pith. machine review for the scientific record. sign in

arxiv: 2605.09666 · v1 · submitted 2026-05-10 · 💻 cs.CV · cs.AI

Recognition: 2 theorem links

· Lean Theorem

Rethinking Evaluation of Multiple Sclerosis (MS) Lesion Segmentation Models

Abdul Basit , Ashir Rashid , Muhammad Abdullah Hanif , Muhammad Shafique
Authors on Pith no claims yet

Pith reviewed 2026-05-12 04:17 UTC · model grok-4.3

classification 💻 cs.CV cs.AI
keywords multiple sclerosislesion segmentationMRIevaluation metricsdeep learningDice scoreproblem fingerprintingclinical deployment
0
0 comments X

The pith

Evaluating multiple sclerosis lesion segmentation models requires going beyond the Dice score to include lesion-wise performance and metrics for complex cases important to neurologists.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper claims that deep learning models for detecting and segmenting MS lesions in brain MRI are routinely judged only by overall overlap scores like Dice, which hides their behavior on individual lesions or on scans that matter most for spotting disease early and tracking its spread. It introduces problem fingerprinting as a way to map exactly what neurologists examine in these images and which additional measurements are needed to match those priorities. An analysis of current leading models on public datasets then shows where their performance drops in ways that could affect real hospital decisions. A sympathetic reader would care because MS has no cure, only ways to slow it, so tools that look good in papers but falter on hard cases may delay useful assistance to patients.

Core claim

The authors establish that standard Dice-only evaluation is insufficient for MS lesion segmentation models because it does not capture lesion-wise detection accuracy, performance on cases that confuse human annotators, or metrics tied to disease detection and progression monitoring. They respond by detailing problem fingerprinting to specify neurologist priorities in MRI scans and by applying a broader set of metrics to state-of-the-art models on two open datasets, revealing gaps in practical usability for hospital deployment.

What carries the argument

Problem fingerprinting, a structured breakdown of the specific scan features and lesion scenarios neurologists prioritize for MS detection and monitoring, paired with lesion-wise and clinical-context metrics that quantify model behavior on those priorities.

If this is right

  • Models must demonstrate reliable detection of individual lesions rather than only aggregate overlap to be considered ready for clinical use.
  • Evaluation protocols will need to test performance on ambiguous or low-contrast lesions that matter for early diagnosis.
  • Progression-monitoring tools will require separate checks for longitudinal consistency across patient scans.
  • Public benchmarks should report both Dice and the lesion-specific metrics to allow direct comparison of real-world readiness.
  • Hospital adoption decisions can shift toward models that pass the expanded tests even if their Dice scores are comparable.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same fingerprinting approach could be adapted to other lesion-based tasks such as tumor or stroke segmentation where per-lesion reliability drives treatment choices.
  • Training pipelines might incorporate the new metrics as auxiliary losses to steer models toward clinically relevant behavior from the start.
  • Regulatory bodies reviewing AI tools for MS could require the expanded metric set as part of safety evidence.
  • If widely adopted, the method would surface systematic weaknesses that current leaderboards obscure, guiding targeted data collection for hard cases.

Load-bearing premise

That adding problem fingerprinting and the extra metrics will produce evaluations that more accurately predict which models will succeed in actual hospital settings than Dice scores alone.

What would settle it

A head-to-head trial in which models chosen by the new fingerprinting metrics show measurably better agreement with neurologist decisions or patient outcome tracking than models chosen solely by high Dice scores.

Figures

Figures reproduced from arXiv: 2605.09666 by Abdul Basit, Ashir Rashid, Muhammad Abdullah Hanif, Muhammad Shafique.

