Daniel Al Mouiee, Ingham Institute Medical Physics

Classifying Retinal Degeneration in Histological Sections using Deep Learning

Abstract:

Artificial Intelligence (AI) techniques are increasingly being used to classify retinal diseases using various imaging techniques. Such applications include the analysis of clinical retinal images like the grading of retinal fundus photography for diabetic retinopathy, assessing retinal Optical Coherence Tomography (OCT) images for Age-related Macular Degeneration (AMD) prognosis. One key area that lacks AI research is retinal histological data, which can illustrate important neuronal re-circuiting that may occur during the progression of retinal diseases. This talk will share the findings of our project in which we investigated the ability of Convolutional Neural Networks (CNNs) in categorizing histological images, of a chemically-induced feline model of retinal dystrophy, into different classes of retinal degeneration.

Bio:

I currently work at the Ingham Institute Medical Physics group as a computer scientist. My role involves developing tools to optimize various radiation therapy clinical procedures for the South West Sydney Local Health District, as well as supporting and contributing to various radiation therapy and medical physics projects that involve image processing, computer vision and deep learning. My interests lay in the application of Artificial intelligence techniques in real world domains, more specifically in Health and cryptocurrency. I graduated last year from UNSW with a Master of Biomedical Engineering and Bachelor of Software Engineering and previously worked at the Vafaee Lab (BABS-UNSW) as a research associate.