[go: up one dir, main page]

WO2024259243A2 - Tomographie à matrice de diffusion avec acquisition hyperspectrale d'instantané - Google Patents

Tomographie à matrice de diffusion avec acquisition hyperspectrale d'instantané Download PDF

Info

Publication number
WO2024259243A2
WO2024259243A2 PCT/US2024/034022 US2024034022W WO2024259243A2 WO 2024259243 A2 WO2024259243 A2 WO 2024259243A2 US 2024034022 W US2024034022 W US 2024034022W WO 2024259243 A2 WO2024259243 A2 WO 2024259243A2
Authority
WO
WIPO (PCT)
Prior art keywords
light signal
sample
image
output
input
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/US2024/034022
Other languages
English (en)
Other versions
WO2024259243A3 (fr
Inventor
Quang Minh DINH
Chia Wei Hsu
Yiwen Zhang
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Southern California USC
Original Assignee
University of Southern California USC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Southern California USC filed Critical University of Southern California USC
Publication of WO2024259243A2 publication Critical patent/WO2024259243A2/fr
Publication of WO2024259243A3 publication Critical patent/WO2024259243A3/fr
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/47—Scattering, i.e. diffuse reflection
    • G01N21/4795—Scattering, i.e. diffuse reflection spatially resolved investigating of object in scattering medium
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B9/00—Measuring instruments characterised by the use of optical techniques
    • G01B9/02—Interferometers
    • G01B9/0209—Low-coherence interferometers
    • G01B9/02091—Tomographic interferometers, e.g. based on optical coherence
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/28—Investigating the spectrum
    • G01J3/2823—Imaging spectrometer

