Optical coherence tomography (OCT) images the retina noninvasively with micrometer-scale volumetric resolutions. It is an invaluable resource in the clinic for identifying, monitoring, and treating blindness-causing diseases. By shortening illumination wavelengths to the visible range, visible-light OCT (vis-OCT) improves image resolution, provides new scattering contrasts, and enables oxygen saturation (sO2) measurements, expanding the potential for clinical care. In this dissertation, I develop vis-OCT technological improvements that bridge the gap between the laboratory and clinic. First, I develop a scanning algorithm robust against eye motions that visualizes anatomical features like Bruch’s membrane and inner plexiform layer (IPL) with unprecedented detail. Both features, previously inaccessible in vivo, are locations of interest in diseases like macular degeneration and glaucoma, respectively. Next, I define a framework of signals, noises, and systemic biases influencing sO2 measurements called spectral contaminants (SCs). I build upon this framework to develop a new retinal sO2 algorithm that senses, adapts to, and removes SCs in the human retina. This technique enables accurate sO2 measurements in 18 human volunteers with repeatability < 2.5%. Accurate and repeatable sO2 measurements provide the potential to sense retinal diseases at the metabolic level before vision loss occurs. Finally, I address fundamental noise and speed limitations in vis-OCT by implementing a detection scheme called balanced detection (BD). BD increases signal-to-noise ratios up to 25.6 dB and increases imaging speeds 5-fold. Together, these advances satisfy the necessary image quality, speed, and sO2 reliability to translate vis-OCT to the clinic.

Creator

• Rubinoff, Ian

DOI

• https://doi.org/10.21985/n2-skqa-wa11

Subject

• Optics

Language

• en

Alternate Identifier

• etdadmin_upload_923820
• http://dissertations.umi.com/northwestern:16193

Date created

• 2022-01-01

Resource type

• Dissertation

Rights statement

• In Copyright