Figure 1
Figure 1. Figure 1: Performance of the nnU-Net model, trained on MSSEG-1 dataset, on two validation samples (Subject 7 and Subject 11) from MSSEG-1. (a) and (e) [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: MS disease stages and the corresponding expectations for an automated lesion segmentation model. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Overview of our proposed evaluation pipeline. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: nnU-Net Performance Evaluation on MSSEG-1 (A) Lesion Performance Scatter: Correlates Dice scores with lesion size, where color indicates the Predicted Lesion Size to GT Lesion Size Ratio. (B) Volumetric Stratification (Dice): Distribution of Dice scores across size bins (Very Small to Large). (C) Volumetric Stratification (HD95): Distribution of HD95 across size bins. (D) Metric Correlation: Relationship b… view at source ↗
Figure 5
Figure 5. Figure 5: nnU-Net Performance Evaluation on MSLesSeg: The same structure as [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
read the original abstract

Multiple Sclerosis (MS) is a chronic autoimmune disease that can significantly reduce the quality of life of a patient. Existing treatment options can only help slow down the progression of the disease. Therefore, early detection and precise monitoring of disease progression are important. Deep learning offers state-of-the-art models for detecting and segmenting MS lesions in brain MRI scans. However, most of these models are evaluated using the Dice score, without accounting for lesion-wise detection and segmentation performance or other metrics that quantify model performance in cases that are complex or confusing for human annotators, or in cases that are essential for disease detection and progression monitoring. In this paper, we highlight the need to rethink the evaluation of MS lesion segmentation models. In this context, we first present problem fingerprinting in detail to highlight what neurologists look for in brain MRI scans for MS detection and progression monitoring, and which metrics are required to properly quantify model performance in these contexts. Additionally, we present an analysis of state-of-the-art models on two open-source datasets using these metrics to highlight their usability for real-world deployment in hospitals.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims that MS lesion segmentation models are primarily evaluated using the Dice score, which overlooks lesion-wise detection and segmentation performance as well as metrics relevant to cases that are complex for human annotators or critical for disease detection and progression monitoring. It introduces 'problem fingerprinting' to detail neurologist priorities in brain MRI scans and proposes additional metrics to better quantify model performance in these contexts. The paper then analyzes state-of-the-art models on two open-source datasets using these metrics to demonstrate differences in their usability for real-world hospital deployment.

Significance. If validated, the emphasis on clinical-context metrics and problem fingerprinting could improve model selection for MS lesion segmentation by better reflecting real-world needs in early detection and monitoring, addressing a known gap in medical image analysis evaluation. The work merits credit for applying the proposed framework to existing SOTA models on public datasets and for grounding the critique in neurologist priorities, though its impact depends on establishing a concrete link to deployment outcomes.

major comments (2)
  1. [Abstract and experimental analysis] Abstract and experimental analysis section: The central claim that the additional metrics and problem fingerprinting better quantify usability for hospital deployment is not supported by evidence. The analysis shows that models differ on lesion-wise and clinical-context metrics compared to Dice, but provides no correlation study, neurologist preference data, inter-rater variability analysis in complex cases, or downstream monitoring accuracy results to demonstrate that adopting these metrics would change model selection or improve real-world outcomes over Dice-only evaluation.
  2. [Problem fingerprinting] Problem fingerprinting section: The description of neurologist priorities and required metrics draws from stated clinical domain knowledge but lacks citations to specific studies, surveys of neurologists, or empirical validation, leaving the completeness of the fingerprint and the choice of proposed metrics open to question as a foundation for the evaluation rethink.
minor comments (2)
  1. [Metrics definitions] Ensure all newly proposed metrics are accompanied by precise mathematical definitions or pseudocode to support reproducibility by other researchers.
  2. [Experimental setup] Clarify the exact lesion-wise metrics used (e.g., lesion detection sensitivity/specificity thresholds) and how they are aggregated across datasets in the reported results.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and indicate where revisions will be made to the manuscript.

read point-by-point responses

  1. Referee: [Abstract and experimental analysis] Abstract and experimental analysis section: The central claim that the additional metrics and problem fingerprinting better quantify usability for hospital deployment is not supported by evidence. The analysis shows that models differ on lesion-wise and clinical-context metrics compared to Dice, but provides no correlation study, neurologist preference data, inter-rater variability analysis in complex cases, or downstream monitoring accuracy results to demonstrate that adopting these metrics would change model selection or improve real-world outcomes over Dice-only evaluation.