Definitions

  • the present invention relates to imaging techniques with improved temporal performance.
  • the Scattering Matrix Tomography (SMT) method for noninvasive 3D imaging within scattering media, such as biological tissue samples, has been previously disclosed.
  • the scattering matrix contains vital information regarding how a sample scatters light at various angles and frequencies. This comprehensive data enables high-resolution 3D imaging deep within scattering materials.
  • the current measurement scheme which necessitates scanning across all input and output channels and frequencies, is time-consuming and impractical for in-vivo imaging.
  • a system for multi-spectral scattering-matrix tomography with snapshot hyperspectral acquisition includes an interference microscope subsystem that includes a broadband light source that provides an input broadband light signal.
  • the interference microscope subsystem is configured to generate a combined broadband interference image signal that includes interference patterns between a sample beam and a reference beam.
  • a hyperspectral imaging subsystem is configured to acquire the interference patterns for a plurality of output channels and a plurality of light frequencies in a single shot collected with a high-speed camera.
  • a computing system is configured to transform the interference patterns into scattering matrix coefficients across the plurality of output channels and the plurality of light frequencies of output.
  • a method for multi-spectral scattering-matrix tomography with snapshot hyperspectral acquisition includes a step of splitting an input broadband light signal into an incident sample light signal and a reference light signal.
  • the input broadband light signal includes a plurality of light frequencies.
  • the incident sample light signal is directed to focus onto a sample such that an output light signal scattered from the sample is generated.
  • the focal point of the incident sample light signal is varied over a predetermined range of areas.
  • the reference light signal is focused onto a reference mirror through an optical path that matches the optical path of the incident sample light signal to form a reflected reference light signal.
  • the reflected reference light signal is passed through a grating which provides a constant wavevector shift for all frequencies.
  • the reflected reference light signal passes through a pinhole used to select a first diffraction order signal from the grating while blocking other orders.
  • the reflected reference light signal and the output sample light signal are combined to form a combined broadband interference image signal.
  • a hyperspectral imaging scheme is applied such that various spectral components of the broadband interference image are separated and captured simultaneously.
  • the hyperspectral imaging scheme includes steps of directing the combined image signal to a spatial light modulator, the spatial light modulator deflecting different image parts into separated angle ranges; collecting the different image parts in the separated angle ranges through a lens array to form a sliced image at a focal plane of the lens array; illuminating the sliced image on a blazed grating where the blazed grating diffracting light signals from the sliced image at different frequencies into different angles to form a dispersed light signal; collecting the dispersed light signal by a lens system; and imaging the dispersed light signal with a high-speed camera, the high-speed camera being placed on a conjugate plane of the spatial light modulator.
  • the camera’s region of interest is determined by a product of output spatial pixels and the number of frequencies in the plurality of light frequencies.
  • a system for multi-spectral scattering-matrix tomography includes a light source configured to generate an input light signal varied over a predetermined frequency range.
  • the system also includes an optical subsystem that splits the input light signal into an incident light signal and a reference light signal and directs the incident light signal to a sample such that an output light signal includes light scattered from or transmitted through the sample, with the incident light signal varied over a predetermined range of incident angles or incident spatial focusing locations.
  • a camera is configured to receive the output light signal and the reference light signal, wherein the reference light signal is directed at a constant angle with respect to the output light signal to allow for amplitude and phase calculation by off-axis holography.
  • a computing device is configured to measure a total light signal as a coherent sum of the reference light signal and the output signal using the camera, collect the total light signal for each light frequency and each incident angle as collected total light signal data, calculate a reflection matrix or transmission matrix from the collected total light signal data, derive an image of the sample from the reflection matrix or transmission matrix by summing over angles and light frequencies, and apply one or more correction algorithms to the reflection matrix or transmission matrix to increase resolution and penetration depth.
  • the one or more correction algorithms includes dispersion compensation, wavefront distortion correction, and spatially varying wavefront distortion correction.
  • a computer-implemented inward-outward progression method involves input-output alternating (IOA) optimization and spatial basis truncation, singular value decomposition (SVD) for multiple scattering removal, Inward progression, and Outward progression.
  • IOA input-output alternating
  • Singular value decomposition (SVD) is used to remove multiple scattering, addressing the non-convex nature of IOA optimization by eliminating local optima.
  • Inward progression focuses on optimizing the brightest zone in the image using IOA optimization, providing an initial guess for adjacent zones.
  • a system for improving image quality in an imaging system by in- out progression includes a light source configured to generate an input light signal varied over a predetermined frequency range and an optical subsystem that splits the input light signal into an incident light signal and a reference light signal and directs the incident light signal to a sample such that an output light signal includes light scattered from or transmitted through the sample, with the incident light signal varied over a predetermined range of incident angles or incident spatial focusing locations.
  • a camera is configured to receive the interference image that combines the output light signal and the reference light signal.
  • a computing device is configured to: a) calculate a reflection matrix or transmission matrix from the collected interference image data and derive an image of the sample from the reflection matrix or transmission matrix by summing over angles and light frequencies; b) preprocess the output image to an initial in-out input image by performing a first singular value decomposition (SVD) to remove multiple scattering; c) identify a brightest location in the initial in-out input image, locate an initial optimization zone as an initial optimization zone around the brightest location, perform spatial basis truncation, and then use input-output alternating (IOA) optimization to optimize correction phases where the c ⁇ n and c° ut are Zernike weights, k 11 is the in-plane wavevector component of the incident light, and is k[j ,ut is the in-plane wavevector component of the scattered or output light, Z n is a wavefront of an n th Zernike polynomials, n is an integer label, initially, the Zernike weights are the
  • FIGURE 1 Schematic of a system for multi-spectral scattering-matrix tomography with snapshot hyperspectral acquisition.
  • FIGURE 2a Depiction of the full scattering matrix.
  • FIGURE 2b Depiction of the truncated scattering matrix.
  • FIGURE 3a Reconstruction formed from the full scattering matrix.
  • FIGURE 3b Reconstruction formed from the truncated scattering matrix.
  • FIGURE 4 Schematic of an input beam module of the system for multi-spectral scattering-matrix tomography of Figure 1.
  • FIGURE 5 Schematic of a sample beam module of the system for multi-spectral scattering-matrix tomography of Figure 1.
  • FIGURE 6 Schematic of a reference beam module and interferometry module of the system for multi-spectral scattering-matrix tomography of Figure 1.
  • FIGURE 7a Schematic of a hyperspectral imaging subsystem of the system for multi- spectral scattering-matrix tomography of Figure 1.
  • FIGURE 7b Schematic of an image slicing component of the hyperspectral imaging subsystem of Figure 7a.
  • FIGURE 7c Schematic of a spectral shearing component of the hyperspectral imaging subsystem of Figure 7 a.
  • FIGURE 8 Schematic of a system for multi-spectral scattering-matrix tomography includes a light source configured to generate an input light signal varied over a predetermined frequency range.
  • FIGURES 9a, 9b, and 9c Virtual wavefront shaping and scattering matrix tomography (SMT).
  • SMT Virtual wavefront shaping and scattering matrix tomography
  • a Conventional imaging and wavefront shaping, with wavefront modulated by lenses and spatial light modulators (SLMs) and using feedback from guidestars
  • b-c (z) Virtual wavefront shaping, with wavefront modulation and optimization performed digitally using noninvasive feedback from the reconstructed image
  • the scattering matrix (k O ut, k , co) relates any incident field Ejn(kj n , a>) to the resulting scattered field E out (k out , o) .
  • SMT corrects all of them digitally through a frequency-dependent phase 7(oi) that acts as a virtual pulse shaper, appropriate momenta and phase coefficients for the medium that act as a virtual index-corrected objective lens, and angle-dependent phases in (kj") and ⁇ > O ut(kfj ,ut ) that act as two virtual SLMs.
  • FIGURES 10a and 10b Measurement of the hyperspectral reflection matrix, a, Off- axis holography is used to measure the phase and amplitude of fields scattered by the sample.
  • BS beam splitter
  • BE beam expander
  • TL tube lens
  • b Construction of the data cube by mapping the output angles with the camera, scanning the input angle with the galvo, and scanning the frequency with the tunable laser.
  • FIGURES I la, 11b, 11c, l id, l ie, I lf, 11g, l lh, Hi, l lj, and I lk.
  • Noninvasive imaging through mouse brain tissue a, Schematic of the sample and a scanning electron microscope image of the buried USAF target, b-d, Reflectance confocal microscopy (RCM), optical coherence tomography (OCT), and optical coherence microscopy (OCM) images at the USAF target plane, synthesized from the measured hyperspectral reflection matrix, e, SMT image.
  • RCM Reflectance confocal microscopy
  • OCT optical coherence tomography
  • OCM optical coherence microscopy
  • the wavefront correction phase maps for the 8 > ⁇ 8 zones in SMT.
  • the SMT PSF of the sample exhibits a peak width comparable to that of the mirror in air and a peak height 70 times that of the OCM PSF.
  • FIGURES 12a, 12b, 12c, 12d, 12e, 12f, 12g, 12h, 12i, 12j, 12k, and 121 Volumetric imaging inside a dense colloid.
  • the sample consists of 500-nm-diameter TiO2 nanoparticles dispersed in PDMS, with an estimated transport mean free path of 1 mm.
  • a-d, SMT, OCM, OCT, and RCM images built from the measured hyperspectral reflection matrix, e-h,
  • a longitudinal slice of the images at y 20.6 pm and close-up views of three particles at different depths in the SMT image, i, Cross sections of the three particles;
  • Ar r - r pea k.
  • j Transverse slices at the depths of the three particles.
  • Scale bars in e and j 10 pm. All images share the same normalization. Volumetric images and 2D slices use the same colorbar.
  • FIGURES 13a and 13b la) is the image before wavefront correction while b) is the image with wavefront correction but not using the inward outward progression.
  • FIGURE 14 Inward progression applying IOA optimization to optimize the entire image.
