(a) Schematic and (b) photograph of the real-time spectrometer-based FDOCT DSP hardware. (c) Schematic of the real-time swept source-based FDOCT DSP hardware. (d) Photograph of the custom high-speed digitizer board used as the analog front end for the swept source-based hardware.
Block diagram of the FDOCT real-time signal processing chain as implemented in Xilinx System Generator within the MATLAB SIMULINKenvironment.
FDOCT image processing simulation results. Shown are the FDOCT spectra after each processing step. Numbers on the plot correspond to the processing steps shown in Fig. 2. The sampling frequency was 100 MHz for the steps performed in the processing board (2–7) and 50 MHz otherwise. Signals are scaled and offset for better visualization.
Images from spectrometer-based FDOCT systems. Human fovea: (a) composite image averaged from ten single OCT frames and (b) single OCT frame. NFL: nerve fiber layer, GCL: ganglion cell layer, IPL: inner plexiform layer, INL: inner nuclear layer, OPL: outer plexiform layer, ONL: outer nuclear layer, ELM: external limiting membrane, IS/OS: junction between inner and outer segments, RPE: retinal pigment epithelium, CC: choriocapillaris, C: choroid. Monkey fovea: (c) composite image averaged from 20 single OCT frames and (d) single OCT frame. Arrow indicates location of full-thickness laser lesion. Monkey retina: (e) composite image average from 20 single OCT frames at eccentricity and (f) single OCT frame. Zebrafish heart: (g) composite image averaged from 100 single OCT frames of 4-day old zebrafish and (h) single OCT frame. V: ventricle, VA: ventral aorta. All scale bars are .
Single swept source FDOCT images of (a) the tip of a human finger and (b) excised human intestinal tissue.
FFT core performance for simulation in cycles. .
Pipeline delay for each FDOCT signal processing block. .
Article metrics loading...
Full text loading...