Retinal inter-eye difference and atrophy progression in multiple sclerosis diagnostics

Diagnosis of multiple sclerosis using optical coherence tomography supported by explainable artificial intelligence

BACKGROUND/OBJECTIVES

Study of retinal structure based on optical coherence tomography (OCT) data can facilitate early diagnosis of relapsing-remitting multiple sclerosis (RRMS). Although artificial intelligence can provide highly reliable diagnoses, the results obtained must be explainable.

SUBJECTS/METHODS

The study included 79 recently diagnosed RRMS patients and 69 age matched healthy control subjects. Thickness (Avg) and inter-eye difference (Diff) features are obtained in 4 retinal layers using the posterior pole protocol. Each layer is divided into six analysis zones. The Support Vector Machine plus Recursive Feature Elimination with Leave-One-Out Cross Validation (SVM-RFE-LOOCV) approach is used to find the subset of features that reduces dimensionality and optimises the performance of the classifier.

RESULTS

SVM-RFE-LOOCV was used to identify OCT features with greatest capacity for early diagnosis, determining the area of the papillomacular bundle to be the most influential. A correlation was observed between loss of layer thickness and increase in functional disability. There was also greater functional deterioration in patients with greater asymmetry between left and right eyes. The classifier based on the top-ranked features obtained sensitivity = 0.86 and specificity = 0.90.

CONCLUSIONS

There was consistency between the features identified as relevant by the SVM-RFE-LOOCV approach and the retinotopic distribution of the retinal nerve fibres and the optic nerve head. This simple method contributes to implementation of an assisted diagnosis system and its accuracy exceeds that achieved with magnetic resonance imaging of the central nervous system, the current gold standard. This paper provides novel insights into RRMS affectation of the neuroretina.

Delimiting MOGAD as a disease entity using translational imaging

The first formal consensus diagnostic criteria for myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) were recently proposed. Yet, the distinction of MOGAD-defining characteristics from characteristics of its important differential diagnoses such as multiple sclerosis (MS) and aquaporin-4 antibody seropositive neuromyelitis optica spectrum disorder (NMOSD) is still obstructed. In preclinical research, MOG antibody-based animal models were used for decades to derive knowledge about MS. In clinical research, people with MOGAD have been combined into cohorts with other diagnoses. Thus, it remains unclear to which extent the generated knowledge is specifically applicable to MOGAD. Translational research can contribute to identifying MOGAD characteristic features by establishing imaging methods and outcome parameters on proven pathophysiological grounds. This article reviews suitable animal models for translational MOGAD research and the current state and prospect of translational imaging in MOGAD.

Diagnostic Performance of Adding the Optic Nerve Region Assessed by Optical Coherence Tomography to the Diagnostic Criteria for Multiple Sclerosis

BACKGROUND AND OBJECTIVES

The optic nerve has been recommended as an additional region for demonstrating dissemination in space (DIS) in diagnostic criteria for multiple sclerosis (MS). The aim of this study was to investigate whether adding the optic nerve region as determined by optical coherence tomography (OCT) as part of the DIS criteria improves the 2017 diagnostic criteria.

METHODS

From a prospective observational study, we included patients with a first demyelinating event who had complete information to assess DIS and a spectral domain OCT scan obtained within 180 days. Modified DIS criteria (DIS + OCT) were constructed by adding the optic nerve to the current DIS regions based on validated thresholds for OCT intereye differences. Time to second clinical attack was the primary endpoint.

RESULTS

We analyzed 267 patients with MS (mean age 31.3 years [SD 8.1], 69% female) during a median observation period of 59 months (range: 13-98). Adding the optic nerve as a fifth region improved the diagnostic performance by increasing accuracy (DIS + OCT 81.2% vs DIS 65.6%) and sensitivity (DIS + OCT 84.2% vs DIS 77.9%) without lowering specificity (DIS + OCT 52.2% vs DIS 52.2%). Fulfilling DIS + OCT criteria (≥2 of 5 DIS + OCT regions involved) indicated a similar risk of a second clinical attack (hazard ratio [HR] 3.6, CI 1.4-14.5) compared with a 2.5-fold increased risk when fulfilling DIS criteria (HR 2.5, CI 1.2-11.8). When the analysis was conducted according to topography of the first demyelinating event, DIS + OCT criteria performed similarly in both optic neuritis and nonoptic neuritis.

