An Automated Method to Characterise Keratoconic Corneas
Authors: Lopes, B., Padmanabhan, P., Zhang, H., Abass, A., Eliasy, A., Bandeira, F., Bao, F., Bühren, J., Elmassry, A., Faria-Correia, F., Rocha, K.
Journal: Journal of Refractive Surgery
Publication Date: Jun 2021
Cone geometry estimations by experts using tangential curvature and relative elevation maps along with results of the automated method. (A to F) Anterior surface of cases 1 to 6. (G to L) Posterior surface of cases 1 to 6. Blue line: cone boundaries estimated by the automated method. Red lines: experts’ cone boundaries estimations on relative elevation maps. Green lines: experts’ cone boundaries estimations on tangential curvature maps. (X and Y axes units: mm)
Summary:
Keratoconus is an eye condition that affects the cornea, causing it to develop a cone-like shape. This distortion can lead to vision problems and difficulties in diagnosing and treating the condition. Currently, eye specialists rely on different maps to identify the cone's centre and boundaries in keratoconic corneas, but these methods can be subjective and highly variable. Our research aimed to develop a more reliable and accurate automated method for detecting the cone shape characteristics in keratoconic corneas.
We compared the results of our automated method with the estimations made by 12 experienced cornea specialists using different types of maps. We found that there was low agreement between the cone centre estimations using different types of maps for 10 of the 12 cases, showing high variability among clinicians' estimations. However, our automated method did not show differences in 11 of the 12 cases when compared to the two map estimations, indicating that it provided a reliable evaluation of the keratoconus shape independent of maps or colour scale.
The development of this automated method is a significant step forward in understanding keratoconus and improving the diagnosis and management of the condition. By providing a more accurate and objective way of identifying the cone's centre and boundaries, this method has the potential to enhance our knowledge of the disease and optimise treatment strategies. Future studies will explore the impact of cone geometry on refractive errors, higher-order aberrations, and image quality, as well as the response of cone characteristics to various corneal procedures.