    Authors: We agree that direct evidence, such as correlation studies with deployment outcomes or neurologist preference data, would provide stronger validation for the claim that these metrics improve real-world usability. Our analysis on two public datasets shows that SOTA models exhibit different performance profiles under lesion-wise detection and clinical-context metrics (e.g., small lesion detection and complex cases) compared to Dice, indicating that Dice-only evaluation may not fully capture aspects relevant to early detection and progression monitoring. The problem fingerprinting framework is presented to systematically identify such priorities from clinical needs. We will revise the abstract, introduction, and discussion to clarify that the metrics are motivated by established clinical requirements and that their adoption could inform better model selection, while explicitly noting that empirical validation against downstream clinical outcomes remains an important avenue for future research. revision: partial

  2. Referee: [Problem fingerprinting] Problem fingerprinting section: The description of neurologist priorities and required metrics draws from stated clinical domain knowledge but lacks citations to specific studies, surveys of neurologists, or empirical validation, leaving the completeness of the fingerprint and the choice of proposed metrics open to question as a foundation for the evaluation rethink.

    Authors: We acknowledge that additional citations would strengthen the grounding of the problem fingerprinting. The described priorities, including the emphasis on lesion detection for disease activity monitoring and handling of complex cases, align with standard MS clinical practices. We will revise the problem fingerprinting section to incorporate specific references to supporting literature, such as studies on MRI lesion criteria in MS diagnosis and monitoring (e.g., McDonald criteria updates and clinical trial endpoints focused on new/enlarging lesions), as well as works on inter-rater variability in lesion annotation. This will provide a more explicit empirical basis for the selected metrics. revision: yes

Circularity Check

0 steps flagged

No circularity; conceptual proposal independent of inputs

full rationale

The paper is a position and analysis piece that argues for expanded evaluation metrics in MS lesion segmentation based on stated clinical priorities for neurologists. It introduces 'problem fingerprinting' as a descriptive framework and reports metric differences on two public datasets for existing models. No equations, derivations, fitted parameters, or self-citation chains appear in the provided text or abstract. The argument draws from external domain knowledge about lesion detection needs rather than reducing any claim to its own inputs by construction. This qualifies as a self-contained non-circular contribution under the evaluation criteria.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a position and critique paper focused on evaluation practices rather than a mathematical or empirical derivation. No free parameters, axioms, or invented entities are introduced or required.

pith-pipeline@v0.9.0 · 5496 in / 1204 out tokens · 52402 ms · 2026-05-12T04:17:27.562627+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

59 extracted references · 59 canonical work pages

  1. [1]

    Multiple sclerosis: pathogenesis, symptoms, diagnoses and cell-based therapy,

    N. Ghasemiet al., “Multiple sclerosis: pathogenesis, symptoms, diagnoses and cell-based therapy,”Cell Journal (Yakhteh), vol. 19, no. 1, p. 1, 2016

    work page 2016

  2. [2]

    Review of multiple sclerosis: Epidemiology, etiology, pathophysiology, and treatment,

    M. Hakiet al., “Review of multiple sclerosis: Epidemiology, etiology, pathophysiology, and treatment,”Medicine, vol. 103, no. 8, p. e37297, 2024

    work page 2024

  3. [3]

    Evidence-based guidelines: Magnims consensus guidelines on the use of mri in multiple sclerosis-establishing disease prognosis and monitoring patients,

    M. P. Wattjeset al., “Evidence-based guidelines: Magnims consensus guidelines on the use of mri in multiple sclerosis-establishing disease prognosis and monitoring patients,”Nature Reviews. Neurology, vol. 11, pp. 597–606, 2015

    work page 2015

  4. [4]

    nnu-net: Self-adapting framework for u-net-based medical image segmentation,

    F. Isenseeet al., “nnu-net: Self-adapting framework for u-net-based medical image segmentation,” 2018

    work page 2018

  5. [5]

    3d mri brain tumor segmentation using autoencoder regularization,

    A. Myronenko, “3d mri brain tumor segmentation using autoencoder regularization,” 2018

    work page 2018

  6. [6]

    Swin unetr: Swin transformers for semantic segmentation of brain tumors in mri images,

    A. Hatamizadehet al., “Swin unetr: Swin transformers for semantic segmentation of brain tumors in mri images,” 2022

    work page 2022

  7. [7]

    Segmentation of multiple sclerosis lesions in intensity corrected multispectral mri,

    B. Johnstonet al., “Segmentation of multiple sclerosis lesions in intensity corrected multispectral mri,”IEEE transactions on medical imaging, vol. 15 2, pp. 154–69, 1996

    work page 1996

  8. [8]

