Deep learning architecture “LightOCT” for diagnostic decision support usingoptical coherence tomography images of biological samples
Optical coherence tomography (OCT) is being increasingly adopted as a non-invasive technique for cancer and ocular diseasediagnosis. Diagnostic information is manifest in textural and geometric features of the OCT images, which are used by human expertise to interpret and triage. Here, a custom deep learning network, LightOCT, isproposed for the classification of OCT images. LightOCTis a convolutional neural network with only two convolutional layers and a fully connectedlayer, but it is shown to provide excellent training and test results for diverse OCT imagedatasets.
Biomedical Optics Express: https://www.osapublishing.org/boe/abstract.cfm?uri=boe-11-9-5017
Ankit Butola, P. Senthilkumaran and Dalip Singh Mehta