Journal of Soft Computing and Information Technology

Journal of Soft Computing and Information Technology

Population-Specific Lesion Patterns in Multiple Sclerosis: Automated Deep Learning Analysis of a Large Iranian Dataset

Document Type : Persian Original Article

Authors
Mahdi Bashiri Bawil 1
Mousa Shamsi 1
Abolhassan Shakeri Bavil 2
1 Department of Biomedical Engineering, Tabriz University of Technology, Tabriz, Iran.
2 Department of Radiology, Tabriz University of Medical Sciences, Tabriz, Iran.
10.22034/jscit.2026.555620.2186
Abstract
Multiple sclerosis impacts approximately 2.8 million people globally, yet Middle Eastern populations remain significantly underrepresented in neuroimaging studies despite higher regional disease prevalence, which constrains understanding of population-specific pathological features and hinders development of tailored treatment strategies. Comprehensive comparative investigations between MS patients and healthy individuals are crucial for identifying disease-specific brain alterations and defining normative biomarker ranges. This retrospective observational investigation examined 1,381 participants (1,000 healthy controls, 381 MS cases) from Northwest Iran, employing deep learning architectures for automated FLAIR sequence analysis. Neuroanatomical classification distinguished periventricular, paraventricular, and juxtacortical white matter lesions, with statistical evaluations incorporating age-gender stratification and comprehensive correlation analyses of ventricular and lesion metrics. The attention U-Net architecture demonstrated clinically viable performance (Dice coefficient: 0.858; Intersection over Union: 0.756). MS participants exhibited 5.5-fold elevated lesion loads across age groups relative to controls, with normalized proportions increasing from 0.13% to 0.71% (p<0.001; effect size r=0.82). Periventricular lesions constituted the majority of total burden (58.02±28.35%), with notable gender-specific anatomical distribution differences. Age-related associations were more pronounced in MS cases (ventricular proportion: r=0.319; lesion load: r=0.230) versus healthy individuals. This investigation establishes validated population-specific reference values for MS neuroimaging biomarkers in Middle Eastern cohorts, demonstrating distinctive lesion accumulation characteristics with marked periventricular predominance. The automated segmentation methodology and statistical approach address significant representational gaps in international MS research.
Keywords
Multiple Sclerosis
Neuroimaging
FLAIR Segmentation
Deep Learning
Attention U-Net
  [1]     G. Khan and M. J. Hashim, “Epidemiology of Multiple Sclerosis: Global, Regional, National and Sub-National-Level Estimates and Future Projections,” Journal of Epidemiology and Global Health, vol. 15, no. 1, Feb. 2025, doi: https://doi.org/10.1007/s44197-025-00353-6.
  [2]     T. J. Simkins, G. J. Duncan, and D. Bourdette, “Chronic Demyelination and Axonal Degeneration in Multiple Sclerosis: Pathogenesis and Therapeutic Implications,” Current Neurology and Neuroscience Reports, vol. 21, no. 6, Apr. 2021, doi: https://doi.org/10.1007/s11910-021-01110-5.
  [3]     C. E. Sabel, J. F. Pearson, D. F. Mason, E. Willoughby, D. A. Abernethy, and B. V. Taylor, “The latitude gradient for multiple sclerosis prevalence is established in the early life course,” Brain, vol. 144, no. 7, pp. 2038–2046, Mar. 2021, doi: https://doi.org/10.1093/brain/awab104.
  [4]     G. Criste, B. Trapp, and R. Dutta, “Axonal loss in multiple sclerosis,” Handbook of Clinical Neurology, vol. 112, pp. 101–113, 2014, doi: https://doi.org/10.1016/b978-0-444-52001-2.00005-4.
  [5]     E. Portaccio et al., “Multiple sclerosis: emerging epidemiological trends and redefining the clinical course,” The Lancet Regional Health - Europe, vol. 44, p. 100977, Sep. 2024, doi: https://doi.org/10.1016/j.lanepe.2024.100977.
  [6]     T. L. Vollmer, K. V. Nair, I. M. Williams, and E. Alvarez, “Multiple Sclerosis Phenotypes as a Continuum: The Role of Neurologic Reserve,” Neurology: Clinical Practice, vol. 11, no. 4, pp. 342–351, Aug. 2021, doi: https://doi.org/10.1212/CPJ.0000000000001045.