    Spatial decision forests for ms lesion segmentation in multi-channel mr images,

    E. Geremiaet al., “Spatial decision forests for ms lesion segmentation in multi-channel mr images,”International Conference on Medical Image Computing and Computer-Assisted Intervention, vol. 13 Pt 1, 2010

    work page 2010

  9. [9]

    Multi-sectional views textural based svm for ms lesion segmentation in multi-channels mris,

    B. A. Abdullahet al., “Multi-sectional views textural based svm for ms lesion segmentation in multi-channels mris,”The Open Biomedical Engineering Journal, vol. 6, pp. 56 – 72, 2012

    work page 2012

  10. [10]

    Deep 3d convolutional encoder networks with shortcuts for multiscale feature integration applied to multiple sclerosis lesion segmentation,

    T. Broschet al., “Deep 3d convolutional encoder networks with shortcuts for multiscale feature integration applied to multiple sclerosis lesion segmentation,”IEEE Transactions on Medical Imaging, vol. 35, 2016

    work page 2016

  11. [11]

    Mri flair lesion segmentation in multiple sclerosis: Does automated segmentation hold up with manual annotation?

    C. Eggeret al., “Mri flair lesion segmentation in multiple sclerosis: Does automated segmentation hold up with manual annotation?”NeuroImage : Clinical, vol. 13, pp. 264 – 270, 2016

    work page 2016

  12. [12]

    Two time point ms lesion segmentation in brain mri: An expectation-maximization framework,

    S. Jainet al., “Two time point ms lesion segmentation in brain mri: An expectation-maximization framework,”Frontiers in Neuroscience, vol. 10, 2016

    work page 2016

  13. [13]

    Improving automated multiple sclerosis lesion segmentation with a cascaded 3d convolutional neural network approach,

    S. Valverdeet al., “Improving automated multiple sclerosis lesion segmentation with a cascaded 3d convolutional neural network approach,” NeuroImage, vol. 155, pp. 159–168, 2017

    work page 2017

  14. [14]

    Longitudinal multiple sclerosis lesion segmentation: Resource and challenge,

    A. Carasset al., “Longitudinal multiple sclerosis lesion segmentation: Resource and challenge,”NeuroImage, vol. 148, pp. 77–102, 2017

    work page 2017

  15. [15]

    Multi-view longitudinal cnn for multiple sclerosis lesion segmentation,

    A. Birenbaumet al., “Multi-view longitudinal cnn for multiple sclerosis lesion segmentation,”Eng. Appl. Artif. Intell., vol. 65, pp. 111–118, 2017

    work page 2017

  16. [16]

    Neural network-based learning kernel for auto- matic segmentation of multiple sclerosis lesions on magnetic resonance images,

    H. Khastavanehet al., “Neural network-based learning kernel for auto- matic segmentation of multiple sclerosis lesions on magnetic resonance images,”Journal of Biomedical Physics & Engineering, vol. 7, 2017

    work page 2017

  17. [17]

    Multi-branch convolutional neural network for multiple sclerosis lesion segmentation,

    S. Aslaniet al., “Multi-branch convolutional neural network for multiple sclerosis lesion segmentation,”NeuroImage, vol. 196, pp. 1–15, 2018

    work page 2018

  18. [18]

    Objective evaluation of multiple sclerosis lesion segmentation using a data management and processing infrastructure,

    O. Commowicket al., “Objective evaluation of multiple sclerosis lesion segmentation using a data management and processing infrastructure,” Scientific Reports, vol. 8, 2018

    work page 2018

  19. [19]

    Brain and lesion segmentation in multiple sclerosis using fully convolutional neural networks: A large-scale study,

    R. Gabret al., “Brain and lesion segmentation in multiple sclerosis using fully convolutional neural networks: A large-scale study,”Multiple Sclerosis Journal, vol. 26, pp. 1217 – 1226, 2019

    work page 2019

  20. [20]

    Comparing lesion segmentation methods in multiple sclerosis: Input from one manually delineated subject is sufficient for accurate lesion segmentation,

    M. M. Weedaet al., “Comparing lesion segmentation methods in multiple sclerosis: Input from one manually delineated subject is sufficient for accurate lesion segmentation,”NeuroImage : Clinical, vol. 24, 2019. Fig. 4.nnU-Net Performance Evaluation on MSSEG-1(A)Lesion Performance Scatter: Correlates Dice scores with lesion size, where color indicates the P...

