At the Vision and Image Processing (VIP) Laboratory at Duke University, our long-term goal is to improve the overall health and vision outcomes of at-risk patients with ocular and neurological diseases through earlier and better-directed therapy. To achieve this goal, we take advantage of recent advances in artificial intelligence (AI) and optics as an integrated technology to capture the highest quality ocular images. We also develop AI-enabled automatic image processing tools to quantitatively measure novel imaging biomarkers of the onset and progression of ophthalmic and neurological diseases.

VIP Lab’s Research Focus

1.  Image Analysis Software Development for Ophthalmology and Vision Sciences:

We collaborate with our clinical colleagues, including Duke Advanced Reasearch in SDOCT Imaging Laboratory. A major focus of our lab is the development of fully automated software to objectively detect and evaluate the biomarkers for onset and progression of ocular diseases in adults (e.g. diabetic retinopathy, age-related macular degeneration (AMD), or glaucoma) and children (e.g. retinopathy or prematurity (ROP)). We also develop automatic segmentation algorithms to detect/segment/quantify ocular anatomical/pathological structures seen on ophthalmic imaging systems such as Optical Coherence Tomography (OCT).

2.  Image Processing Theory and Application:

We study efficient signal processing based methods to overcome the theoretical and practical limitations that constrain the achievable resolution of any imaging device. Our approach, which is based on adaptive extraction and robust fusion of relevant information from the expensive and sophisticated as well as simple and cheap sensors, has found wide applications in improving the quality of imaging systems such as ophthalmic SD-OCT, video indirect ophthalmoscopy, digital X-ray mammography, electronic and optical microscopes, and commercial digital camcorders. When not developing a mathematical model of the procrastination theory, we take part in some statistical signal processing ideas, mainly super-resolution, demosaicing/deblurring/denoising, motion estimation, compressive sensing/adaptive sampling, and sensor fusion.

3.  Advanced Ophthalmic Imaging Hardware Development:

In collaboration with our colleagues at the department of biomedical engineering, especially the Laboratory for Biophotonics, we develop the next generation ophthalmic imaging systems, including advanced handheld SDOCT and adaptive optics ocular imaging systems