Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 35 min ago
Peer-reviewedCC BY 4.0Ophthalmology Science

OCT-Derived Radiomic Features for Differentiating Optic Neuritis Etiologies

  • Hoda El Mohyne1
  • Thierry David2,3
  • Kevin Mairot2,4
  • Sarah Demortière5
  • Shailesh Appukuttan1
  • Mario Borrelli2
  • Bertrand Audoin1,5
  • Jan-Patrick Stellmann1,6
  • Natacha Stolowy1,2
6 affiliations
  1. 1Aix Marseille Université, Centre d'Exploration Métabolique par Résonance Magnétique-Center for Magnetic Resonance in Biology and Medicine, Centre National de la Recherche Scientifique, Marseille, France
  2. 2Department of Ophthalmology, Assistance Publique Hôpitaux de Marseille, Hopital de la Conception, Marseille France
  3. 3Aix Marseille Université, Centre National de la Recherche Scientifique, Marseille, France
  4. 4Aix Marseille Université, Institut de Neurosciences de la Timone, Centre National de la Recherche Scientifique, Marseille, France
  5. 5Department of Neurology, Assistance Publique Hôpitaux de Marseille, Hopital de la Timone, Marseille, France
  6. 6Department of Radiology, Assistance Publique Hôpitaux de Marseille, Hopital de la Timone, Marseille, France
Read full textPDFDOICite

Abstract

Purpose: To assess the feasibility of an OCT-based radiomics approach in optic neuritis (ON) and determine whether peripapillary retinal nerve fiber layer (pRNFL) OCT-derived radiomic features could help differentiate between multiple sclerosis–associated ON (MS-ON) and myelin oligodendrocyte glycoprotein antibody–associated disease ON (MOGAD-ON).

Design: Retrospective cohort study.

Participants: Patients with ON associated with either MS (MS-ON) or MOGAD-ON were selected from the existing Optic Neuropathies of la Timone cohort. In total, 45 patients (61 eyes with ON; 28 with MS and 17 with MOGAD) contributing 266 OCT examinations were included.

Methods: A radiomics pipeline was applied to pRNFL OCT data to extract quantitative imaging features. A total of 102 radiomic features were generated and reduced using feature selection methods. Unsupervised and supervised machine-learning approaches were used to identify relevant radiomic patterns and evaluate their ability to discriminate MS-ON from MOGAD-ON. Models were trained using radiomic features alone, clinical variables alone, and a combination of both.

Main Outcome Measures: The primary outcome measure was the diagnostic performance of machine-learning models in distinguishing MS-ON from MOGAD-ON, assessed using the area under the receiver operating characteristic curve (AUC).

Results: Among the 102 extracted radiomic features, 62 were retained after feature reduction, and 3 features were selected through machine-learning–based analysis. Supervised machine-learning models (logistic regression) using radiomic features outperformed models based on clinical variables alone (AUC, 0.763 vs. 0.584; P = 0.001). Combining radiomics with clinical variables yielded a numerically higher AUC of 0.788 than radiomics alone (0.763), but without statistical significance (P = 0.361). The combined model remained significantly better than the clinical-variable-only model (P < 0.001).

Conclusions: This study establishes a reproducible OCT radiomics pipeline and demonstrates the feasibility of applying unsupervised and supervised machine-learning approaches to pRNFL OCT data. Radiomic features show promising potential for improving the differentiation between MOGAD-ON and MS-ON.