  • FIGURE 15 Inward progression step showing small region around the brightest spot.
  • FIGURE 16 Outward progression step applying correction phases of the brightest zone.
  • FIGURE 17 Outward progression step applying weights of the brightest zone.
  • FIGURE 18 Outward progression steps to optimize the outer zones.
  • FIGURE 19 Outward progression steps to optimize the outer zones.
  • integer ranges explicitly include all intervening integers.
  • the integer range 1-10 explicitly includes 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10.
  • the range 1 to 100 includes 1, 2, 3, 4. . . . 97, 98, 99, 100.
  • intervening numbers that are increments of the difference between the upper limit and the lower limit divided by 10 can be taken as alternative upper or lower limits. For example, if the range is 1.1. to 2.1 the following numbers 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, and 2.0 can be selected as lower or upper limits.
  • the term “less than” includes a lower non-included limit that is 5 percent of the number indicated after “less than.”
  • a lower nonincludes limit means that the numerical quantity being described is greater than the value indicated as a lower non-included limited.
  • “less than 20” includes a lower non-included limit of 1 in a refinement. Therefore, this refinement of “less than 20” includes a range between 1 and 20.
  • the term “less than” includes a lower non-included limit that is, in increasing order of preference, 20 percent, 10 percent, 5 percent, 1 percent, or 0 percent of the number indicated after “less than.”
  • connection to means that the electrical components referred to as connected to are in electrical communication.
  • connected to means that the electrical components referred to as connected to are directly wired to each other.
  • connected to means that the electrical components communicate wirelessly or by a combination of wired and wirelessly connected components.
  • connected to means that one or more additional electrical components are interposed between the electrical components referred to as connected to with an electrical signal from an originating component being processed (e.g., filtered, amplified, modulated, rectified, attenuated, summed, subtracted, etc.) before being received to the component connected thereto.
  • the term “electrical communication” means that an electrical signal is either directly or indirectly sent from an originating electronic device to a receiving electrical device. Indirect electrical communication can involve processing of the electrical signal, including but not limited to, filtering of the signal, amplification of the signal, rectification of the signal, modulation of the signal, attenuation of the signal, adding of the signal with another signal, subtracting the signal from another signal, subtracting another signal from the signal, and the like. Electrical communication can be accomplished with wired components, wirelessly connected components, or a combination thereof. [0046] The term “one or more” means “at least one” and the term “at least one” means “one or more.” The terms “one or more” and “at least one” include “plurality” as a subset.
  • computing device refers generally to any device that can perform at least one function, including communicating with another computing device.
  • a computing device includes a central processing unit that can execute program steps and memory for storing data and a program code.
  • Examples of computing devices include, but are not limited to, desktop computers, notebook computers, laptop computers, mainframes, mobile phones, headsets such as augmented reality headsets, virtual reality headsets, mixed reality headsets, augmented reality devices, virtual reality devices, mixed reality devices, and the like.
  • a computing device When a computing device is described as performing an action or method step, it is understood that the one or more computing devices are operable to and/or configured to perform the action or method step typically by executing one or more lines of source code.
  • the actions or method steps can be encoded onto non-transitory memory (e.g., hard drives, optical drive, flash drives, and the like).
  • the processes, methods, or algorithms disclosed herein can be deliverable to/implemented by a processing device, controller, or computer, which can include any existing programmable electronic control unit or dedicated electronic control unit.
  • the processes, methods, or algorithms can be stored as data and instructions executable by a controller or computer in many forms including, but not limited to, information permanently stored on non-writable storage media such as ROM devices and information alterably stored on writeable storage media such as floppy disks, magnetic tapes, CDs, RAM devices, and other magnetic and optical media.
  • the processes, methods, or algorithms can also be implemented in a software executable object.
  • the processes, methods, or algorithms can be embodied in whole or in part using suitable hardware components, such as Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), state machines, controllers, or other hardware components or devices, or a combination of hardware, software and firmware components.
  • suitable hardware components such as Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), state machines, controllers, or other hardware components or devices, or a combination of hardware, software and firmware components.
  • PBS means polarizing beam splitter
  • SMT Scattering Matrix Tomography
  • SMT single -spectral scattering-matrix tomography
  • Traditional SMT involves a method for 3D imaging within scattering media, such as biological tissues, by capturing detailed information on how the sample scatters light at various angles and frequencies. This process typically requires scanning across all input and output channels and frequencies, which can be time-consuming and impractical for certain applications, particularly in-vivo imaging.
  • Traditional SMT systems use a comprehensive scattering matrix that includes numerous coefficients to represent the scattering characteristics of the sample. The scattering matrix contains vital information about how light interacts with the sample, detailing the scattering behavior at different angles and frequencies. This matrix is used to reconstruct high -resolution 3D images of the sample's internal structure.
  • an incident broadband light signal is split into incident and reference signals.
  • the incident light is directed onto the sample, generating scattered output light, while the reference light follows a separate optical path.
  • the scattered output light from the sample and the reflected reference light are then combined to form a broadband interference image signal.
  • the setup typically includes various optical components like beam splitters, gratings, and lenses to direct and manipulate the light signals.
  • the comprehensive scattering matrix requires scanning and measuring across a wide range of spatial points and frequencies, which involves significant data collection and processing time. This approach, while providing detailed imaging, is often limited by the extensive time required for data acquisition and reconstruction.
  • system 10 is designed for multi -spectral scattering-matrix tomography with snapshot hyperspectral acquisition.
  • System 10 includes an interference microscope subsystem 12 and a hyperspectral imaging subsystem 14.
  • the interference microscope subsystem 12 includes a broadband light source.
  • the interference microscope subsystem is configured to generate a combined broadband interference image signal that includes interference patterns between a sample beam and a reference beam.
  • the hyperspectral imaging subsystem 14 is configured to acquire the interference patterns for a plurality of output channels and a plurality of light frequencies in a single shot collected with a high-speed camera.
  • the hyperspectral imaging subsystem 14 is configured to transform the interference patterns into two-dimensional (2D) spectral image blocks and collect the 2D spectral image blocks with a high-speed camera.
  • the new system is designed to expedite the imaging process and facilitate in-vivo imaging.
  • This approach employs a truncated scattering matrix scheme, which maintains sufficient information for high-quality imaging while significantly reducing the number of coefficients requiring measurement.
  • the snap-shot hyperspectral imaging method is introduced. This method enables the simultaneous measurement of scattering matrix coefficients across hundreds of output channels and frequencies in a single shot. This method also takes advantage of the higher recording efficiency (# pixels per second) of the larger region of interest of the high-speed camera.
  • a computing system 16 is configured to transform the interference patterns into scattering matrix coefficients across the plurality of output channels and the plurality of light frequencies of output.
  • Sf u n(r out , r in ) is the full scattering matrix in the spatial basis where r in is the input focused location and r out is the output detection location.
  • Sf U n( r out' r in)> r in and r out scan over the entire field of view.
  • ⁇ Ar, r in ) is the truncated scattering matrix that contains only the near diagonal elements, with the input and output distance
  • the truncated scattering matrix can provide sufficient information for high-quality imaging.
  • S full (r out , r in ) including all the spatial points across the entire field of view of 50x50 pm 2 S truncated (
  • ⁇ Ar 5 pm , r in ) can maintain a good imaging quality.
  • the two benefits from using 5 truncated includes (1) reducing data size by at least 25 times (2) within the sample’s decorrelation time, the area size needs to be scanned and measured is 25 times smaller (10 x 10 compared to 50 x 50 pm 2 ).
  • Figure 2a provides a reconstruction using the full scattering matrix while Figure 2b provides an analogous reconstructions using the truncated scattering matrix.
  • multi-spectral scattering-matrix tomography system 10 includes an interference microscope subsystem 12 and a hyperspectral imaging subsystem 14.
  • Interference microscope subsystem 12 includes input beam module 20, sample beam module 22 and reference beam module 24.
  • Input beam module 20 is configured to provide and collimate the input broadband light signal, polarize it, and split it into an incident sample light signal and a reference light signal.
  • Interference microscope subsystem 12 forms the reference beam and sample beam from broadband source 26.
  • Sample beam module 22 is configured to focus the incident sample light signal onto the sample and scan over a sample plane. The sample beam is the light directed at the sample, where it interacts with the sample and scatters (including reflection and transmission), carrying information about the sample's internal structure.
  • Interferometry module 28 combines the reflected reference light signal 30 and the output sample light signal 32 to form a combined broadband interference image signal.
  • Hyperspectral imaging subsystem 14 includes a spatial light modulator 34 configured to deflect different image parts of a combined broadband interference image signal into separated angle ranges.
  • a lens array 36 is configured to collect different spatial regions in the separated angle ranges and form a sliced image at a focal plane of plane of the lens array 36.
  • a blazed grating 38 is configured to diffract light signals from the sliced image at different frequencies into different angles to form a dispersed light signal.
  • a lens system 40 is configured to collect the dispersed light signal.
  • a highspeed camera 42 is positioned on a conjugate plane of the spatial light modulator 34 and is configured to image the dispersed light signal.
  • the hyperspectral imaging scheme includes the step of directing the combined image signal to the spatial light modulator 34.