  [7]     F. D. Lublin et al., “Defining the clinical course of multiple s clerosis: The 2013 revisions,” Neurology, vol. 83, no. 3, pp. 278–286, May 2014, doi: https://doi.org/10.1212/wnl.0000000000000560.
  [8]     A. J. Thompson et al., “Diagnosis of Multiple sclerosis: 2017 Revisions of the McDonald Criteria,” The Lancet Neurology, vol. 17, no. 2, pp. 162–173, Feb. 2018, doi: https://doi.org/10.1016/s1474-4422(17)30470-2.
  [9]     M. A. Rocca et al., “Current and future role of MRI in the diagnosis and prognosis of multiple sclerosis,” The Lancet Regional Health - Europe, vol. 44, no. 100978, pp. 100978–100978, Aug. 2024, doi: https://doi.org/10.1016/j.lanepe.2024.100978.
[10]     F. Dehghani, H. Arabi, and A. Karimian, “Automated brain tumor segmentation on multi-mr sequences to determine the most efficient sequence using a deep learning method,” Computational Intelligence in Electrical Engineering, vol. 14, Art. no. 1, 2023, doi: https://doi.org/10.22108/isee.2021.126101.1427.
[11]     M. Filippi et al., “MRI criteria for the diagnosis of multiple sclerosis: MAGNIMS consensus guidelines,” The Lancet. Neurology, vol. 15, no. 3, pp. 292–303, 2016, doi: https://doi.org/10.1016/S1474-4422(15)00393-2.
[12]     A. Jankowska, Kamil Chwojnicki, and Edyta Szurowska, “The diagnosis of multiple sclerosis: what has changed in diagnostic criteria?,” Polish Journal of Radiology, vol. 88, no. 1, pp. 574–581, Jan. 2023, doi: https://doi.org/10.5114/pjr.2023.133677.
[13]     P. Schwenkenbecher et al., “Impact of the McDonald Criteria 2017 on Early Diagnosis of Relapsing-Remitting Multiple Sclerosis,” Frontiers in Neurology, vol. 10, Mar. 2019, doi: https://doi.org/10.3389/fneur.2019.00188.
[14]     C. H. Polman et al., “Diagnostic criteria for multiple sclerosis: 2010 Revisions to the McDonald criteria,” Annals of Neurology, vol. 69, no. 2, pp. 292–302, Feb. 2011, doi: https://doi.org/10.1002/ana.22366.
[15]     O. I. Alomair, “Conventional and Advanced Magnetic Resonance Imaging Biomarkers of Multiple Sclerosis in the Brain,” Cureus, vol. 17, no. 3, Mar. 2025, doi: https://doi.org/10.7759/cureus.79914.
[16]     M. Filippi et al., “Assessment of lesions on magnetic resonance imaging in multiple sclerosis: practical guidelines,” Brain, vol. 142, no. 7, pp. 1858–1875, Jun. 2019, doi: https://doi.org/10.1093/brain/awz144.
[17]     M. A. Sahraian and A. Eshaghi, “Role of MRI in diagnosis and treatment of multiple sclerosis,” Clinical Neurology and Neurosurgery, vol. 112, no. 7, pp. 609–615, Sep. 2010, doi: https://doi.org/10.1016/j.clineuro.2010.03.022.
[18]     A. Fahmi Jafargholkhanloo, M. Shamsi, and M. Bashiri Bawil, “A novel fuzzy-based clustering algorithm for segmentation of brain tissues with bias correction based on MRI images,” Journal of Machine Vision and Image Processing, vol. 12, Art. no. 2, 2025, Available: https://jmvip.sinaweb.net/article_229851.html.
[19]     M. Filippi et al., “Present and future of the diagnostic work-up of multiple sclerosis: the imaging perspective,” Journal of Neurology, vol. 270, no. 3, Nov. 2022, doi: https://doi.org/10.1007/s00415-022-11488-y.