Financial Disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Related

  1. Evaluation of MRI Radiomic Features as Surrogates of Tumor Immune Status in Breast Cancer: A Pilot Study of MRI-Derived Radiomic Features for Non-Invasive Immune Profiling in Breast Cancer.
    Mimansha Shah, Soaham Kumar, Narendra Kumar +1
    CancersPeer-reviewedCC BYFull textMedical imaging & radiologyEurope PMC
  2. CT radiomic features detect mineral, collagen, and water contributions to cortical bone texture
    Youngjun Lee, Wikum R. Bandara, Peter K. Jalaie +6
    Scientific ReportsPeer-reviewedCC BY-NC-ND 4.0Full textMedical imaging & radiologyCrossref
  3. Hybrid deep learning and meta learning for Alzheimer disease classification via multimodal radiomic and visual features
    Areej Y. Bayahya, Fares Jammal
    Scientific ReportsPeer-reviewedCC BYFull textPsychiatry & mental healthDOAJ
  4. Effect of Intensity Discretization on Radiomic Features in En Face EDI-OCT of Choroidal Nevi
    M. Vila Franca, J. P. Nobre, D. Saraiva +4
    medRxivPreprintCC BY-NC-NDFull textGeneral medicinemedRxiv
  5. Lobar analysis of classical and deep learning radiomic features in chronic airflow limitation
    Emelie Bäcklin, Birgitta Janerot-Sjoberg, Örjan Smedby +2
    BMC Medical ImagingPeer-reviewedCC BY 4.0Full textMedical imaging & radiologyPMC

Cite

BibTeXRIS
BibTeX
@article{mohyne2026octderived,
  title = {{OCT-Derived Radiomic Features for Differentiating Optic Neuritis Etiologies}},
  author = {Hoda El Mohyne and Thierry David and Kevin Mairot and Sarah Demortière and Shailesh Appukuttan and Mario Borrelli and Bertrand Audoin and Jan-Patrick Stellmann and Natacha Stolowy},
  journal = {Ophthalmology Science},
  year = {2026},
  month = sep,
  doi = {10.1016/j.xops.2026.101401},
  url = {https://doi.org/10.1016/j.xops.2026.101401},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - OCT-Derived Radiomic Features for Differentiating Optic Neuritis Etiologies
AU  - Hoda El Mohyne
AU  - Thierry David
AU  - Kevin Mairot
AU  - Sarah Demortière
AU  - Shailesh Appukuttan
AU  - Mario Borrelli
AU  - Bertrand Audoin
AU  - Jan-Patrick Stellmann
AU  - Natacha Stolowy
PY  - 2026
DA  - 2026/09/07
JO  - Ophthalmology Science
DO  - 10.1016/j.xops.2026.101401
UR  - https://doi.org/10.1016/j.xops.2026.101401
AB  - Purpose: To assess the feasibility of an OCT-based radiomics approach in optic neuritis (ON) and determine whether peripapillary retinal nerve fiber layer (pRNFL) OCT-derived radiomic features could help differentiate between multiple sclerosis–associated ON (MS-ON) and myelin oligodendrocyte glycoprotein antibody–associated disease ON (MOGAD-ON).  Design: Retrospective cohort study.  Participants: Patients with ON associated with either MS (MS-ON) or MOGAD-ON were selected from the existing Optic Neuropathies of la Timone cohort. In total, 45 patients (61 eyes with ON; 28 with MS and 17 with MOGAD) contributing 266 OCT examinations were included.  Methods: A radiomics pipeline was applied to pRNFL OCT data to extract quantitative imaging features. A total of 102 radiomic features were generated and reduced using feature selection methods. Unsupervised and supervised machine-learning approaches were used to identify relevant radiomic patterns and evaluate their ability to discriminate MS-ON from MOGAD-ON. Models were trained using radiomic features alone, clinical variables alone, and a combination of both.  Main Outcome Measures: The primary outcome measure was the diagnostic performance of machine-learning models in distinguishing MS-ON from MOGAD-ON, assessed using the area under the receiver operating characteristic curve (AUC).  Results: Among the 102 extracted radiomic features, 62 were retained after feature reduction, and 3 features were selected through machine-learning–based analysis. Supervised machine-learning models (logistic regression) using radiomic features outperformed models based on clinical variables alone (AUC, 0.763 vs. 0.584; P = 0.001). Combining radiomics with clinical variables yielded a numerically higher AUC of 0.788 than radiomics alone (0.763), but without statistical significance (P = 0.361). The combined model remained significantly better than the clinical-variable-only model (P < 0.001).  Conclusions: This study establishes a reproducible OCT radiomics pipeline and demonstrates the feasibility of applying unsupervised and supervised machine-learning approaches to pRNFL OCT data. Radiomic features show promising potential for improving the differentiation between MOGAD-ON and MS-ON.  Financial Disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
SN  - 2666-9145
ER  -