum disorder and myelin-oligodendrocyte-glycoprotein-IgG-associated disorder as well as discriminating the latter two diseases.

• The three-disease discrimination pipeline showcased remarkable accuracy, surpassing the performance of experienced radiologists, highlighting its potential as a valuable diagnostic tool.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Abbreviations

AQP4-IgG:

Aquaporin-4-immunoglobulin-G

IBSI:

Imaging biomarker standardization initiative

MOGAD:

Myelin-oligodendrocyte-glycoprotein-IgG-associated disorder

MS:

Multiple sclerosis

NMOSD:

Neuromyelitis optica spectrum disorder

RIL:

The Radiomics Index for Lesion

RIS:

The Radiomics Index for Scan

SDI:

The Spatial Distribution Index

References

  1. Chawla S, Kister I, Wuerfel J et al (2016) Iron and non-iron-related characteristics of multiple sclerosis and neuromyelitis optica lesions at 7T MRI. AJNR Am J Neuroradiol 37:1223–1230

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Sinnecker T, Schumacher S, Mueller K et al (2016) MRI phase changes in multiple sclerosis vs neuromyelitis optica lesions at 7T. Neurol Neuroimmunol Neuroinflamm 3:e259

    Article  PubMed  PubMed Central  Google Scholar 

  3. Wegner C (2013) Recent insights into the pathology of multiple sclerosis and neuromyelitis optica. Clin Neurol Neurosurg 115:S38–S41

    Article  PubMed  Google Scholar 

  4. Rotstein D, Montalban X (2019) Reaching an evidence-based prognosis for personalized treatment of multiple sclerosis. Nat Rev Neurol 15:287–300

    Article  PubMed  Google Scholar 

  5. Palace J, Leite MI, Nairne A, Vincent A (2010) Interferon beta treatment in neuromyelitis optica. Arch Neurol 67:1016–1017

    Article  PubMed  Google Scholar 

  6. Sechi E, Cacciaguerra L, Chen JJ et al (2022) Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD): a review of clinical and MRI features, diagnosis, and management. Front Neurol 13:885218

    Article  PubMed  PubMed Central  Google Scholar 

  7. Reindl M, Waters P (2019) Myelin oligodendrocyte glycoprotein antibodies in neurological disease. Nat Rev Neurol 15:89–102

    Article  CAS  PubMed  Google Scholar 

  8. Jarius S, Paul F, Aktas O et al (2018) MOG encephalomyelitis: international recommendations on diagnosis and antibody testing. J Neuroinflammation 15:134

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Matthews L, Marasco R, Jenkinson M et al (2013) Distinction of seropositive NMO spectrum disorder and MS brain lesion distribution. Neurology 80:1330–1337

    Article  PubMed  PubMed Central  Google Scholar 

  10. Juryńczyk M, Tackley G, Kong Y et al (2017) Brain lesion distribution criteria distinguish MS from AQP4-antibody NMOSD and MOG-antibody disease. J Neurol Neurosurg Psychiatry 88:132–136

    Article  PubMed  Google Scholar 

  11. Hyun J-W, Huh S-Y, Shin H-J et al (2019) Evaluation of brain lesion distribution criteria at disease onset in differentiating MS from NMOSD and MOG-IgG-associated encephalomyelitis. Mult Scler 25:585–590

    Article  PubMed  Google Scholar 

  12. Juryńczyk M, Geraldes R, Probert F et al (2017) Distinct brain imaging characteristics of autoantibody-mediated CNS conditions and multiple sclerosis. Brain 140:617–627

    Article  PubMed  Google Scholar 

  13. Cacciaguerra L, Meani A, Mesaros S et al (2019) Brain and cord imaging features in neuromyelitis optica spectrum disorders. Ann Neurol 85:371–384

    Article  PubMed  Google Scholar 

  14. Cai M-T, Zheng Y, Shen C-H et al (2021) Evaluation of brain and spinal cord lesion distribution criteria at disease onset in distinguishing NMOSD from MS and MOG antibody-associated disorder. Mult Scler 27:871–882

    Article  CAS  PubMed  Google Scholar 

  15. Patel J, Pires A, Derman A et al (2022) Development and validation of a simple and practical method for differentiating MS from other neuroinflammatory disorders based on lesion distribution on brain MRI. J Clin Neurosci 101:32–36

