June 2015
Volume 56, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2015
A new Zernike algorithm to link asymmetric corneal thickness to corneal wavefront aberrations for diagnosis of keratoconus
Author Affiliations & Notes
  • Purnima R Srivatsa
    Narayana Nethralaya, Bangalore, India, Bangalore, India
  • Rohit Shetty
    Narayana Nethralaya, Bangalore, India, Bangalore, India
    Singapore Eye Research Institute, Singapore, Singapore
  • Himanshu Matalia
    Narayana Nethralaya, Bangalore, India, Bangalore, India
  • Abhijit Sinha Roy
    Narayana Nethralaya, Bangalore, India, Bangalore, India
  • Footnotes
    Commercial Relationships Purnima R Srivatsa, None; Rohit Shetty, None; Himanshu Matalia, None; Abhijit Sinha Roy, Avedro (C), Carl Zeiss (C), Cleveland Clinic Cole Eye Institute (P), Topcon (C)
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2015, Vol.56, 1618. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Purnima R Srivatsa, Rohit Shetty, Himanshu Matalia, Abhijit Sinha Roy; A new Zernike algorithm to link asymmetric corneal thickness to corneal wavefront aberrations for diagnosis of keratoconus. Invest. Ophthalmol. Vis. Sci. 2015;56(7 ):1618.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract
 
Purpose
 

To correlate Zernike interpolation of spatial corneal thickness with corneal wavefront aberrations and assess its diagnostic sensitivity and specificity

 
Methods
 

Corneal tomography of 44 normal and 92 keratoconus (KC) corneas was assessed with Pentacam (OCULUS Optikgerate Gmbh, Germany). Using mean keratometry, a custom severity scale was used to grade KC corneas from 1 to 3. Zernike analysis of anterior surface aberrations was performed using ray tracing. A novel Zernike formulation was used to map the spatially varying thickness of the corneas. Both aberrations and thickness mapping with Zernike yielded familiar coefficients, e.g., defocus, coma. The 2nd, 3rd and higher order root mean square (RMS) of the Zernike coefficients from both aberrations and thickness mapping were analyzed with stepwise logistic regression. Also, cone location magnitude index of the anterior and posterior surface using both axial (aCLMI) and tangential curvature (tCLMI) were analyzed with logistic regression. Area under the receiver operating characteristics curve (AUROC) of each regression was compared along with their corresponding sensitivity (Se) and specificity (Sp). A p-value<0.05 was considered to be statistically significant.

 
Results
 

Among aberrations variables, 2nd and 3rd order Zernike RMS of aberrations were the best indicators of KC (p<0.0001, AUROC=0.998, Se=98.8%, Sp=97.7%). Among thickness variables, minimum corneal thickness, 2nd and 3rd order Zernike RMS of thickness map were the best indicators of KC (p<0.0001, AUROC=0.978, Se=94.2%, Sp=95.4%). Among CLMI variables, aCLMI of posterior and tCLMI of anterior surface were the best indicators of KC (p<0.0001, AUROC=0.965, Se=98.8%, Sp=97.7%). Comparison of the ROC curves of the three regression models did not yield any statistically significant difference between them (CLMI vs. corneal Zernike aberrations: p=0.77; CLMI vs. thickness Zernike: p =0.06; Corneal Zernike aberrations vs. thickness Zernike: p=0.06). ROC curves are shown in figure. There was significant linear correlation between RMS of corneal aberration and thickness Zernike coefficients (minimum r = 0.69, p<0.0001).

 
Conclusions
 

A new Zernike method of mapping corneal thickness in KC was developed. The Zernike mapping of thickness had diagnostic efficacy similar to using Zernike mapping of corneal wavefront aberrations and CLMI indices.  

 
ROC curves of study indices
 
ROC curves of study indices

 
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×