  • the spatial light modulator 34 deflects different image parts into separated angle ranges.
  • the scheme also includes a step of collecting the different image parts in the separated angle ranges through the lens array to form the sliced image at the focal plane of the lens array.
  • the scheme also includes a step of illuminating the sliced image on the blazed grating 38, where the blazed grating diffracts light signals from the sliced image at different frequencies into different angles to form the dispersed light signal.
  • the scheme also includes steps of collecting the dispersed light signal by the lens system 40 and imaging the dispersed light signal with the high-speed camera 42.
  • the high-speed camera 42 is placed on the conjugate plane of the spatial light modulator 34.
  • an input broadband light signal 46 from broadband source 26 is split into a reference light signal 48 and an incident sample light signal 50 (step a).
  • the input broadband light signal 46 includes a plurality of light frequencies.
  • the incident sample light signal 50 is directed to focus onto sample 52 such that an output light signal scattered from the sample is generated.
  • the focal point of the incident sample light signal is varied over a predetermined range of areas (step b).
  • the reference light signal 48 is focused onto a reference mirror 54 through an optical path that matches the optical path of the incident sample light signal to form the reflected reference light signal 30 (step c).
  • the reflected reference light signal 30 is passed through a grating 56 which provides a constant wavevector shift for all frequencies (step d).
  • the reflected reference light signal is passed through a pinhole 58 used to select a first diffraction order signal from the grating 56 while blocking other orders (step e).
  • the reflected reference light signal 30 and the output sample light signal 32 are combined to form a combined broadband interference image signal by interferometry module 28 (step f).
  • a hyperspectral imaging scheme which includes steps of directing the combined image signal to a spatial light modulator 34 where the spatial light modulator deflects different image parts into separated angle ranges (step g 1 ); collecting the different image parts in the separated angle ranges through a lens array 36 to form a sliced image at a focal plane of the lens array (step g2); illuminating the sliced image on a blazed grating 38 where the blazed grating diffracting light signals from the sliced image at different frequencies into different angles to form a dispersed light signal (step g3); collecting the dispersed light signal by a lens system 40 (step g4); and imaging the dispersed light signal with a high-speed camera 42 where the high-speed camera being placed on a conjugate plane of the spatial light modulator 34 (step g5).
  • a region of interest is determined by a product of output spatial pixels and the number of frequencies in the plurality of light frequencies.
  • step a) is performed by an input beam module 20
  • step b) is performed by a sample beam module 22
  • steps c) through e) are performed by a reference beam module 24
  • step f) is performed by an interferometry module 28
  • step g) is performed by a hyperspectral imaging subsystem 14.
  • the input beam module 20 includes a broadband source 26 that provides the input broadband light signal 46 and a collimating lens 64 that receives and collimates the input broadband light signal 46.
  • Input beam module 20 also includes a first polarized beam splitter 66 that linearly polarizes the input broadband light signal from the collimating lens 64.
  • Input beam module 20 also includes a second polarized beam splitter 67 that splits the input broadband light signal into the incident sample light signal and the reference light signal and a Fresnel rhomb positioned 70 between the first polarized beam splitter and the second beam splitter for adjusting a power ratio between the incident sample light signal and the reference light signal.
  • the sample beam module 22 includes a pinhole 72 placed on the conjugate plane of the sample plane to truncate the output, selectively allowing outputs near the input location while blocking outputs distant from the incident focus location.
  • a MEMS scanner 74 is placed on the conjugate plane of a back focal plane of the objective lens 68. Characteristically, the MEMS scanner 74 is configured to scan a focused input over a sample plane in sample 52.
  • Sample beam module 22 also includes a quarter wave plate 76 through which the incident sample light signal and the output scattered light signal pass.
  • Objective lens (OL) 68 focuses the incident sample light signal on the sample plan.
  • reference beam module 24 includes reference mirror 54, grating 56, and a pinhole 58 as described above.
  • Reference beam module 24 also includes a quarter wave plate 82 through which the reference light signal and the reflected reference light signal pass.
  • the reference beam light is focused on a reference mirror 54.
  • Reference beam module 24 also includes optical elements configured to match an optical path of the reference light signal to the optical path of the incident sample light signal.
  • Lens 1, Lens 2, Lens 3, Lens 4, and OL are the same models as the ones used in the sample beam to match the optical path.
  • the reflected light from the reference mirror 54 goes through a grating 56 which provides a constant wavevector shift for all the frequencies.
  • a pinhole 58 is used to select the first diffraction order signal from the grating and block other orders.
  • the use of the polarized beam splitter 55 and the quarter wave plate eliminates power loss during beam splitting, thereby providing a 4-fold increase in power efficiency.
  • the interferometry imaging module 28 includes a non-polarized beam splitter 84 that combines the reflected reference light signal and the output sample light signal.
  • hyperspectral imaging subsystem 14 includes the spatial light modulator 34, a first lens 86 that provides the different image parts in the separated angle ranges through the lens array 36 and lens 90, the blazed grating 38, the lens system 40, and the high-speed camera 42.
  • the input image is projected on a spatial light modulator 34, which is placed at the front focal plane of lens 86.
  • the SLM 34 deflects different image parts, parti, part2, part3, ... into separated angle ranges: angle range 1, angle range 2, angle range 3, ... All the angle ranges are collected by lens 86 and go through the lensletl, lenslet2, lenslet3, ... of a lens array 36.
  • a sliced image is formed at its focal plane.
  • the sliced image then goes through lens 90 and is illuminated on a blazed grating 38.
  • the blazed grating 38 diffracts the light signals at different frequencies into different angles.
  • the dispersed light signal is collected by lens system 40 and imaged by a high-speed camera.
  • the camera sensor is placed on the conjugate plane of the SLM.
  • the combination of a polarized beam splitter and a quarter wave plate gives a 4 times power efficiency than the case with a regular nonpolarized beam splitter (NBS).
  • NBS nonpolarized beam splitter
  • a method for multi-spectral scattering-matrix tomography with snapshot hyperspectral acquisition includes combining a reflected reference light signal and an output sample light signal to form a combined broadband interference image signal.
  • the reflected reference light signal and the output sample light signal include a plurality of frequencies.
  • the method also includes applying a hyperspectral imaging scheme. This scheme involves directing the combined broadband interference image signal to a spatial light modulator, which deflects different image parts into separated angle ranges. The different image parts are collected through a lens array to form a sliced image at a focal plane of the lens array.
  • the sliced image is illuminated on a blazed grating, which diffracts light signals from the sliced image at different frequencies into different angles to form a dispersed light signal that includes interference patterns.
  • the dispersed light signal is collected by a lens system and imaged with a high-speed camera placed on a conjugate plane of the spatial light modulator.
  • the region of interest of the camera is determined by a product of output spatial pixels and the number of frequencies in the plurality of light frequencies.
  • a computing device calculates scattering matrix coefficients across a plurality of output channels and the plurality of light frequencies by transforming the images collected by the camera into a plurality of interference patterns at different light frequencies and processing these interference patterns to calculate the phase and amplitude of scattering matrix coefficients across the plurality of output channels and the plurality of light frequencies.
  • the method for multi-spectral scattering-matrix tomography includes creating the combined broadband interference image signal. This is done by splitting an input broadband light signal into an incident sample light signal and a reference light signal, where the input broadband light signal includes a plurality of light frequencies.
  • the incident sample light signal is directed to focus onto a sample, generating an output light signal scattered from the sample, with the focal point of the incident sample light signal varied over a predetermined range of areas.
  • the reference light signal is focused onto a reference mirror through an optical path matching the optical path of the incident sample light signal to form a reflected reference light signal.
  • the reflected reference light signal passes through a grating providing a constant wavevector shift for all frequencies, and then through a pinhole used to select a first diffraction order signal from the grating while blocking other orders.
  • the reflected reference light signal is then combined with the output sample light signal to form an interference pattern.
  • the steps of the method for creating the combined broadband interference image signal are performed by specific subsystems.
  • Step a) is performed by an input beam subsystem
  • step b) is performed by a sample beam subsystem
  • steps c) through e) are performed by a reference beam subsystem
  • step f) is performed by an interferometry subsystem.
  • the input beam subsystem includes a broadband source that provides the input broadband light signal.
  • the input beam subsystem also includes a collimating lens that receives and collimates the input broadband light signal.
  • a first polarized beam splitter linearly polarizes the input broadband light signal from the collimating lens.
  • a second polarized beam splitter splits the input broadband light signal into the incident sample light signal and the reference light signal.
  • a Fresnel rhomb is positioned between the first polarized beam splitter and the second polarized beam splitter to adjust the power ratio between the incident sample light signal and the reference light signal.
  • the sample beam subsystem includes an objective lens that focuses the incident sample light signal on the sample.
  • a pinhole is placed on the conjugate plane of a sample plane to truncate the output sample light signal, selectively allowing outputs near an input location while blocking outputs distant from an incident focus location.
  • a MEMS scanner is placed on the conjugate plane of a back focal plane of the objective lens, and the MEMS scanner is configured to scan a focused input over the sample plane.
  • the sample beam subsystem also includes a quarter wave plate through which the incident sample light signal and the output scattered light signal pass.
  • the reference beam subsystem includes a quarter wave plate through which the reference light signal and the reflected reference light signal pass.
  • the use of a polarized beam splitter and the quarter wave plate in the reference beam subsystem eliminates power loss during beam splitting, thereby providing a four-fold increase in power efficiency.