    work page 2019

  21. [21]

    Multiple sclerosis lesion segmentation with tiramisu and 2.5d stacked slices,

    H. Zhanget al., “Multiple sclerosis lesion segmentation with tiramisu and 2.5d stacked slices,”International Conference on Medical Image Computing and Computer-Assisted Intervention, vol. 11766, pp. 338–346, 2019

    work page 2019

  22. [22]

    Rsanet: Recurrent slice-wise attention network for multiple sclerosis lesion segmentation,

    ——, “Rsanet: Recurrent slice-wise attention network for multiple sclerosis lesion segmentation,”ArXiv, vol. abs/2002.12470, 2019

    work page arXiv 2002

  23. [23]

    Simultaneous lesion and neuroanatomy segmen- tation in multiple sclerosis using deep neural networks,

    R. McKinleyet al., “Simultaneous lesion and neuroanatomy segmen- tation in multiple sclerosis using deep neural networks,”ArXiv, vol. abs/1901.07419, 2019

    work page arXiv 1901

  24. [24]

    A contrast-adaptive method for simultaneous whole-brain and lesion segmentation in multiple sclerosis,

    S. Cerriet al., “A contrast-adaptive method for simultaneous whole-brain and lesion segmentation in multiple sclerosis,”NeuroImage, vol. 225, pp. 117 471 – 117 471, 2020

    work page 2020

  25. [25]

    Spatio-temporal learning from longitudinal data for multiple sclerosis lesion segmentation,

    S. Denneret al., “Spatio-temporal learning from longitudinal data for multiple sclerosis lesion segmentation,” inBrainLes@MICCAI, 2020

    work page 2020

  26. [26]

    State-of-the-art segmentation techniques and future directions for multiple sclerosis brain lesions,

    A. Kauret al., “State-of-the-art segmentation techniques and future directions for multiple sclerosis brain lesions,”Archives of Computational Methods in Engineering, vol. 28, pp. 951 – 977, 2020

    work page 2020

  27. [27]

    All-net: Anatomical information lesion-wise loss function integrated into neural network for multiple sclerosis lesion segmentation,

    H. Zhanget al., “All-net: Anatomical information lesion-wise loss function integrated into neural network for multiple sclerosis lesion segmentation,”NeuroImage : Clinical, vol. 32, 2021

    work page 2021

  28. [28]

    Multiple sclerosis lesion analysis in brain magnetic resonance images: Techniques and clinical applications,

    Y . Maet al., “Multiple sclerosis lesion analysis in brain magnetic resonance images: Techniques and clinical applications,”IEEE Journal of Biomedical and Health Informatics, vol. 26, pp. 2680–2692, 2021

    work page 2021

  29. [29]

    Multiple sclerosis lesion segmentation in brain mri using inception modules embedded in a convolutional neural network,

    S. U. Ansariet al., “Multiple sclerosis lesion segmentation in brain mri using inception modules embedded in a convolutional neural network,” Journal of Healthcare Engineering, vol. 2021, 2021

    work page 2021

  30. [30]

    Simultaneous lesion and brain segmentation in multiple sclerosis using deep neural networks,

    R. McKinleyet al., “Simultaneous lesion and brain segmentation in multiple sclerosis using deep neural networks,”Scientific Reports, vol. 11, 2021

    work page 2021

  31. [31]

    An open-source tool for longitudinal whole-brain and white matter lesion segmentation,

    S. Cerriet al., “An open-source tool for longitudinal whole-brain and white matter lesion segmentation,”NeuroImage : Clinical, vol. 38, 2022

    work page 2022

  32. [32]

    Framework to segment and evaluate multiple sclerosis lesion in mri slices using vgg-unet,

    S. Krishnamoorthyet al., “Framework to segment and evaluate multiple sclerosis lesion in mri slices using vgg-unet,”Computational Intelligence and Neuroscience, vol. 2022, 2022

    work page 2022

  33. [33]

    Multiple sclerosis lesions segmentation using attention-based cnns in flair images,

    M. SadeghiBakhiet al., “Multiple sclerosis lesions segmentation using attention-based cnns in flair images,”IEEE Journal of Translational Engineering in Health and Medicine, vol. 10, 2022

    work page 2022

  34. [34]