DISCUSSION

Addition of the optic nerve, assessed by OCT, as a fifth region in the current DIS criteria improves diagnostic performance by increasing sensitivity without lowering specificity.

CLASSIFICATION OF EVIDENCE

This study provides Class II evidence that adding the optic nerve as determined by OCT as a fifth DIS criterion to the 2017 McDonald criteria improves diagnostic accuracy.

Longitudinal stability of inter-eye differences in optical coherence tomography measures for identifying unilateral optic nerve lesions in multiple sclerosis

INTRODUCTION

Optical coherence tomography (OCT)-derived peripapillary retinal nerve fiber layer (pRNFL) and ganglion cell+inner plexiform layer (GCIPL) thickness inter-eye differences (IEDs) are robust measurements for identifying clinical history acute ON in people with MS (PwMS). This study investigated the utility and durability of these measures as longitudinal markers to identify optic nerve lesions.

METHODS

Prospective, multi-center international study of PwMS (with/without clinical history of ON) and healthy controls. Data from two sites in the International MS Visual System Consortium (IMSVISUAL) were analyzed. Mixed-effects models were used to compare inter-eye differences based on MS and acute ON history.

RESULTS

Average age of those with MS (n = 210) was 39.1 ± 10.8 and 190 (91%) were relapsing-remitting. Fifty-nine (28.1%) had a history of acute unilateral ON, while 9/210 (4.3%) had >1 IB episode. Median follow-up between OCT scans was 9 months. By mixed-effects modeling, IEDs were stable between first and last visits within groups for GCIPL for controls (p = 0.18), all PwMS (p = 0.74), PwMs without ON (p = 0.22), and PwMS with ON (p = 0.48). For pRNFL, IEDs were within controls (p = 0.10), all PwMS (p = 0.53), PwMS without ON history (p = 0.98), and PwMS with history of ON (p = 0.81).

CONCLUSION

We demonstrated longitudinal stability of pRNFL and GCIPL IEDs as markers for optic nerve lesions in PwMS, thus reinforcing the role for OCT in demonstrating optic nerve lesions.

Diagnosis of multiple sclerosis using optical coherence tomography supported by artificial intelligence

BACKGROUND

Current procedures for diagnosing multiple sclerosis (MS) present a series of limitations, making it critically important to identify new biomarkers. The aim of the study was to identify new biomarkers for the early diagnosis of MS using spectral-domain optical coherence tomography (OCT) and artificial intelligence.

METHODS

Spectral domain OCT was performed on 79 patients with relapsing-remitting multiple sclerosis (RRMS) (disease duration ≤ 2 years, no history of optic neuritis) and on 69 age-matched healthy controls using the posterior pole protocol that incorporates the anatomic Positioning System. Median retinal thickness values in both eyes and inter-eye difference in healthy controls and patients were evaluated by area under the receiver operating characteristic (AUROC) curve analysis in the foveal, parafoveal and perifoveal areas and in the overall area spanned by the three rings. The structures with the greatest discriminant capacity - retinal thickness and inter-eye difference - were used as inputs to a convolutional neural network to assess the diagnostic capability.

RESULTS

Analysis of retinal thickness and inter-eye difference in RRMS patients revealed that greatest alteration occurred in the ganglion cell (GCL), inner plexiform (IPL), and inner retinal (IRL) layers. By using the average thickness of the GCL (AUROC = 0.82) and the inter-eye difference in the IPL (AUROC = 0.71) as inputs to a two-layer convolutional neural network, automatic diagnosis attained accuracy = 0.87, sensitivity = 0.82, and specificity = 0.92.

CONCLUSION

This study adds weight to the argument that neuroretinal structure analysis could be incorporated into the diagnostic criteria for MS.