Automatic segmentation of seven retinal layers in SDOCT images congruent with expert manual segmentation.

RESEARCH INTERESTS

Dr. Toth specializes in the evaluation and surgical treatment of vitreoretinal diseases in infants, children and adults, and in novel research resulting in the clinical application of optical coherence tomography (OCT) imaging in surgery and at the bedside. Her clinical interests and skills include the surgical treatment of macular diseases (such as, macular hole, epiretinal membrane and vitreomacular traction), retinal detachment, proliferative diabetic retinopathy, proliferative vitreoretinopathy (PVR), and retinopathy of prematurity (ROP). 

Dr. Toth is a world expert in retinal imaging with optical coherence tomography (OCT) and pioneered both the first use of a research hand-held spectral domain OCT system for infant examination and the first intraoperative OCT-guided ophthalmic surgical system. For infants and children, Dr. Toth's multidisciplinary team has demonstrated novel eye findings that are visible only with OCT imaging and that are often associated with brain disease or challenges of brain development. In surgery, Dr. Toth performed the world's first intraoperative OCT imaging and the first swept-source OCT imaging with heads-up display during retinal surgery. With colleagues in the Duke Eye Center and in Biomedical Engineering, she perfecting such techniques. She has been repeatedly honored among the Best Doctors in America.

Dr. Toth is also professor in the Department of Biomedical Engineering in the Pratt School of Engineering. Her primary research interests are in translational research and early-application clinical trials with a focus on novel retinal imaging with spectral domain and swept source optical coherence tomography (SD and SSOCT). Dr. Toth's Laboratory, the Duke Advanced Research in Spectral Domain/Swept Source OCT Imaging (DARSI) Laboratory centers on improving early diagnostic methods, imaging biomarkers and therapies for both age-related macular degeneration (AMD) and for retinal diseases in children. Sina Farsiu, PhD, has collaborated to provide advanced image processing for OCT with in the DARSI Laboratory. In collaboration with Joseph Izatt, PhD  in Biomedical Engineering, the DARSI team is currently applying OCT to the diagnosis and care of retinal diseases and especially in microsurgery in adults and in children in several studies including NIH funded investigations. 

Dr. Toth was also co-founder and has been the Director of Grading for OCT for the Duke Reading Center and has designed and directed OCT analysis for numerous multicenter clinical trials including the Comparisons of AMD Treatment Trials (CATT). The Duke Reading Center provides support in training, data acquisition, and grading for multicenter clinical trials utilizing optical coherence tomography as an outcome measure.

Dr. Toth chaired the multicenter Age Related Eye Disease Study 2 Ancillary SDOCT (A2ASDOCT) Study and has participated as site PI in the AREDS2. She also led studies of macular translocation surgery (MT360) for patients with severe AMD, along with co-investigator Dr. Sharon Freedman. Macular translocation surgery was a salvage treatment for AMD patients who lost vision due to neovascular AMD, prior to the current era of anti-Vascular Endothelial Growth Factor treatments. The surgery resulted in an auto-transplant of the retina, isolating the retina from the underlying choroidal and retinal pigment epithelial pathology. Imaging and retinal function data from those studies have contributed to teasing out events in the macula related to vision loss.

Learn More about the Toth-DARSI Lab

I am the director of the Vision and Image Processing (VIP) Laboratory. Along with my colleagues, we investigate how to improve early diagnostic methods and find new imaging biomarkers of ocular and neurological diseases in adults (e.g. age-related macular degeneration, diabetic retinopathy, Glaucoma, Alzheimer) and children (e.g. retinopathy or prematurity). We also develop automatic artificial intelligence machine learning and deep learning algorithms to detect/segment/quantify anatomical/pathological structures seen on medical images.

On another front, 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, digital X-ray mammography, electronic and optical microscopes, and commercial digital camcorders. We are also interested in pursuing statistical signal processing based projects, including super-resolution, demosaicing, deblurring, denoising, motion estimation, compressive sensing/adaptive sampling, and sensor fusion.