The detection of glaucoma using an artificial neural network

The scanning laser ophthalmoscope is a device used by ophthalmologists to obtain topographic images of patients' optic nerve heads (ONHs). Measurements are taken from these images that quantitatively describe the shape of the ONH. Glaucoma involves the loss of retinal nerve fibers, which in turn produces a change in the ONH shape. However, it is not known which shape parameters are most relevant to the diagnosis of glaucoma. To solve this problem, a feedforward artificial neural network (ANN) was designed to discriminate between patient data. Patients were first independently classified using perimetry data (visual fields) into normal and abnormal (glaucomatous) groups. The ANN was trained using error backpropagation (n=89 samples) and the classification model was cross validated using one normal, and one abnormal sample. The entire data set (45 normals and 46 abnormals) was utilized for cross validation and each time the error rate of the training set was required to be less than 15%. The ANN gave an overall classification rate of 86.7%, with a specificity (correct normals) of 88.9% and a sensitivity (correct abnormals) of 84.4%. The ANN classification model, with only two hidden units, generalized well which indicates that the ONH measurements are useful for the detection of glaucoma.