Compressed single pixel imaging in the spatial frequency domain

Abstract

Abstract. We have developed compressed sensing single pixel spatial frequency domain imaging (cs-SFDI) to characterize tissue optical properties over a wide field of view (35 mm × 35 mm) using multiple near-infrared (NIR) wavelengths simultaneously. Our approach takes advantage of the relatively sparse spatial content required for mapping tissue optical properties at length scales comparable to the transport scattering length in tissue (l tr ∼ 1 mm) and the high bandwidth available for spectral encoding using a single-element detector. cs-SFDI recovered absorption (μ a ) and reduced scattering (μ 0 s ) coefficients of a tissue phantom at three NIR wavelengths (660, 850, and 940 nm) within 7.6% and 4.3% of absolute values determined using camera-based SFDI, respectively. These results suggest that cs-SFDI can be developed as a multi-and hyperspectral imaging modality for quantitative, dynamic imaging of tissue optical and physiological properties.

Characterisation of impaired wound healing in a preclinical model of induced diabetes using wide‐field imaging and conventional immunohistochemistry assays

“…Based on these values, it calculates the StO 2 values . More details on SFDI instrumentation can be found in the literature . Briefly, a 250 W quartz‐tungsten lamp (Newport Oriel, Stratford, Connecticut) coupled to a digital micromirror device (DMD) (Texas Instruments, Dallas, Texas) projects a series of sinusoidal intensity patterns onto a 90 × 66 mm field of view.…”

“…Data were calibrated to a reference phantom having known optical properties . MATLAB (MathWorks, Inc., Natick, Massachusetts) was used to process the data …”

“…28,30,[38][39][40] More details on SFDI instrumentation can be found in the literature. 31,41,42 Briefly, a 250 W quartztungsten lamp (Newport Oriel, Stratford, Connecticut) coupled to a digital micromirror device (DMD) (Texas Instruments, Dallas, Texas) projects a series of sinusoidal intensity patterns onto a 90 × 66 mm field of view. Five spatial frequency patterns equally spaced from 0 to 0.2 mm −1 were projected onto the illumination field, and reflected light from the sample was captured by Charged Coupled Device (CCD) camera (Nuance, CRI, Inc., Woburn, Massachusetts) with a liquid-crystal tunable filter (λ = 650-1100 nm, FWHM = 10 nm).…”

“…43,44 MATLAB (MathWorks, Inc., Natick, Massachusetts) was used to process the data. 42 At each wavelength, optical properties (reduced scattering and absorption coefficients) of the measured tissue were determined by fitting the collected reflectance data to a lookup table for scattering and absorption at both spatial frequencies generated by forward Monte Carlo simulations. 45 A linear least-squares fit algorithm was used to transform the absorption coefficient values into concentration of HbO 2 and Hb using Beer's law.…”

Hyperspectral imaging in the spatial frequency domain with a supercontinuum source

“…As a result, new SFDI strategies for increasing spectral content have been introduced using temporally encoded light sources 23 and single-pixel compressive sensing. 24 To maximize information content from tissues, these methods should provide sufficient bandwidth and resolution to probe multiple endogenous and exogenous chromophores simultaneously, typically in the 100 to 200 nm bandwidth range, with resolution that is on the order of 5 to 10 nm. In addition, for characterizing endogenous tissue components, spectral regions with high-value information content can span broadly from the visible to the near-infrared (NIR), which puts significant demands on strategies for achieving quantitative hyperspectral content.…”