Elsevier

Ophthalmology

Volume 119, Issue 11, November 2012, Pages 2231-2238
Ophthalmology

Original article
Use of a Support Vector Machine for Keratoconus and Subclinical Keratoconus Detection by Topographic and Tomographic Data

https://doi.org/10.1016/j.ophtha.2012.06.005Get rights and content

Purpose

To define a new classification method for the diagnosis of keratoconus based on corneal measurements provided by a Scheimpflug camera combined with Placido corneal topography (Sirius, CSO, Florence, Italy).

Design

Retrospective case series.

Participants

We analyzed the examinations of 877 eyes with keratoconus, 426 eyes with subclinical keratoconus, 940 eyes with a history of corneal surgery (defined as abnormal), and 1259 healthy control eyes.

Methods

For each group, eyes were divided into a training and a validation set. A support vector machine (SVM) was used to analyze the corneal measurements and classify the eyes into the 4 groups of participants. The classifier was trained to consider the indices obtained from both the anterior and posterior corneal surfaces or only from the anterior corneal surface.

Main Outcome Measures

Symmetry index of front and back corneal curvature, best fit radius of the front corneal surface, Baiocchi Calossi Versaci front index (BCVf) and BCV back index (BCVb), root mean square of front and back corneal surface higher order aberrations, and thinnest corneal point were analyzed. The diagnostic performance of the classifier was evaluated.

Results

The accuracy of the classifier was excellent both with and without the data generated from the posterior corneal surface and corneal thickness because the number of true predictions was greater than 95% and 93%, respectively, in all classes. Precision improved most when posterior corneal surface data were included, especially in cases of subclinical keratoconus. Using the data from both anterior and posterior corneal surfaces and pachymetry allowed the SVM to increase its sensitivity from 89.3% to 96.0% in abnormal eyes, 92.8% to 95.0% in eyes with keratoconus, 75.2% to 92.0% in eyes with subclinical keratoconus, and 93.1% to 97.2% in normal eyes.

Conclusions

The classification algorithm showed high accuracy, precision, sensitivity, and specificity in discriminating among abnormal eyes, eyes with keratoconus or subclinical keratoconus, and normal eyes. Including the posterior corneal surface and thickness parameters markedly improved the sensitivity in the diagnosis of subclinical keratoconus. Classification may be particularly useful in excluding eyes with early signs of corneal ectasia when screening patients for excimer laser surgery.

Financial Disclosure(s)

Proprietary or commercial disclosure may be found after the references.

Section snippets

Materials and Methods

This was a retrospective case series study. Clinical data and corneal examinations were retrieved from clinical records at the Muscat Eye Laser Center (Muscat, Oman) and Studio Oculistico d'Azeglio (Bologna, Italy). The study was conducted in accordance with the ethical standards stated in the 1964 Declaration of Helsinki and approved by the local clinical research ethics committee with informed consent obtained.

Results

Overall, 3502 eyes were enrolled. According to the clinical diagnosis, they were classified as follows:

  • 1

    Keratoconus: 877 eyes of 451 patients (mean age, 34.8±12.6 years; range, 15–71 years).

  • 2

    Subclinical keratoconus: 426 eyes of 340 patients (mean age, 40.4±17.1 years; range, 15–65 years). This group included 229 eyes with early keratoconus and 197 eyes with suspect keratoconus.

  • 3

    Abnormal: 940 eyes of 486 patients (mean age, 43.6±13.6 years; range, 14–78 years).

  • 4

    Normal: 1259 eyes of 756 normal

Discussion

Preoperative screening of patients undergoing corneal refractive surgery requires correct identification of eyes with subclinical keratoconus because subjects with this condition are known to be at increased risk of developing iatrogenic ectasia.3, 4, 5, 6, 7 When no signs of keratoconus are detected in either eye, this task represents a challenge for the ophthalmologist, given the lack—by definition—of any clinical difference between normal eyes and eyes with subclinical keratoconus. The only

References (42)

  • D.P. Piñero et al.

    Corneal volume, pachymetry, and correlation of anterior and posterior corneal shape in subclinical and different stages of clinical keratoconus

    J Cataract Refract Surg

    (2010)
  • M. Milla et al.

    Pachymetric measurements with a new Scheimpflug photography–based system: intraobserver repeatability and agreement with optical coherence tomography pachymetry

    J Cataract Refract Surg

    (2011)
  • G. Savini et al.

    Repeatability of automatic measurements by a new Scheimpflug camera combined to Placido topography

    J Cataract Refract Surg

    (2011)
  • X. Li et al.

    Keratoconus: classification scheme based on videokeratography and clinical signs

    J Cataract Refract Surg

    (2009)
  • S.N. Rao et al.

    Role of Orbscan II in screening keratoconus suspects before refractive corneal surgery

    Ophthalmology

    (2002)
  • M.W. Belin et al.

    Keratoconus: it is hard to define, but …

    Am J Ophthalmol

    (2007)
  • J.H. Krachmer et al.

    Keratoconus and related noninflammatory corneal thinning disorders

    Surv Ophthalmol

    (1994)
  • R. Ambrosio et al.

    Complications of laser in situ keratomileusis: etiology, prevention, and treatment

    J Refract Surg

    (2001)
  • J.B. Randleman et al.

    Validation of the Ectasia Risk Score System for preoperative laser in situ keratomileusis screening

    Am J Ophthalmol

    (2008)
  • J.B. Randleman et al.

    Risk assessment for ectasia after corneal refractive surgery

    Ophthalmology

    (2008)
  • N. Maeda et al.

    Automated keratoconus screening with corneal topography analysis

    Invest Ophthalmol Vis Sci

    (1994)
  • Cited by (182)

    • Artificial intelligence for detecting keratoconus

      2023, Cochrane Database of Systematic Reviews
    View all citing articles on Scopus

    Manuscript no. 2012-78.

    Financial Disclosure(s): The author(s) have made the following disclosure(s):

    Francesco Versaci and Gabriele Vestri are employees of CSO srl.

    View full text