A comparative study of transformation models for the sequential mosaicing of long retinal sequences of slit-lamp images obtained in a closed-loop motion

PurposeNavigated panretinal photocoagulation is a standard care for proliferative diabetic retinopathy. Slit-lamp-based systems used for this treatment provide a narrow view of the retina. Retinal mosaics are used for view expansion and treatment planning. Mosaicing slit-lamp images is a hard task due to the absence of a physical model of the imaging process, large textureless regions and imaging artifacts, mostly reflections.MethodsWe present a comparative study of various geometric transformation models applied to retinal image mosaicing in computer-assisted slit-lamp imaging. We propose an efficient point correspondence-based framework for transformation model evaluation in a typical closed-loop motion scenario. We compare the performance of multiple linear and nonlinear models of different complexities and assess the effect of the number of points used for parameter estimation. We use a local fitting error (LFE) metric to estimate the models’ performance in pairwise registration. Because LFE alone is not conclusive regarding the problem of accumulated drift, we propose a loop closure error (LCE) metric to quantify the effect of accumulated local registration errors. We also provide a new normalization procedure for the quadratic transformation model, widely used in retinal image registration.ResultsIn total, seven transformation models were evaluated on three datasets of long image sequences. LFE decreases with increasing complexity of the model, while LCE, in contrast, shows superior performance of simple models. Varying the number of point correspondences did not reveal a common trend for the LCE metric, showing an increase in the error for simple models and an unstable behavior of the complex models.ConclusionOur results show that simple models are less sensitive to drift and preferable for sequential mosaicing in slit-lamp imaging, while more complex models are the best choice for short-term registration.

[1]  P. Anandan,et al.  About Direct Methods , 1999, Workshop on Vision Algorithms.

[2]  Charles V. Stewart,et al.  Covariance-driven mosaic formation from sparsely-overlapping image sets with application to retinal image mosaicing , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[3]  Andrew Zisserman,et al.  Feature Based Methods for Structure and Motion Estimation , 1999, Workshop on Vision Algorithms.

[4]  J W Berger,et al.  Mosaicking and enhancement of slit lamp biomicroscopic fundus images , 2001, The British journal of ophthalmology.

[5]  Philip F. McLauchlan,et al.  Image mosaicing using sequential bundle adjustment , 2002, Image Vis. Comput..

[6]  Koen A. Vermeer,et al.  A Hierarchical Coarse-to-Fine Approach for Fundus Image Registration , 2014, WBIR.

[7]  Charles L. Lawson,et al.  Solving least squares problems , 1976, Classics in applied mathematics.

[8]  Sangyeol Lee,et al.  Validation of Retinal Image Registration Algorithms by a Projective Imaging Distortion Model , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[9]  Charles V. Stewart,et al.  A Feature-Based Technique for Joint Linear Estimation of High-Order Image-to-Mosaic Transformations: Mosaicing the Curved Human Retina , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Matthijs C. Dorst.  Distinctive Image Features from Scale-Invariant Keypoints , 2011 .

[11]  Yulong Shen,et al.  Registration and fusion of retinal images-an evaluation study , 2003, IEEE Transactions on Medical Imaging.

[12]  Marcus Kernt,et al.  Focal and panretinal photocoagulation with a navigated laser (NAVILAS®) , 2011, Acta ophthalmologica.

[13]  R. Klein,et al.  Retinopathy in a population-based study. , 1992, Transactions of the American Ophthalmological Society.

[14]  Chia-Ling Tsai,et al.  The dual-bootstrap iterative closest point algorithm with application to retinal image registration , 2003, IEEE Transactions on Medical Imaging.

[15]  Russell H. Taylor,et al.  Fundus Image Mosaicking for Information Augmentation in Computer-Assisted Slit-Lamp Imaging , 2014, IEEE Transactions on Medical Imaging.

[16]  Jay Chhablani,et al.  Comparison of conventional pattern and novel navigated panretinal photocoagulation in proliferative diabetic retinopathy. , 2014, Investigative ophthalmology & visual science.

[17]  Martina Broehan,et al.  Real-Time Multimodal Retinal Image Registration for a Computer-Assisted Laser Photocoagulation System , 2011, IEEE Transactions on Biomedical Engineering.

[18]  James C. Gee,et al.  Landmark matching based retinal image alignment by enforcing sparsity in correspondence matrix , 2014, Medical Image Anal..

[19]  Eros Comunello,et al.  Robust Visual Tracking for Retinal Mapping in Computer-Assisted Slit-Lamp Imaging , 2014, 2014 IEEE 27th International Symposium on Computer-Based Medical Systems.

[20]  Marcus Kernt,et al.  Pain and accuracy of focal laser treatment for diabetic macular edema using a retinal navigated laser (Navilas®) , 2012, Clinical ophthalmology.

[21]  Adrien Bartoli,et al.  Maximizing the Predictivity of Smooth Deformable Image Warps through Cross-Validation , 2008, Journal of Mathematical Imaging and Vision.