Machine Learning in Optic Neuritis Differentiation: Human Study
In a study of 45 patients with optic neuritis related to multiple sclerosis and MOG antibody disease, supervised machine-learning models using radiomic features were found to better distinguish between conditions than those using clinical variables alone (AUC 0.763 vs. 0.584, P = 0.001). Combining both approaches did yield a higher score (AUC 0.788) but not statistically significant (P = 0.361).
- Tags
- #MSResearch
- Most relevant to
- —
- Human efficacy
- Not established from this source
- Mechanism
- Not reported
- Evidence strength
- Limited
- Outcome
- Positive
Why it matters
These findings suggest that machine learning utilizing radiomic data may enhance diagnostic accuracy in differentiating optic neuritis types, which is crucial for understanding the underlying cause more accurately and potentially guiding treatment strategies.
What this does not prove
However, the study is retrospective, with a small sample size of 45 patients, meaning the results may not apply broadly to other populations.
Next milestone
No next milestone was established from the available source.
Study facts
- Study design
- Observational
- Participants / samples
- 45 · basis not reported
- Randomised
- Not reported
- Controlled
- Not reported
- Primary endpoint met
- Yes
- Relevant MS type
- Not reported
- Publication date
- 2026-09-07
- Evidence reviewed
- Abstract only
- Regulatory approval
- Not reported
- Research areas
- Not classified
- Biological target
- Not reported
- Pathway
- Not reported
- Population
- Patients with optic neuritis associated with multiple sclerosis or MOGAD
- Size of effect
- Not reported
- Secondary endpoints
- Not reported
- MRI endpoints
- Not reported
- Biomarkers
- Not reported
- Safety
- Not reported
Original sources
- Primary evidenceOCT-Derived Radiomic Features for Differentiating Optic Neuritis Etiologies: An Unsupervised and Supervised Machine-Learning Approach. ↗DOI: 10.1016/j.xops.2026.101401
Supporting passages (11)
study phaseDESIGN: Retrospective cohort study.
study designDESIGN: Retrospective cohort study.
subjectsPARTICIPANTS: Patients with ON associated with either MS (MS-ON) or MOGAD-ON were selected from the existing Optic Neuropathies of la Timone cohort.
sample sizeIn total, 45 patients (61 eyes with ON; 28 with MS and 17 with MOGAD) contributing 266 OCT examinations were included.
peer reviewedPublication types: Journal Article
primary endpoint metMAIN 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).
findingsSupervised machine-learning models (logistic regression) using radiomic features outperformed models based on clinical variables alone (AUC, 0.763 vs. 0.584; P = 0.001).
populationPARTICIPANTS: Patients with ON associated with either MS (MS-ON) or MOGAD-ON were selected from the existing Optic Neuropathies of la Timone cohort.
outcomeRadiomic features show promising potential for improving the differentiation between MOGAD-ON and MS-ON.
publication datePublication date: 2026-09-07
limitationsTitle: OCT-Derived Radiomic Features for Differentiating Optic Neuritis Etiologies: An Unsupervised and Supervised Machine-Learning Approach.
AI assessment, not yet reviewed by a person · version 1 · Community votes are separate from evidence review.