  • the interferometry subsystem includes a non-polarized beam splitter that combines the reflected reference light signal and the output sample light signal.
  • the hyperspectral imaging scheme is performed by a hyperspectral imaging subsystem.
  • the hyperspectral imaging subsystem includes the spatial light modulator, a first lens that provides the different image parts in the separated angle ranges through the lens array, the blazed grating, the lens system, and the high-speed camera.
  • a system and method for multi-spectral scattering-matrix tomography includes a light source 102 configured to generate an input light signal varied over a predetermined frequency range.
  • the SMT system 100 also includes an optical subsystem 104 that splits the input light signal 106 into an incident light signal 108 and a reference light signal 110.
  • Optical subsystem 104 directs the incident light signal to sample 112 such that an output light signal 114 includes light scattered from or transmitted through the sample, with the incident light signal varied over a predetermined range of incident angles.
  • Sample 112 can be moved by movable sample holder 116.
  • An imaging camera 118 is configured to receive the output light signal and the reference light signal, wherein the reference light signal is directed at a constant angle with respect to the output light signal to allow for amplitude and phase calculation by off-axis holography.
  • System 100 includes a computing device 120 (i.e., a computing device) configured to measure a total light signal as a coherent sum of the reference light signal and the output signal using the camera.
  • Computing device 120 is configured to collect a digitized total light signal for each light frequency and each incident angle as collected total light signal data.
  • Computing device 120 is configured to calculate a reflection matrix or transmission matrix from the collected total light signal data.
  • the computing device 120 is also configured to derive an image of the sample from the reflection matrix or transmission matrix by summing over angles and light frequencies.
  • the computing device 120 is configured to apply one or more correction algorithms to the reflection matrix or transmission matrix to increase resolution and penetration depth, including dispersion compensation, wavefront distortion correction, and spatially varying wavefront distortion correction.
  • the computing device 120 is configured to determine the reflection matrix or transmission matrix by Fourier transforming the collected total light signal data to form transformed collected total signal data and performing an inverse Fourier transform on a first-order region of the transformed collected total signal data to determine the amplitude and phase of the output signal.
  • optical subsystem 104 includes a beam splitter 126 configured to split the input light signal into an incident light signal and a reference light signal.
  • a method for multi-spectral scattering-matrix tomography includes a step of splitting an input light signal into an incident light signal and a reference light signal, wherein the input light signal is varied over a predetermined frequency range.
  • the method also includes steps of directing the incident light signal to a sample in either a reflection configuration or a transmission configuration such that an output light signal includes light scattered from or transmitted through the sample, wherein the incident light signal is varied over a predetermined range of incident angles; directing the output light signal and the reference light signal to a camera, the output light signal directed at a constant angle with respect to the reference light signal to allow for amplitude and phase to be calculated by off-axis holography; measuring with the camera a total light signal that is a coherent sum of the reference light signal and the output signal; collecting the total light signal for each light frequency and each incident angle as collected total light signal data; calculating with a computing device a reflection matrix or transmission matrix from the collected total light signal data; and deriving an image of the sample from the reflection matrix or transmission matrix by summing over angles and summing over light frequencies.
  • one or more computer-implemented correction methods are applied to increase resolution and penetration depth.
  • the reflection matrix or transmission matrix is determined by Fourier transforming the collected total light signal data to form a transformed collected total signal data and performing an inverse Fourier transform on a first-order region of the transformed collected total signal data to determine amplitude and phase of the output signal.
  • image intensity is determined from: where:
  • I SMT is the image intensity as a function of position r in the sample; r is a position vector of a point in the sample;
  • S(k out , kin, ⁇ ) is an element of the scattering matrix for an incidence channel with k in and a reflection channel with k out ;
  • kin is a wavevector of the incident light signal;
  • kout is the wavevector of the output (i.e., reflected) light signal;
  • ⁇ ( ⁇ o) is a spectral phase;
  • ⁇ in is an input correction phase;
  • ⁇ out is an output correction phase;
  • ⁇ is the light frequency.
  • the one or more correction algorithms include a dispersion compensation algorithm comprising: introducing a frequency-dependent phase shift ⁇ ( ⁇ ) to the scattering matrix; precomputing an angular summation from: such that wherein 0 (co) is determined by maximizing an image quality metric is the scattering matrix.
  • one variable at a time is optimized such that first a temporal gate is aligned with a spatial gates by scanning to obtain while keeping 0, then symmetric pulse compression is performed by scanning to obtain while keeping 0, and finally, asymmetric pulse compression is performed by scanning to obtain while keeping x .
  • a local gradient based optimization is performed to optimize all three variables simultaneously, using ( , ) as an initial guess wherein a gradient of M with respect to is where I o is a reference intensity that is a normalization constant.
  • the one or more correction algorithms include a wavefront distortion correction algorithm comprising an alternating optimization scheme including steps of: a) determining a target’s depth from a dispersion-compensated volumetric image by scanning a longitudinal direction z and selecting the depth where a 2D en face image yields a highest M. b) building a depth-resolved time-gated scattering matrix at a target’s depth zo from: where k° u1: and kj” are tranverse input and output wavenumbers, and are the longitudinal input and output wavenumbers, and c is the speed of light in the sample.
  • e) determine the gradient of AY with respect to the Zemike weights e) optimize AY; f) building a matrix , co) with each column containing the image from a single output wherein these single-output images, denoted as are built by: g) determining the gradient of M with respect to the Zemike weights h) optimize AY; and i) alternate between optimizing ⁇ and until A converges.
  • the one or more correction algorithms include a spatially varying wavefront distortion correction algorithm comprising: a) dividing the image into small zones, each zone being corrected with its own correction phase maps wherein optimization of these zones’ images is non-convex; b) optimizing sharpness of a whole image is the input-output alternating optimization scheme; c) dividing the whole image into 2 x 2 smaller zones with equal sizes wherein these zones are slightly overlapped; d) optimizing each zone separately with Zemike weights of the whole image being an initial guess; e) after optimization, further dividing each zone into 2 x 2 smaller zones, where the Zernike weights of each big zone serve as the initial guesses for the smaller zones it contains f) repeating steps c) to e) until no further improvement of image quality is observed, wherein the number of optimized Zernike polynomials is also increased after each division step with the weights of newly added Zernike polynomials are initialized as zeros.
  • Figure 9a provides a schematic of conventional imaging and wavefront shaping, with wavefront modulated by lenses and spatial light modulators (SLMs) and using feedback from guidestars.
  • Figure 9b-c illustrates the virtual spatiotemporal wavefront shaping approach, as described above. This approach not only combines all the strengths of the conventional methods but goes beyond them as schematically illustrated in Figure 9b-c, z.
  • the hyperspectral scattering matrix of the sample was measured and used to virtually perform spatiotemporal focusing with high-speed Guidestar-free wavefront optimization for every isoplanatic patch, pulse compression, and refractive index mismatch correction.
  • the focus digitally can be scanned to yield a phase-resolved 3D image of the sample with no depth-of-field trade-off.
  • the scattering matrix [59, 60] encapsulates the sample’s complete linear response ( Figure 9b-c (zz)): any incident wave is a superposition of plane waves over momentum ki n and frequency co, and the resulting outgoing wave is given by the scattering matrix through .
  • the angular summations are restricted to in a background medium with speed of light c/n bg .
  • the sample’s response is digitally synthesized to virtually perform tailored measurements in space-time for customized spatiotemporal inputs.
  • Digitally scanning r forms a phase-resolved 3D image of the sample where the three gates align at every point ( Figure 9b-c, /). This enables high lateral resolution and high axial resolution across a wide field of view and large depth of field, with no restriction on any focal plane.
  • the non-uniform fast Fourier transform [61] is used to efficiently evaluate these summations and the spatial scan. This is the minimal form of “scattering matrix tomography” (SMT).
  • MT efficiently suppresses multiple scattering.
  • Such single-scattering contributions add up in phase in Eq. (1) to form the image, similar to an inverse Fourier transform from i](q) to r](r); meanwhile, the multiple-scattering contributions add up with quasi-random phases.
  • the triple summations over co, kout, and k in boost the single-to- multiple-scattering ratio enable imaging even when multiple scattering is orders of magnitude stronger than single scattering in the raw data 5(k O ut, ki n , co).
  • SMT also allows customized spatial and spectral corrections.
  • the frequency dependence of the refractive index (i.e., dispersion) in the optical elements of the system and the sample creates a frequency-dependent phase that misaligns and broadens the temporal gate.
  • SMT can overcome dispersion using a spectral phase 0(a), acting as a virtual pulse shaper ( Figure 9b-c, iv).
  • the refractive index mismatch between the sample medium (e.g., biological tissue), the far field (e.g., air), and the coverslip (if there is one) refracts the rays and degrades the gates.
  • the sample medium e.g., biological tissue
  • the far field e.g., air
  • the coverslip if there is one
  • SMT can achieve an ideal focus even in the presence of refraction, effectively creating a virtual dry objective lens that perfectly focuses inside any refractive index at any depth ( Figure 9b-c, v) without expensive hardware or liquid immersion.
  • the general form of SMT reads
  • the measurement can use angular [64] or frequency [65, 66] scans at high speed, limited only by the camera frame rate (which can go beyond MHz for commercial cameras, orders of magnitude faster than the fastest SLMs).
  • the subsequent wavefront optimization requires no additional measurement and outpaces SLM-based optimization even more.
  • the absence of spatial or temporal focus during measurement prevents photodamage to the sample and also avoids localized saturation of the camera to allow for a higher signal-to-noise ratio.
