Elsevier

Ophthalmology

Volume 119, Issue 8, August 2012, Pages 1705-1711
Ophthalmology

Original article
Diagnostic Ability of a Linear Discriminant Function for Spectral-Domain Optical Coherence Tomography in Patients with Multiple Sclerosis

https://doi.org/10.1016/j.ophtha.2012.01.046Get rights and content

Purpose

To calculate and validate a linear discriminant function (LDF) for spectral-domain optical coherence tomography (OCT) to improve the diagnostic ability of retinal nerve fiber layer (RNFL) thickness parameters for the detection of multiple sclerosis (MS).

Design

Observational cross-sectional study.

Participants

Patients with multiple sclerosis (n = 115) and age-matched healthy subjects (n = 115) were enrolled in the study.

Methods

The Spectralis OCT system (Heidelberg Engineering, Heidelberg, Germany) was used to obtain the circumpapillary RNFL thickness in both eyes of each participant.

Main Outcome Measures

A validating set including 60% of the study subjects (69 healthy individuals and 69 patients with MS) was used to test the performance of the LDF in an independent population. Receiver operating characteristic (ROC) curves were plotted and compared with the RNFL parameters measured using OCT. Sensitivity and specificity were used to evaluate diagnostic performance.

Results

The optimized function was 4.965 − 0.40 × (mean thickness 15–30 degrees) − 0.17 × (mean thickness 300–315 degrees) + 2.743 − 0.032 × (mean thickness 105–120 degrees) − 0.031 × (mean thickness 120–135 degrees) − 0.018 × (mean thickness 225–240 degrees). The largest area under the ROC curve was 0.834 for our LDF in the validating population. At 95% fixed specificity, the LDF yielded the highest sensitivity values.

Conclusions

Measurements of RNFL thickness obtained with Spectralis OCT had good ability to differentiate between healthy individuals and individuals with MS. On the basis of the area under the ROC curve, the LDF performed better than any single parameter.

Financial Disclosure(s)

The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Section snippets

Materials and Methods

The design of the study followed the Declaration of Helsinki Principles. The study protocol was approved by the Clinical Research Ethics Committee of Aragon (Zaragoza, Spain), and informed written consent was obtained from all participants.

Results

A total of 115 eyes from 115 patients with relapsing-remitting MS were examined. A previous acute optic neuritis attack was reported for 35 eyes (30%), whereas 80 eyes (70%) were studied from patients with no history of optic neuritis. The duration of the MS ranged from 6 months to 40 years with a median of 9.3 years since diagnosis. The ages of patients ranged from 20 to 62 years with a mean of 41.2 years (Table 1). The ratio of women to men was 2:1 (76 female, 39 male). Mean intraocular

Discussion

Previous studies have reported the sensitivity and specificity of OCT for discriminating between healthy and MS eyes.18, 19 In addition, some studies have attempted to increase the diagnostic ability of OCT for some pathologies using learning classifiers,9, 20, 21 yet the sensitivity and specificity of OCT to diagnose MS have not improved. We were unable to find published studies aimed at calculating an LDF solely on the basis of RNFL parameters measured with OCT in patients with MS. The

Conclusions

Although we were unable to find other combinations of RNFL parameters with better diagnostic ability, other statistical analyses could provide alternative formulas that would increase the diagnostic performance of the Spectralis OCT. The AUC of the proposed LDF was not significantly different from the temporal and the temporal inferior sector thicknesses, but the sensitivity values were higher for the LDF at a high fixed specificity. Also, the lowest negative LR was found for the LDF, and these

References (25)

  • E. Garcia-Martin et al.

    Risk factors for progressive axonal degeneration of the retinal nerve fibre layer in multiple sclerosis patients

    Br J Ophthalmol

    (2011)
  • E. Garcia-Martin et al.

    Effect of optic neuritis on progressive axonal damage in multiple sclerosis patients

    Mult Scler

    (2011)
  • Cited by (34)

    • Shannon entropy-based complexity quantification of nonlinear stochastic process

      2022, Multi-Chaos, Fractal and Multi-Fractional Artificial Intelligence of Different Complex Systems
    • Past, present and future role of retinal imaging in neurodegenerative disease

      2021, Progress in Retinal and Eye Research
      Citation Excerpt :

      Second, there are numerous manufacturers of OCT systems. It is well known that scan patterns, segmentation algorithms and outcome measures vary significantly among manufacturers (Garcia-Martin et al., 2012a,b; Garcia-Martin et al., 2014a,b; Satue et al., 2014; Chan et al., 2019). Therefore, when comparing across studies, it is critical to keep in mind the details of the OCT system and scan parameters used.

    • Machine learning in diagnosis and disability prediction of multiple sclerosis using optical coherence tomography

      2021, Computers in Biology and Medicine
      Citation Excerpt :

      Thinning of these layers, which are not covered by myelin, constitutes a direct readout of axonal damage [7,13,14]. Studies have shown that RNFL thickness, measured by OCT, is a useful parameter to distinguish MS patients from healthy controls [4,15]. Garcia-Martin et al. [4,16] used an artificial neural network (ANN) in combination with OCT data to diagnose MS. Pérez del Palomar et al. [17] also demonstrated that machine learning techniques, specially decision tree (DT), combined with OCT data could be used to diagnose this disease.

    • Optical coherence tomography findings in Parkinson's disease

      2018, Kaohsiung Journal of Medical Sciences
      Citation Excerpt :

      Optical coherence tomography is a cost-effective, rapid, and non-invasive optical interferometric method generating cross-sectional images of the retina. It is a reliable tool which enables quantitative assessment of RNFLT for diagnose and follow-up of neurological and neurodegenerative conditions including migraine, optic neuritis, Alzheimer's disease, and multiple sclerosis [20–24]. Although IPD is a common disease among aging population, accuracy of the clinical diagnosis of IPD is still limited.

    • Retinal layer segmentation in patients with multiple sclerosis using spectral domain optical coherence tomography

      2014, Ophthalmology
      Citation Excerpt :

      The new prototype technique for automated retinal segmentation of the Spectralis OCT device was performed to identify each retinal layer and quantify its thickness. An exhaustive literature search using PubMed revealed that this automatic segmentation technique has not been used in previous studies with patients with MS. Images were acquired using image alignment eye-tracking software (TruTrack; Heidelberg Engineering Inc.) to obtain perifoveal volumetric retinal scans comprising 25 single horizontal axial scans (scanning area: 666 mm2, centered on the fovea).9,10 Segmentation of the retinal layers in a single horizontal foveal scan was performed automatically by the new segmentation application (Segmentation Technology; Heidelberg Engineering, Inc.) to separate the retina automatically into 10 layers: 1 = inner limiting membrane; 2 = nerve fiber layer; 3 = ganglion cell layer; 4 = inner plexiform layer; 5 = inner nuclear layer; 6 = outer plexiform layer; 7 = outer nuclear layer; 8 = outer limiting membrane; 9 = photoreceptors (rods and cones); 10 = retinal pigment epithelium (Fig 1).

    View all citing articles on Scopus

    Manuscript no. 2011-1704.

    Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article.

    View full text