Automatic Measurement of Wet AMD's Imaging Biomarkers
Principal Investigator
Co-Principal Investigator
-
Duke University
Project Goals
We are developing an open source fully-automated software program with demonstrated high accuracy that is able to detect, segment, and analyze Neovascular AMD (NVAMD) pathology seen on optical coherence tomography and compare these data to corresponding features on other imaging modalities. We anticipate that the software tools developed in this proposal will be readily adopted by clinicians, clinical study sites, and image Reading Centers to better identify NVAMD at the earliest stages, to quantify disease progression, and to measure response to therapy.
Project Summary
We are developing a fully-automated software program with demonstrated high accuracy that is able to detect, segment, and analyze neovascular AMD (NVAMD) pathology seen on spectral domain optical coherence tomography (SDOCT) and compare these data to corresponding features on other imaging modalities.
Progress Updates
So far, we have achieved significant progress on all aims of this project, often exceeding our proposed project timeline. Moreover, our experience in this project helped us to achieve exciting progress in related dry-AMD projects. We have submitted two articles on automatic segmentation of normal and AMD eyes to the most prestigious journals of our field (one already published and one under review). Moreover, one of our abstracts received the National Eye Institute's travel award at the Association for Research in Vision and Ophthalmology (ARVO) Annual Meeting, May 2011. The automated segmentation technology that we have developed for AMD eyes, has also resulted in a serendipitous discovery of a novel technique for automatic segmentation of corneal images, resulting in yet another submission of a journal article.
Overall, in part based on our work in this project, we have submitted 5 journal papers. Based on the year one results, we are confident to attain all our major goals based on the proposed timeline. We anticipate that the software tools developed in this proposal will be readily adopted by clinicians, clinical study sites, and image Reading Centers to better identify NVAMD at the earliest stages, to quantify disease progression, and to measure response to therapy.
Publications
S. J. Chiu, C. A. Toth, C. Bowes Rickman, J.A. Izatt, S. Farsiu. "Automatic Segmentation of Closed Contour Features in Ophthalmic Images using Graph Theory and Dynamic Programming", Biomedical Optics Express, Vol. 3, Issue 5, pp. 1127-1140. May 2012
http://www.opticsinfobase.org/boe/abstract.cfm?uri=boe-3-5-927
L. Fang, S. Li, Q. Nie, J.A. Izatt, C.A. Toth, and S. Farsiu, "Sparsity Based Denoising of Spectral Domain Optical Coherence Tomography Images", Biomedical Optics Express, 3(5), pp. 927-942, May, 2012
http://www.opticsinfobase.org/boe/abstract.cfm?uri=boe-3-5-1127
P. Lin, P.S. Mettu, D.L. Pomerleau, S.J. Chiu, R. Maldonado, S. Stinnett, C. Toth, S. Farsiu, P.Mruthyunjaya, Image inversion on spectral-domain optical coherence tomography optimizes choroidal thickness measurement and visualization of outer choroidal detail through improved outer choroidal contrast,Investigative Ophthalmology & Visual Science, 53(4), April, pp.1874-1882 2012.
First published on: April 01, 2010
Last modified on: November 21, 2024