  • the synthesized pulse is concentrated at the target arrival time instead of spreading out across different times, providing another signal -to-noise- ratio advantage similar to that of frequency-domain OCT over time-domain OCT.
  • the triple summation of SMT additionally suppresses uncorrelated noises in the scattering matrix data.
  • the detection sensitivity currently limited by the residual reflection from the objective lens, is 90 dB.
  • the 8th group (whose sixth element has a bar width of 1.1 pm) of a 1951 USAF resolution target underneath 0.98 mm of mouse brain tissue (Figure I la) is imaged.
  • the imaging plane is chosen to maximize the total signal.
  • corrections for the dispersion and input aberration of the optical system are included, as well as the dispersion of the sample.
  • none of the three methods can reveal any group-8 element due to the overwhelming scattering from the mouse brain tissue.
  • the SMT image also corrects the air-glass-sample index mismatches and incorporates optimized wavefronts that correct the multiple scattering from the sample and the output aberration of the optical system.
  • the image quality metric over the full 50x50 pm 2 image with the same pin out; the summation over the many image pixels suppresses local oscillations of the objective function and makes the optimization problem more convex.
  • the field of view 50 x 50 pm2
  • the spatial dependence of the optimal wavefront is significant. Therefore, the image is progressively bisected and c
  • the gradient-based optimization algorithm is used to perform local optimizations, using up to 275 Zernike polynomials in each of the 64 zos.
  • the resulting (j) m and ⁇ ou t are shown in Figure I lf; note out(k
  • ) # i n (— k
  • SMT can image the USAF target with near perfection down to the smallest element of group 8 ( Figure He).
  • the speckled OCM PSF averages to be 70 dB below the peak PSF of a mirror without the brain tissue ( Figure I lk), so the signal (which is buried beneath the speckled background and not visible here) has been reduced by at least ten-million-fold due to multiple scattering.
  • the peak of the SMT PSF is 70 times higher than the averaged OCM PSF, indicating that the input and output wavefront corrections and the index-mismatch correction increase the signal by at least 70-fold.
  • the SMT peak’s full width at half maximum (FWHM) is 1.08 pm, close to the 0.93 pm FWHM of the mirror PSF, demonstrating diffraction limited resolution despite the overwhelming multiple scattering.
  • the depth-over-resolution ratio is 910 here.
  • the lateral FWHM resolution is found to be submicron and the axial FWHM resolution close to the theoretical limit of 1.28 pm for the 206 nm bandwidth here, across the whole volume.
  • SMT can work with reflection, transmission, remission, or a combination of them (such as a 4Pi microscope), in any basis.
  • One may use the phase information of ipsm for digital staining and to resolve small nm-scale displacements for neuro imaging.
  • One may incorporate polarization gating to select birefringent objects such as directionally oriented tissues.
  • the hyperspectral scattering matrix can additionally resolve spectral information of the sample, such as the oxygenation of the hemoglobin.
  • Additional details are found in Zhang Y, Dinh M, Wang Z, Zhang T, Chen T, Hsu CW. Deep imaging inside scattering media through virtual spatiotemporal wavefront shaping. arXiv preprint arXiv:2306.08793. 2023 Jun 15; the entire disclosure of which is hereby incorporated by reference
  • system 100 includes a light source 103 configured to generate an input light signal varied over a predetermined frequency range.
  • the system also includes an optical subsystem 14 that splits the input light signal into an incident light signal and a reference light signal and directs the incident light signal to a sample such that an output light signal includes light scattered from or transmitted through the sample, with the incident light signal varied over a predetermined range of incident angles.
  • a camera is configured to receive the output light signal and the reference light signal and form an output image of a field of view.
  • the system includes a computing device 120 configured to preprocess the output image to an initial in-out input image by performing a first singular value decomposition (SVD) to remove multiple scattering.
  • the Zernike weights initially are for the entire field of view.
  • the computing device iteratively shrinks the optimization zone and performs spatial basis truncation and IOA optimization for each progressively smaller zone using more Zemike polynomials than the initial optimization zone, where the c ⁇ ” and c° ut of a previous zone are initial parameters for the current zone. This shrinking of zones is repeated until a zone’s image attains a predefined sharpness.
  • the computing device then iteratively expands the optimization zone outward by applying optimized correction phases (Zemike weights), ensuring that each newly included zone is optimized by performing additional IOA optimization steps, thereby progressively enhancing the image quality across the entire image. Finally, the system stitches the zones together to form a complete image of the field of view.
  • Inward progression involves identifying the brightest zone in the image. This includes performing spatial basis truncation to limit the optimization to the relevant zones of the image. The process iteratively shrinks the zone and performs input-output alternating (IO A) optimization with increased correction detail using more Zemike polynomials.
  • the spatial basis truncation involves converting the time-gated reflection matrix to spatial basis using a Fourier transform. The matrix elements outside the zone of interest are set to zero, and the truncated matrix is then converted back to angular basis for focused optimization.
  • Out outward progression involves using the correction phases from the brightest zone to optimize surrounding zones.
  • Singular Value Decomposition can be performed on the timegated reflection matrix to remove multiple scattering background.
  • the optimized correction phases are iteratively applied to further zones to enhance.
  • the images input to inward-outward progression are preprocessed with IOA optimization and spatial basis truncation and SVD for multiple scattering.
  • the purpose of IOA optimization and spatial basis truncation is to iteratively optimize the correction phases for input and output wavefronts in an imaging system to improve the image quality by removing distortions caused by multiple scattering and other aberrations.
  • the process aims to find the best possible correction phases that maximize the figure of merit M, which is a measure of image quality.
  • M which is a measure of image quality.
  • This step is crucial for achieving high-resolution, high-quality images in environments where wavefront distortions are significant, such as in biological tissues or other highly scattering media.
  • the spatial basis truncation ensures that the optimization is focused on the relevant zones of the image, further enhancing the accuracy and efficiency of the correction.
  • This step starts from the time-gated reflection matrix ) . Each row is one output with wavenumber 1 , each column is one input with wavenumber .
  • the optimization finds the correction phases and for each input and each output.
  • the corrected image is built from the by
  • the correction phases are parameterized using Zernike polynomials and and are the weights ofthe n th Zernike polynomials, Z coordinate is the wavefront of the n th Zernike polynomials.
  • the number of Zernike polynomials is a free parameter that should be sufficient to capture fast-varying wavefront distortion.
  • the computing device 120 is further configured to optimize a figure of merit M for input optimization and output optimization.
  • the figure of merit M is a function of and examples include a normalization constant).
  • the gradient of M with respect to is:
  • M is optimized using a gradient-based optimization algorithm. Then, perform output optimization to find . Precompute the single-output complex images: such that and c is the result of the previous input optimization.
  • the gradient of M with respect to is:
  • computing device 120 is further configured to: precompute single-output complex images: such that and Cn is a result of the previous input optimization; return to find c, 1 " using the same process, with the initial estimate being the cj from the previous step; find c° ut following the same method, using the c° ut from the prior step as the initial value, where the result of each input optimization, serves as the initial estimate for a next input optimization, the process of alternating between input and output optimization is repeated until the figure of merit M converges.
  • the purpose for multiple scattering removal is to enhance image quality by removing the effects of multiple scattering, which manifests as a speckled background in images taken through scattering media.
  • Singular Value Decomposition Singular Value Decomposition (SVD) to the time-gated reflection matrix, the method isolates and removes small singular values that correspond to multiple scattering contributions. Retaining only the large singular values, which represent the stronger single scattering signals, allows for a clearer, more accurate image.
  • the singular value decomposition includes setting a threshold value to differentiate between small singular values corresponding to multiple scattering and large singular values corresponding to single scattering and removing singular values below the threshold to eliminate multiple scattering contributions.
  • computing device 120 is configured to convert the optimized matrix back to the angular basis, resulting in a reflection matrix that highlights single scattering signals and reduces noise from multiple scattering, thereby improving the overall image resolution and quality.
  • the time-gated reflection matrix S(k[j >ut , kj", z 0 ) is converted to spatial basis S(r
  • the reflection matrix after SVD is:
  • the output image of a field of view is processed by input-output alternating (IOA) optimization followed by a first SVD.
  • IOA input-output alternating
  • the IOA optimization involves precomputing single-input and single-output complex images for each input and output wavefront, constructing a total image intensity from the precomputed images, and maximizing an image quality metric
  • the purpose of the inward progression is to incrementally refine the image quality by focusing on and optimizing the brightest zone, gradually shrinking the zone while increasing the correction detail using more Zernike polynomials. This process ensures high-quality wavefront correction phases for the central region, which can then be used as accurate initial guesses for optimizing neighboring regions.
  • the purpose of outward progression is to systematically optimize image quality across all zones by using the correction phases of the brightest zone as initial guesses for surrounding areas. This iterative process ensures that wavefront distortions are minimized throughout the entire image, leading to high-resolution, high-quality imaging.
  • outward progression involves using the correction phases of the brightest zone to optimize the surrounding zones and subsequently applying these optimized phases to further zones.
  • a second Singular Value Decomposition SVD
  • the second SVD focuses solely on removing the multiple scattering background, requiring more singular values to be retained.
  • the correction phases of the brightest zone are applied to the new reflection matrix after the second SVD, providing good initial guesses for the regions around the brightest zone, which only need minor improvements.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Health & Medical Sciences (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Optics & Photonics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