[20]     Fahmi Jafargholkhanloo, M. Shamsi, and M. Bashiri Bawil, “Segmentation of brain tissues in MRI images using improved gustafson-kessel clustering algorithm,” Journal of Soft Computing and Information Technology, vol. 13, Art. no. 4, 2025, Available: https://jscit.nit.ac.ir/article_217009.html.
[21]     O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional networks for biomedical image segmentation,” arXiv (Cornell University), May 2015, doi: https://doi.org/10.48550/arxiv.1505.04597.
[22]     O. Oktay et al., “Attention u-net: Learning where to look for the pancreas,” 2018. https://arxiv.org/abs/1804.03999.
[23]     T. Brosch, L. Y. W. Tang, Y. Yoo, D. K. B. Li, A. Traboulsee, and R. Tam, “Deep 3D Convolutional Encoder Networks With Shortcuts for Multiscale Feature Integration Applied to Multiple Sclerosis Lesion Segmentation,” IEEE Transactions on Medical Imaging, vol. 35, no. 5, pp. 1229–1239, May 2016, doi: https://doi.org/10.1109/tmi.2016.2528821.
[24]     Z. Chen, X. Wang, J. Huang, J. Lu, and J. Zheng, “Deep Attention and Graphical Neural Network for Multiple Sclerosis Lesion Segmentation From MR Imaging Sequences,” IEEE Journal of Biomedical and Health Informatics, vol. 26, no. 3, pp. 1196–1207, Mar. 2022, doi: https://doi.org/10.1109/jbhi.2021.3109119.
[25]     M. Sadeghibakhi, H. Pourreza, and H. Mahyar, “Multiple Sclerosis Lesions Segmentation Using Attention-Based CNNs in FLAIR Images,” IEEE Journal of Translational Engineering in Health and Medicine, vol. 10, pp. 1–11, 2022, doi: https://doi.org/10.1109/jtehm.2022.3172025.
[26]     M. Bashiri Bawil, M. Shamsi, A. Shakeri Bavil, and S. Danishvar, “Specialized gray matter segmentation via a generative adversarial network: application on brain white matter hyperintensities classification,” Frontiers in Neuroscience, vol. Volume 18 - 2024, 2024, doi: https://doi.org/10.3389/fnins.2024.1416174.
[27]     S. Umirzakova, Muksimova Shakhnoza, Mardieva Sevara, and Taeg Keun Whangbo, “Deep learning for multiple sclerosis lesion classification and stratification using MRI,” Computers in Biology and Medicine, vol. 192, pp. 110078–110078, Apr. 2025, doi: https://doi.org/10.1016/j.compbiomed.2025.110078.
[28]     F. Isensee, P. F. Jaeger, S. A. A. Kohl, J. Petersen, and K. H. Maier-Hein, “nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation,” Nature Methods, vol. 18, no. 2, pp. 203–211, Dec. 2020, doi: https://doi.org/10.1038/s41592-020-01008-z.
[29]     P. De et al., “Consensus of algorithms for lesion segmentation in brain MRI studies of multiple sclerosis,” Scientific Reports, vol. 14, no. 1, Sep. 2024, doi: https://doi.org/10.1038/s41598-024-72649-9.
[30]     O. Mirmosayyeb, V. Shaygannejad, S. Bagherieh, A. M. Hosseinabadi, and M. Ghajarzadeh, “Prevalence of multiple sclerosis (MS) in Iran: a systematic review and meta-analysis,” Neurological Sciences, vol. 43, no. 1, pp. 233–241, Nov. 2021, doi: https://doi.org/10.1007/s10072-021-05750-w.
[31]     C. Walton et al., “Rising Prevalence of Multiple Sclerosis worldwide: Insights from the Atlas of MS, Third Edition,” Multiple Sclerosis Journal, vol. 26, no. 14, pp. 1816–1821, Nov. 2020, doi: https://doi.org/10.1177/1352458520970841.
[32]     M. Saadatnia, Masoud Etemadifar, and Amir-Hadi Maghzi, “Multiple Sclerosis in Isfahan, Iran,” Elsevier eBooks, pp. 357–375, Jan. 2007, doi: https://doi.org/10.1016/s0074-7742(07)79016-5.
