2D Biomedical Hyperspectral Imager - Phase II

DESCRIPTION (provided by applicant): This proposal from KESTREL CORPORATION continues the research work on a new three dimensional (two spatial, one spectral) hyperspectral imager for use in clinical and biomedical research applications. Kestrel's three dimensional hyperspectral retinal imager concept merges two developing technologies, Fourier Transform hyperspectral imaging technology and Integral Field Spectroscopy. The two technologies allow an imaged region to be spatially sampled in a two dimensional area while retaining the spectral information in each sample. These samples are then processed and reassembled to create a full three dimensional hyperspectral data cube containing two spatial dimensions with full spectral content of the selected area. In Phase 1 Kestrel successfully demonstrated the feasibility of the proposed technology. Phase 2 builds on this success by implementing a prototype of a new retinal imager that can be used to advance the knowledge the mechanisms of retinal diseases and the clinical practice to manage such illnesses. The Specific Aims of this proposal are: Aim 1. Address the effects of fabrication of the Integral Field Unit on the spectral data. Aim 2. Obtain the optimum configuration of the Integral Field Unit for medical applications. Aim 3. Build a prototype of a new kind of fundus imager that can collect complete hyperspectral data cubes in single exposures. Aim 4. Collect a comprehensive database of spectral signatures of retinal features from normal subjects and subjects presenting different degrees of diabetic retinopathy and age-related macular degeneration. Aim 5. Demonstrate the clinical value of spectral imaging of the retina by determining features in the spectral information that correlate with the current assessment of retinal health status. Applications for this technology exist across a wide range of biomedical imaging. The value of hyperspectral medical imaging and especially its application in ocular uses is slowly gaining recognition. However, the current implementation of these hyperspectral devices limits their transition to wide spread clinical and research in vivo uses because of the techniques employed in collecting a three dimensional data cube are for practical purposes unusable. The technology proposed offers the potential of overcoming this limitation.