Un système de tomographie à matrice de diffusion multi-spectrale avec acquisition hyperspectrale d'instantané comprend un sous-système de microscope à interférence avec une source de lumière à large bande générant un signal d'image d'interférence à large bande combiné à partir de motifs d'interférence entre un faisceau d'échantillon et un faisceau de référence. Un sous-système d'imagerie hyperspectrale acquiert lesdits motifs d'interférence pour de multiples canaux de sortie et des fréquences lumineuses en une seule prise à l'aide d'une caméra à grande vitesse. Un système informatique transforme ensuite les motifs d'interférence en coefficients de matrice de diffusion à travers les canaux de sortie et les fréquences lumineuses. L'invention concerne également des systèmes mettant en œuvre des procédés de correction d'image.
PCT/US2024/034022 2023-06-14 2024-06-14 Tomographie à matrice de diffusion avec acquisition hyperspectrale d'instantané Pending WO2024259243A2 (fr)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US202363472900P 2023-06-14 2023-06-14
US63/472,900 2023-06-14
US202463549045P 2024-02-02 2024-02-02
US63/549,045 2024-02-02

Publications (2)

Publication Number Publication Date
WO2024259243A2 true WO2024259243A2 (fr) 2024-12-19
WO2024259243A3 WO2024259243A3 (fr) 2025-03-27

