Abstract
This work reports a deep-learning based registration algorithm that aligns multi-modal retinal images collected from longitudinal clinical studies to achieve accuracy and robustness required for analysis of structural changes in large-scale clinical data. Deep-learning networks that mirror the architecture of conventional feature-point-based registration were evaluated with different networks that solved for registration affine parameters, image patch displacements, and patch displacements within the region of overlap. The ground truth images for deep learning-based approaches were derived from successful conventional feature-based registration. Cross-sectional and longitudinal affine registrations were performed across color fundus photography (CFP), fundus autofluorescence (FAF), and infrared reflectance (IR) image modalities. For mono-modality longitudinal registration, the conventional feature-based registration method achieved mean errors in the range of 39-53 µm (depending on the modality) whereas the deep learning method with region overlap prediction exhibited mean errors in the range 54-59 µm. For cross-sectional multi-modality registration, the conventional method exhibited gross failures with large errors in more than 50% of the cases while the proposed deep-learning method achieved robust performance with no gross failures and mean errors in the range 66-69 µm. Thus, the deep learning-based method achieved superior overall performance across all modalities. The accuracy and robustness reported in this work provide important advances that will facilitate clinical research and enable a detailed study of the progression of retinal diseases such as age-related macular degeneration.
Highlights
Retinal diseases such as age-related macular degeneration (AMD) affect ∼3 million Americans and the associated vision loss among the ageing population creates a growing healthcare burden [1]
Color fundus photography (CFP), fundus autofluorescence (FAF), and infrared reflectance (IR) are 2D imaging modalities that are routinely acquired in the clinic
We investigate the convolutional neural network (CNN)-based architecture resembling the conventional feature-based registration process proposed by Rocco et al
Summary
Retinal diseases such as age-related macular degeneration (AMD) affect ∼3 million Americans and the associated vision loss among the ageing population creates a growing healthcare burden [1]. Diagnosis and prediction of disease progression is important in developing therapeutics to prevent late stage disease and irrecoverable vision loss. In this context, longitudinal clinical studies are critical in understanding the natural history of AMD and other retinal diseases with multi-modal imaging playing an important role in measuring the structural changes that occur in the course of the disease. Co-registering these longitudinally acquired, multi-modal images facilitates the characterization of disease progression as well as assists the exploration of imaging biomarkers that precede vision loss. This work aims to develop a registration framework that automatically aligns several 2D imaging modalities of the retina to advance research in large-scale clinical studies
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
