KeratoDetect: Keratoconus Detection Algorithm Using Convolutional Neural Networks

Abstract

Keratoconus (KTC) is a noninflammatory disorder characterized by progressive thinning, corneal deformation, and scarring of the cornea. The pathological mechanisms of this condition have been investigated for a long time. In recent years, this disease has come to the attention of many research centers because the number of people diagnosed with keratoconus is on the rise. In this context, solutions that facilitate both the diagnostic and treatment options are quickly needed. The main contribution of this paper is the implementation of an algorithm that is able to determine whether an eye is affected or not by keratoconus. The KeratoDetect algorithm analyzes the corneal topography of the eye using a convolutional neural network (CNN) that is able to extract and learn the features of a keratoconus eye. The results show that the KeratoDetect algorithm ensures a high level of performance, obtaining an accuracy of 99.33% on the data test set. KeratoDetect can assist the ophthalmologist in rapid screening of its patients, thus reducing diagnostic errors and facilitating treatment.

Figure 1

The keratoconus (a) and normal (b) cornea.

Figure 2

Structure of a CNN.

Figure 3

Cornea topographies. (a) Keratoconus eye. (b) Healthy eye.

Figure 4

CNN-proposed algorithm.

Figure 5

Algorithm input.

Figure 6

CNN-designed algorithm.

Figure 7

Algorithm accuracy parameter.

Figure 8

Accuracy parameter for 630 learning iterations.

Figure 9

Losses for 630 learning iterations.

Figure 10

Accuracy parameter for 798 learning iterations.

Figure 11

Losses for 798 learning iterations.

Figure 12

Initial convolution layer weights (before training) (a) and the first convolutional layer weights (after training) (b).