    Noise invariant convolution neural network for segmentation of multiple sclerosis lesions from brain magnetic resonance imaging,

    S. M. Det al., “Noise invariant convolution neural network for segmentation of multiple sclerosis lesions from brain magnetic resonance imaging,”International Journal of Online and Biomedical Engineering (iJOE), 2022

    work page 2022

  35. [35]

    Triplanar u-net with lesion-wise voting for the segmentation of new lesions on longitudinal mri studies,

    S. Hitzigeret al., “Triplanar u-net with lesion-wise voting for the segmentation of new lesions on longitudinal mri studies,”Frontiers in Neuroscience, vol. 16, 2022

    work page 2022

  36. [36]

    Using convolutional neural networks for segmen- tation of multiple sclerosis lesions in 3d magnetic resonance imaging,

    A. A. Hamadet al., “Using convolutional neural networks for segmen- tation of multiple sclerosis lesions in 3d magnetic resonance imaging,” Advances in Materials Science and Engineering, 2022

    work page 2022

  37. [37]

    Boosting multiple sclerosis lesion segmentation through attention mechanism,

    A. Rondinellaet al., “Boosting multiple sclerosis lesion segmentation through attention mechanism,”Computers in biology and medicine, vol. 161, p. 107021, 2023

    work page 2023

  38. [38]

    Coactseg: Learning from heterogeneous data for new multiple sclerosis lesion segmentation,

    Y . Wuet al., “Coactseg: Learning from heterogeneous data for new multiple sclerosis lesion segmentation,” inInternational Conference on Medical Image Computing and Computer-Assisted Intervention, 2023

    work page 2023

  39. [39]

    Enhancing multiple sclerosis lesion segmentation Fig. 5.nnU-Net Performance Evaluation on MSLesSeg: The same structure as Figure 4 in multimodal mri scans with diffusion models,

    A. Rondinellaet al., “Enhancing multiple sclerosis lesion segmentation Fig. 5.nnU-Net Performance Evaluation on MSLesSeg: The same structure as Figure 4 in multimodal mri scans with diffusion models,”2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2023

    work page 2023

  40. [40]

    Multiple sclerosis lesion segmentation: revisiting weighting mechanisms for federated learning,

    D. Liuet al., “Multiple sclerosis lesion segmentation: revisiting weighting mechanisms for federated learning,”Frontiers in Neuroscience, vol. 17, 2023

    work page 2023

  41. [41]

    Scanner agnostic large-scale evaluation of ms lesion delineation tool for clinical mri,

    A. Hindsholmet al., “Scanner agnostic large-scale evaluation of ms lesion delineation tool for clinical mri,”Frontiers in Neuroscience, vol. 17, 2023

    work page 2023

  42. [42]

    Toward more accurate diagnosis of multiple sclerosis: Automated lesion segmentation in brain magnetic resonance image using modified u-net model,

    B. Amaludinet al., “Toward more accurate diagnosis of multiple sclerosis: Automated lesion segmentation in brain magnetic resonance image using modified u-net model,”International Journal of Imaging Systems and Technology, vol. 34, 2023

    work page 2023

  43. [43]

    Towards an accurate and generalizable multiple sclerosis lesion segmentation model using self-ensembled lesion fusion,

    J. Zhanget al., “Towards an accurate and generalizable multiple sclerosis lesion segmentation model using self-ensembled lesion fusion,”2024 IEEE International Symposium on Biomedical Imaging (ISBI), 2023

    work page 2024

  44. [44]

    Consensus of algorithms for lesion segmentation in brain mri studies of multiple sclerosis,

    A. P. D. Rosaet al., “Consensus of algorithms for lesion segmentation in brain mri studies of multiple sclerosis,”Scientific Reports, vol. 14, 2024

    work page 2024

  45. [45]

    Diagnosis of multiple sclerosis lesion using deep learning models,

    S. Balanet al., “Diagnosis of multiple sclerosis lesion using deep learning models,”Journal of Electrical Systems, 2024

    work page 2024

  46. [46]

    Icpr 2024 competition on multiple sclerosis lesion segmentation - methods and results,

