Assessment of morphology and function in retinal degeneration by multi-modal mapping

Various morphological and functional techniques for retinal examination have been established in recent years. For an integrative characterization of retinal structure in clinical routine, the registration of morphological and functional data as well as the extraction of parameters, such as the thickness of certain retinal layers, need to be available in a practicable and efficient way for clinicians. In this work, the development and use of a novel software solution is described, allowing, for the first time, a comprehensive mapping of a multitude of functional and morphological ophthalmological data from a wide range of established routine clinical examination devices, including fundus photography, optical coherence tomography, scanning laser ophthalmoscopy images, GDx retinal nerve fiber layer thickness profiles, multifocal electroretinography, multifocal pattern electroretinography, perimetry, and microperimetry. Different quantitative parameters that can be used for visual status assessment, such as the structural properties of the vascular network, and the integration of the corresponding extraction functionality into the software are presented. The software supports graphics hardware accelerated semi-automatic image registration and provides a special multi-modal mapping view for regional parameter extraction across data from different devices. The main chapter of this work gives a detailed description of the software functionality and its implementation, including data management, import and loading of device data, data visualization, automatic and semi-automatic parameter extraction from 2D and 3D retinal images, registration and mapping of functional and morphological data, and data export for statistical analysis. The use of the software as a research tool in different clinical studies demonstrates its usability by clinicians and provides the proof of concept for the role of the software to generate unprecedented information on the functional impact of retinal pathology. The software has proven to be highly valuable in defining surrogate endpoints in clinical trials and the versatility achieved by supporting various modality types and extractable parameters greatly differentiates the software presented from existing pure retinal image analysis tools.