TY - GEN

T1 - A fast regularized least squares method for retinal vascular oxygen tension estimation

AU - Gunay, Gokhan

AU - Yildirim, Isa

PY - 2013

Y1 - 2013

N2 - Regularization of least squares estimation method was proposed earlier in order to improve estimation of oxygen tension in retinal vessels and it was shown to be very effective. Optimum points for cost function of the regularized least squares (RLS) estimation were found using iterative methods which do not require considerable computations for small oxygen tension (pO2) maps. However, as pO2 maps get larger, iterative solutions become computationally expensive. In this study, to find optimum points of the RLS cost functions of larger pO2 images, we propose an accelerated solution derived from closed form solution of the RLS cost function.

AB - Regularization of least squares estimation method was proposed earlier in order to improve estimation of oxygen tension in retinal vessels and it was shown to be very effective. Optimum points for cost function of the regularized least squares (RLS) estimation were found using iterative methods which do not require considerable computations for small oxygen tension (pO2) maps. However, as pO2 maps get larger, iterative solutions become computationally expensive. In this study, to find optimum points of the RLS cost functions of larger pO2 images, we propose an accelerated solution derived from closed form solution of the RLS cost function.

KW - Accelerated closed form solution

KW - Regularized least squares estimation

KW - Retinal oxygen tension estimation

UR - http://www.scopus.com/inward/record.url?scp=84887939677&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84887939677

SN - 9789881925299

T3 - Lecture Notes in Engineering and Computer Science

SP - 2244

EP - 2248

BT - Proceedings of the World Congress on Engineering 2013, WCE 2013

T2 - 2013 World Congress on Engineering, WCE 2013

Y2 - 3 July 2013 through 5 July 2013

ER -