[33]     S. Eskandarieh, P. Heydarpour, S.-R. Elhami, and M. A. Sahraian, “Prevalence and Incidence of Multiple Sclerosis in Tehran, Iran,” Iranian journal of public health, vol. 46, no. 5, pp. 699–704, May 2017, Available: https://pubmed.ncbi.nlm.nih.gov/28560202/
[34]     M. Azami, M. H. YektaKooshali, M. Shohani, A. Khorshidi, and L. Mahmudi, “Epidemiology of multiple sclerosis in Iran: A systematic review and meta-analysis,” PLOS ONE, vol. 14, no. 4, p. e0214738, Apr. 2019, doi: https://doi.org/10.1371/journal.pone.0214738.
[35]     N. Fattahi et al., “Burden of multiple sclerosis in Iran from 1990 to 2017,” BMC Neurology, vol. 21, no. 1, Oct. 2021, doi: https://doi.org/10.1186/s12883-021-02431-1.
[36]     F. Spagnolo et al., “How far MS lesion detection and segmentation are integrated into the clinical workflow? A systematic review,” NeuroImage: Clinical, vol. 39, pp. 103491–103491, Jan. 2023, doi: https://doi.org/10.1016/j.nicl.2023.103491.
[37]     A. J. Solomon, R. Pettigrew, R. T. Naismith, S. Chahin, S. Krieger, and B. Weinshenker, “Challenges in multiple sclerosis diagnosis: Misunderstanding and misapplication of the McDonald criteria,” Multiple Sclerosis Journal, p. 135245852091049, Mar. 2020, doi: https://doi.org/10.1177/1352458520910496.
[38]     O. Ronneberger et al., "U-Net: Convolutional Networks for Biomedical Image Segmentation," in Medical Image Computing and Computer-Assisted Intervention, pp. 234-241, 2015.doi: https://doi.org/10.48550/arXiv.1505.04597.
[39]     L.-C. Chen et al., "Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation," in European Conference on Computer Vision, pp. 833-851, 2018. doi: https://doi.org/10.48550/arXiv.1802.02611.
[40]     J. Chen et al., "TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation," arXiv preprint arXiv:2102.04306, 2021. doi: https://doi.org/10.48550/arXiv.2102.04306.
[41]     G. Arrambide et al., “Lesion topographies in multiple sclerosis diagnosis,” Neurology, vol. 89, no. 23, pp. 2351–2356, Nov. 2017, doi: https://doi.org/10.1212/wnl.0000000000004715.
[42]     A. Traboulsee et al., “Revised Recommendations of the Consortium of MS Centers Task Force for a Standardized MRI Protocol and Clinical Guidelines for the Diagnosis and Follow-Up of Multiple Sclerosis,” American Journal of Neuroradiology, vol. 37, no. 3, pp. 394–401, Nov. 2015, doi: https://doi.org/10.3174/ajnr.a4539.
[43]     K. W. Kim, J. R. MacFall, and M. E. Payne, “Classification of white matter lesions on magnetic resonance imaging in elderly persons,” Biological Psychiatry, vol. 64, no. 4, pp. 273–280, Aug. 2008, doi: https://doi.org/10.1016/j.biopsych.2008.03.024.
[44]     C. DeCarli, E. Fletcher, V. Ramey, D. Harvey, and W. J. Jagust, “Anatomical Mapping of White Matter Hyperintensities (WMH),” Stroke, vol. 36, no. 1, pp. 50–55, Jan. 2005, doi: https://doi.org/10.1161/01.str.0000150668.58689.f2.
[45]     W. I. McDonald et al., “Recommended diagnostic criteria for multiple sclerosis: Guidelines from the international panel on the diagnosis of multiple sclerosis,” Annals of Neurology, vol. 50, no. 1, pp. 121–127, 2001, doi: https://doi.org/10.1002/ana.1032.
[46]     Y. Shan et al., “Risk Factors and Clinical Manifestations of Juxtacortical Small Lesions: A Neuroimaging Study,” Frontiers in Neurology, vol. 8, Sep. 2017, doi: https://doi.org/10.3389/fneur.2017.00497.
Journal of Soft Computing and Information Technology
Volume 15, Issue 1
Spring - April
Spring 2026
Pages 58-75