Mosaicking retinal images
Brief Description of Project:
OCT angiography (OCT-A) methods allow us to image the retinal vascular perfusion. There is a fundamental trade-off between vascular resolution and imaging field of view in OCT-A such that microvasculature are only visible when imaging small volumes. This is particularly detrimental for retinopathy of prematurity (ROP) screening because microvascular changes occur in early-stage disease before the onset of retinal hemorrhage and scar tissue formation. However, successful disease screening requires surveillance of the entire retina out to the periphery over multiple longitudinal follow-up exams. Robust volumetric co-registration can overcome these trade-offs between resolution and field-of-view by enabling multi-volumetric mosaicking. Several small field-of-view, high-resolution OCT-A volumes can be acquired over the entire retina and combined in post-processing into a single three-dimensional dataset for quantitative analysis. The goal of this project is to explore registration methods for enabling robust mosaicking. I expect this project to lead to a submission to the SPIE Medical Imaging Conference by the end of the summer.
Matlab, preferably C++ programming
Nature of Supervision:
I will meet with the student weekly, and additionally, we have a joint lab meeting with the Tao lab (BME) who work on the image acquisition side. I have additional meetings as needed (e.g., near conference deadlines, etc). Both me and the other senior lab members (grad students/postdocs) are available for more frequent help as needed.
A Brief Research Plan (period is for 10 weeks):
1 week - project overview and plan development
6 weeks - project implementation
1 week - statistical evaluation
2 week - project write-up
Number of Open Slots: 2
Name: Ipek Oguz
Department: Electrical Engineering and Computer Science