Real-Time Multimodal Retinal Image Registration for a Computer-Assisted Laser Photocoagulation System

Abstract

An algorithm for the real-time registration of a retinal video sequence captured with a scanning digital ophthalmoscope (SDO) to a retinal composite image is presented. This method is designed for a computer-assisted retinal laser photocoagulation system to compensate for retinal motion and hence enhance the accuracy, speed, and patient safety of retinal laser treatments. The procedure combines intensity and feature-based registration techniques. For the registration of an individual frame, the translational frame-to-frame motion between preceding and current frame is detected by normalized cross correlation. Next, vessel points on the current video frame are identified and an initial transformation estimate is constructed from the calculated translation vector and the quadratic registration matrix of the previous frame. The vessel points are then iteratively matched to the segmented vessel centerline of the composite image to refine the initial transformation and register the video frame to the composite image. Criteria for image quality and algorithm convergence are introduced, which assess the exclusion of single frames from the registration process and enable a loss of tracking signal if necessary. The algorithm was successfully applied to ten different video sequences recorded from patients. It revealed an average accuracy of 2.47 ± 2.0 pixels (∼23.2 ± 18.8 μm) for 2764 evaluated video frames and demonstrated that it meets the clinical requirements.

Retinal image registration using topological vascular tree segmentation and bifurcation structures

“…To verify and correct the initial correspondence, the eigen-decomposition methods exploit spatial relations between features with the aid of the coordinate proximity matrices. The technique works well on simple shape matching, but suffers form the spurious feature points in the inexact matching [32][33][34]. In the following section, we are interested in developing an efficient correspondence verification scheme.…”

Real-time retinal vessel mapping and localization for intraocular surgery

“…Key point algorithms use image feature descriptors such as SIFT [38] to find and match unique points between retinal images [39–42]. Vasculature landmark matching algorithms find distinctive points based on vessel networks, such as vein crossings or bifurcations, and match custom descriptors across multiple images [43–45]. Other approaches augment or eschew key points and use the shape of extracted vessels to match vasculature trees [46], [47].…”

“…With optimization, the algorithms of [43], [45] could be run in real time on modem hardware; however, they only use sparse retinal vessel landmarks, which are relatively few or non-existent at high magnifications. More importantly, they only perform localization and do not build a map of the vasculature.…”

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

“…However, sequential improvements in the past ten years led to the development of computer-guided photocoagulation systems with integrated imaging and automatic navigation [6,9,10]. Although the visible area of the retina is smaller with the slit-lamp, compared with images acquired by non-mydriatic fundus photography [4,9], magnification, expansion and control offered by the slit-lamp make it a very popular choice for laser delivery in the clinical environment [2]. The imaging setup is based on the eyepiece and microscope optics of the slit-lamp and the magnifying contact lens attached to the eye such that slit illumination is projected onto the retina (Fig.…”