    Article  CAS  PubMed  Google Scholar 

  16. Cortese R, Prados Carrasco F, Tur C et al (2023) Differentiating multiple sclerosis from AQP4-neuromyelitis optica spectrum disorder and MOG-antibody disease with imaging. Neurology 100:e308–e323

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Filippi M, Preziosa P, Banwell BL et al (2019) Assessment of lesions on magnetic resonance imaging in multiple sclerosis: practical guidelines. Brain 142:1858–1875

    Article  PubMed  PubMed Central  Google Scholar 

  18. Lambin P, Leijenaar RTH, Deist TM et al (2017) Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol 14:749–762

    Article  PubMed  Google Scholar 

  19. Ma X, Zhang L, Huang D et al (2019) Quantitative radiomic biomarkers for discrimination between neuromyelitis optica spectrum disorder and multiple sclerosis. J Magn Reson Imaging 49:1113–1121

    Article  PubMed  Google Scholar 

  20. Luo X, Piao S, Li H et al (2022) Multi-lesion radiomics model for discrimination of relapsing-remitting multiple sclerosis and neuropsychiatric systemic lupus erythematosus. Eur Radiol 32:5700–5710

    Article  CAS  PubMed  Google Scholar 

  21. Yang L, Li H, Xia W et al (2020) Quantitative brain lesion distribution may distinguish MOG-ab and AQP4-ab neuromyelitis optica spectrum disorders. Eur Radiol 30:1470–1479

    Article  PubMed  Google Scholar 

  22. Thompson AJ, Banwell BL, Barkhof F et al (2018) Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol 17:162–173

    Article  PubMed  Google Scholar 

  23. Wingerchuk DM, Banwell B, Bennett JL et al (2015) International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology 85:177–189

    Article  PubMed  PubMed Central  Google Scholar 

  24. Ashrafinia S (2019) Quantitative nuclear medicine imaging using advanced image reconstruction and radiomics. Johns Hopkins University, Baltimore, US. Available via https://jscholarship.library.jhu.edu/handle/1774.2/61551. Accessed 20 Feb 2022

  25. Zwanenburg A, Vallieres M, Abdalah MA et al (2020) The Image Biomarker Standardization Initiative: standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology 295:328–338

    Article  PubMed  Google Scholar 

  26. Zhang Y, Moore GW, Laule C et al (2013) Pathological correlates of magnetic resonance imaging texture heterogeneity in multiple sclerosis. Ann Neurol 74:91–99

    Article  CAS  PubMed  Google Scholar 

  27. Sinnecker T, Dörr J, Pfueller CF et al (2012) Distinct lesion morphology at 7-T MRI differentiates neuromyelitis optica from multiple sclerosis. Neurology 79:708–714

    Article  PubMed  Google Scholar 

  28. Sinnecker T, Clarke MA, Meier D et al (2019) Evaluation of the central vein sign as a diagnostic imaging biomarker in multiple sclerosis. JAMA Neurol 76:1446–1456

    Article  PubMed  Google Scholar 

  29. Jeong IH, Choi JY, Kim S-H et al (2016) Comparison of myelin water fraction values in periventricular white matter lesions between multiple sclerosis and neuromyelitis optica spectrum disorder. Mult Scler 22:1616–1620

    Article  PubMed  Google Scholar 

  30. Ito S, Mori M, Makino T, Hayakawa S, Kuwabara S (2009) “Cloud-like enhancement” is a magnetic resonance imaging abnormality specific to neuromyelitis optica. Ann Neurol 66:425–428

    Article  PubMed  Google Scholar 

  31. Kaneko K, Sato DK, Nakashima I et al (2018) CSF cytokine profile in MOG-IgG+ neurological disease is similar to AQP4-IgG+ NMOSD but distinct from MS: a cross-sectional study and potential therapeutic implications. J Neurol Neurosurg Psychiatry 89:927–936

    Article  PubMed  Google Scholar 

  32. Chen X, Zeng C, Luo T et al (2012) Iron deposition of the deep grey matter in patients with multiple sclerosis and neuromyelitis optica: a control quantitative study by 3D-enhanced susceptibility-weighted angiography (ESWAN). Eur J Radiol 81:e633–e639

    Article  PubMed  Google Scholar 

  33. Cooray GK, Sundgren M, Brismar T (2020) Mechanism of visual network dysfunction in relapsing-remitting multiple sclerosis and its relation to cognition. Clin Neurophysiol 131:361–367

