Dimitri Azar

Speaker

Dimitri Azar, MD, MBA
Professor
College of Medicine, The University of Illinois
Senior Director of Ophthalmic Innovations, and
Clinical Lead, Ophthalmology Programs, Alphabet Verily Life Sciences/Google

Date and Time

Monday, October 7, 2019
11:10 am - 12:30 pm

Location

489 Minor Hall
Berkeley, CA

Dimitri Azar's Abstract

Artificial Intelligence in Diagnosis of Ophthalmic Diseases

In this talk we will describe how Artificial Intelligence (AI) can be helpful in the diagnosis and screening of various Ophthalmic conditions. AI-based grading algorithm may be used to identify which patients should be referred to an eye care provider for evaluation and treatment. AI can improve work efficiency in settings of limited healthcare resources and can generate novel disease-specific patterns and novel features to gain innovative scientific insights. We will describe various AI learning algorithms, including unsupervised learning, supervised learning (including deep learning convolutional neural networks) and semi-supervised learning. These are helpful to interpret ophthalmological images. Convolutional neural networks start off any fine tuning and progressively learn the combinations and permutations of important features. AI applications will be described for the following conditions: refractive error prediction, corneal topography, dry eye diagnosis, diabetic retinopathy, and AMD.