Amplitude-scan classification using artificial neural networks

Optical coherence tomography (OCT) images semi-transparent tissues noninvasively. Relying on backscatter and interferometry to calculate spatial relationships, OCT shares similarities with other pulse-echo modalities. There is considerable interest in using machine learning techniques for automated image classifcation, particularly among ophthalmologists who rely heavily on diagnostic OCT.Artifcial neural networks (ANN) consist of interconnected nodes and can be employed as classifers after training on large datasets. Conventionally, OCT scans are rendered as 2D or 3D humanreadable images of which the smallest depth-resolved unit is the amplitude-scan refectivity-function profle which is difcult for humans to interpret. We set out to determine whether amplitude-scan refectivity-function profles representing disease signatures could be distinguished and classifed by a feed-forward ANN. Our classifer achieved high accuracies after training on only 24 eyes, with evidence of good generalization on unseen data. The repertoire of our classifer can now be expanded to include rare and unseen diseases and can be extended to other disciplines and industries