    Article  PubMed  Google Scholar 

  34. Höftberger R, Guo Y, Flanagan EP et al (2020) The pathology of central nervous system inflammatory demyelinating disease accompanying myelin oligodendrocyte glycoprotein autoantibody. Acta Neuropathol 139:875–892

    Article  PubMed  PubMed Central  Google Scholar 

  35. Amiry-Moghaddam M, Ottersen OP (2003) The molecular basis of water transport in the brain. Nat Rev Neurosci 4:991–1001

    Article  CAS  PubMed  Google Scholar 

  36. Hinson SR, Pittock SJ, Lucchinetti CF et al (2007) Pathogenic potential of IgG binding to water channel extracellular domain in neuromyelitis optica. Neurology 69:2221–2231

    Article  CAS  PubMed  Google Scholar 

  37. Takai Y, Misu T, Suzuki H et al (2021) Staging of astrocytopathy and complement activation in neuromyelitis optica spectrum disorders. Brain 144:2401–2415

    Article  PubMed  Google Scholar 

  38. Brunner C, Lassmann H, Waehneldt TV, Matthieu J-M, Linington C (1989) Differential ultrastructural localization of myelin basic protein, myelin/oligodendroglial glycoprotein, and 2′,3′-cyclic nucleotide 3′-phosphodiesterase in the CNS of adult rats. J Neurochem 52:296–304

    Article  CAS  PubMed  Google Scholar 

  39. Fujimori J, Takai Y, Nakashima I et al (2017) Bilateral frontal cortex encephalitis and paraparesis in a patient with anti-MOG antibodies. J Neurol Neurosurg Psychiatry 88:534–536

    Article  PubMed  Google Scholar 

Download references

Funding

This work has received funding from the National Natural Science Foundation of China (82102132, 81771296, 82171341, 82102016, 82202122, 8237071280), the Science and Technology Commission of Shanghai Municipality (20S31904300, 22TS1400900, 23S31904100, 22ZR1409500), and the Greater Bay Area Institute of Precision Medicine (Guangzhou) (KCH2310094).

Author information

Author notes

  1. Xiao Luo and Haiqing Li contributed equally to this work.

  2. Yuxin Li and Liqin Yang contributed equally to this work as the corresponding authors.

Authors and Affiliations

  1. Academy for Engineering and Technology, Fudan University, Shanghai, China

    Xiao Luo, Wei Xia & Daoying Geng

  2. Department of Radiology, Huashan Hospital, Fudan University, 12 Wulumuqi Middle Road, Shanghai, 200040, China

    Haiqing Li, Na Wang, Yifang Bao, Daoying Geng, Yuxin Li & Liqin Yang

  3. Department of Neurology, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai, China

    Chao Quan, Jingzi ZhangBao & Hongmei Tan

  4. National Center for Neurological Disorders, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai, China

    Chao Quan, Jingzi ZhangBao & Hongmei Tan

  5. Institute of Functional and Molecular Medical Imaging, Fudan University, Shanghai, China

    Daoying Geng, Yuxin Li & Liqin Yang

Authors
  1. Xiao Luo
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Haiqing Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Wei Xia
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Chao Quan
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Jingzi ZhangBao
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Hongmei Tan
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Na Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Yifang Bao
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Daoying Geng
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Yuxin Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Liqin Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Yuxin Li or Liqin Yang.

Ethics declarations

Guarantor

The scientific guarantor of this publication is Daoying Geng.

Conflict of interest

The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.

Statistics and biometry

One of the authors has significant statistical expertise.

Informed consent

Written informed consent was obtained from all subjects (patients) in this study.

Ethical approval

Approval was obtained from Huashan Hospital Institutional Review Board.

Study subjects or cohorts overlap

Some study subjects or cohorts have been previously reported published in two studies: https://doi.org/10.1007/s00330-019-06506-z, and https://doi.org/10.1007/s00330-022-08653-2.

Methodology

• retrospective

• cross sectional study

• performed at one institution

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (PDF 1674 KB)

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Luo, X., Li, H., Xia, W. et al. Joint radiomics and spatial distribution model for MRI-based discrimination of multiple sclerosis, neuromyelitis optica spectrum disorder, and myelin-oligodendrocyte-glycoprotein-IgG-associated disorder. Eur Radiol (2023). https://doi.org/10.1007/s00330-023-10529-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00330-023-10529-y

Keywords

Search

Navigation