In This Video
- Nazlee Zebardast, MD, MSc, is a clinician scientist and director of Glaucoma Imaging at Mass Eye and Ear/Massachusetts General Hospital Department of Ophthalmology and an assistant professor of Ophthalmology at Harvard Medical School
- Dr. Zebardast discusses her research that uses health data science and big data to determine endophenotypes for disease in glaucoma to better understand the genetics of the disease
- With no image-based subtypes for glaucoma currently available, cutting edge machine learning methods will define the novel architecture for glaucoma that can explain the underlying disease and help with the discovery of new genetic associations
- The genetic information could be combined with patient information to determine an individual's risk score for glaucoma and help guide the clinician in developing a more precise treatment plan for each patient
Nazlee Zebardast, MD, MSc, is a clinician scientist and director of Glaucoma Imaging at Mass Eye and Ear/Massachusetts General Hospital Department of Ophthalmology and an assistant professor of Ophthalmology at Harvard Medical School. In the video, she discusses the use of big data to determine endophenotypes for disease in glaucoma. This research could lay the groundwork for individualized patient risk scores for glaucoma and more precise treatment.
Subscribe to the latest updates from Ophthalmology Advances in Motion
My research focuses on the use of health data science and big data. One of the main projects I'm working on right now—as part of my K12 Award—is understanding endophenotypes for glaucoma. Glaucoma is a disease of the optic nerve. It causes irreversible damage to the optic nerve and is one of the major causes of irreversible blindness worldwide. Statistics show that 50% of cases in the United States go unrecognized because of the relatively slow loss of vision. The problem is that glaucoma is fairly heterogeneous, and we also aren't able to determine pre-symptomatically who is going to be the person who will progress to blindness and who is going to be the person whose vision remains stable.
It turns out that glaucoma is highly genetic, and we can use the fact that it's highly heritable to determine individually who is most at risk for developing it. My current project uses images from two large biobanks to determine endophenotypes—or markers for disease—in glaucoma that we think align with disease subtype, but also with disease worsening and progression. Using these endophenotypes, we can determine genetic variants that may be associated with disease worsening and refine what we know already about glaucoma genetics. We want to then combine this information to determine for any individual, what is your background genetic risk, and what is your phenotypic risk to come up with an individualized risk score for glaucoma. From a population standpoint, I think this work will lay the foundation for individualized screening and diagnostic testing. This will allow for more effective glaucoma screening, and we hope will eventually decrease the cost and burden of vision loss.
From a clinical standpoint, if we can integrate polygenic risk scores or background genetic risk, in addition to someone's other risk factors like their age and what their phenotype looks like, then the clinician can potentially take that into account when deciding how often this patient need to be monitored and how aggressively they potentially need to be treated.
We currently have no image-based subtypes for glaucoma. These machine learning methods that we're going to use will define novel architecture for glaucoma that can explain the underlying disease, but also help us discover new genetic associations. This research is incredibly important because once we're able to identify endophenotypes for glaucoma and determine the genetic pathways, we can better understand both the genetic architecture and clinical phenotype of the disease, and ultimately move towards our end goal, which is individualized medicine.
Learn more about the Department of Ophthalmology at Mass Eye and Ear/Mass General
Refer a patient to Mass Eye and Ear/Mass General