    A. Rondinellaet al., “Icpr 2024 competition on multiple sclerosis lesion segmentation - methods and results,” inInternational Conference on Pattern Recognition, 2024

    work page 2024

  47. [47]

    Longitudinal segmentation of ms lesions via temporal difference weighting,

    M. Rokusset al., “Longitudinal segmentation of ms lesions via temporal difference weighting,” inLDTM/MMMI/ML4MHD/ML-CDS@MICCAI, 2024

    work page 2024

  48. [48]

    Lst-ai: A deep learning ensemble for accurate ms lesion segmentation,

    T. Wiltgenet al., “Lst-ai: A deep learning ensemble for accurate ms lesion segmentation,”NeuroImage : Clinical, vol. 42, 2024

    work page 2024

  49. [49]

    Seghed+: Segmentation of heterogeneous data for multiple sclerosis lesions with anatomical constraints and lesion-aware augmentation,

    B. D. Basaranet al., “Seghed+: Segmentation of heterogeneous data for multiple sclerosis lesions with anatomical constraints and lesion-aware augmentation,”ArXiv, vol. abs/2412.10946, 2024

    work page arXiv 2024

  50. [50]

    Seghed: Segmentation of heterogeneous data for multiple sclerosis lesions with anatomical constraints,

    ——, “Seghed: Segmentation of heterogeneous data for multiple sclerosis lesions with anatomical constraints,”ArXiv, vol. abs/2410.01766, 2024

    work page arXiv 2024

  51. [51]

    A novel convolutional neural network for automated multiple sclerosis brain lesion segmentation,

    E. Dereskewiczet al., “A novel convolutional neural network for automated multiple sclerosis brain lesion segmentation,”Journal of Neuroimaging, vol. 35, 2025

    work page 2025

  52. [52]

    Enhancing precision in multiple sclerosis lesion segmentation: A u-net based machine learning approach with data augmentation,

    O. Cetinet al., “Enhancing precision in multiple sclerosis lesion segmentation: A u-net based machine learning approach with data augmentation,”Neuroimage: Reports, vol. 5, 2025

    work page 2025

  53. [53]

    Flames: A robust deep learning model for automated multiple sclerosis lesion segmentation,

    E. Dereskewiczet al., “Flames: A robust deep learning model for automated multiple sclerosis lesion segmentation,”medRxiv, 2025

    work page 2025

  54. [54]

    Mslesseg: baseline and benchmarking of a new multiple sclerosis lesion segmentation dataset,

    F. Guarneraet al., “Mslesseg: baseline and benchmarking of a new multiple sclerosis lesion segmentation dataset,”Scientific Data, vol. 12, 2025

    work page 2025

  55. [55]

    Enhanced segmentation of active and nonactive mul- tiple sclerosis plaques in t1 and flair mri images using transformer-based encoders,

    N. Davaraniet al., “Enhanced segmentation of active and nonactive mul- tiple sclerosis plaques in t1 and flair mri images using transformer-based encoders,”International Journal of Imaging Systems and Technology, vol. 35, 05 2025

    work page 2025

  56. [56]

    Shallow vs deep learning architectures for white matter lesion segmentation in the early stages of multiple sclerosis,

    F. La Rosaet al., “Shallow vs deep learning architectures for white matter lesion segmentation in the early stages of multiple sclerosis,” in International MICCAI Brainlesion Workshop. Springer, 2018

    work page 2018

  57. [57]

    Multi-scale convolutional-stack aggregation for robust white matter hyperintensities segmentation,

    H. Liet al., “Multi-scale convolutional-stack aggregation for robust white matter hyperintensities segmentation,” inInternational MICCAI Brainlesion Workshop. Springer, 2018, pp. 199–207

    work page 2018

  58. [58]

    Survey of the distribution of lesion size in multiple sclerosis: implication for the measurement of total lesion load,

    L. Wanget al., “Survey of the distribution of lesion size in multiple sclerosis: implication for the measurement of total lesion load,”Journal of Neurology, Neurosurgery & Psychiatry, vol. 63, no. 4, 1997

    work page 1997

  59. [59]

    Lst-ai: a deep learning ensemble for accurate ms lesion segmentation,

    T. Wiltgenet al., “Lst-ai: a deep learning ensemble for accurate ms lesion segmentation,”medRxiv, 2023

    work page 2023