Family

ID=93852741

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2024/034022 Pending WO2024259243A2 (fr) 2023-06-14 2024-06-14 Tomographie à matrice de diffusion avec acquisition hyperspectrale d'instantané

Country Status (1)

Country Link
WO (1) WO2024259243A2 (fr)

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9442015B2 (en) * 2010-09-03 2016-09-13 The Arizona Board Of Regents On Behalf Of The University Of Arizona Snapshot spatial heterodyne imaging polarimetry
KR20140096262A (ko) * 2011-08-21 2014-08-05 데이비드 레비츠 스마트폰에 장착된 광 간섭 단층 촬영 시스템
US11896303B2 (en) * 2017-07-14 2024-02-13 Wavesense Engineering Gmbh Optical apparatus
US10219700B1 (en) * 2017-12-15 2019-03-05 Hi Llc Systems and methods for quasi-ballistic photon optical coherence tomography in diffusive scattering media using a lock-in camera detector

Also Published As

Publication number Publication date
WO2024259243A3 (fr) 2025-03-27

Similar Documents

Publication Publication Date Title
US11422503B2 (en) Device and method for iterative phase recovery based on pixel super-resolved on-chip holography
US12237094B2 (en) Fourier ptychographic imaging systems, devices, and methods
US12052518B2 (en) Multi-modal computational imaging via metasurfaces
JP5721195B2 (ja) 光学特性測定装置及び光学特性測定方法
US7336353B2 (en) Coding and modulation for hyperspectral imaging
CN106164784B (zh) 数字全息设备
KR102271053B1 (ko) 수차를 야기하는 샘플 내의 타겟 오브젝트를 이미징하기 위한 초점 스캔 방식의 이미징 장치
US11368608B2 (en) Compressed sensing based object imaging system and imaging method therefor
KR20210048951A (ko) 초분광 이미지 센서 및 이를 포함하는 초분광 촬상 장치
JP6192017B2 (ja) デジタルホログラフィ装置
KR101916577B1 (ko) 산란과 수차를 동시에 야기하는 매질 내의 타겟 오브젝트를 이미징하는 방법
CN1688944A (zh) 离轴照明直接数字全息术
WO2019025759A1 (fr) Dispositif d'imagerie spectrale à ouverture codée
US12360039B2 (en) Multi-spectral scattering-matrix tomography
WO2024259243A2 (fr) Tomographie à matrice de diffusion avec acquisition hyperspectrale d'instantané
US11012643B2 (en) System and method for spectral imaging
JP6984736B2 (ja) 撮像装置及び撮像方法
Hagen et al. Using polarization cameras for snapshot imaging of phase, depth, and spectrum
Petrov et al. Terahertz multiple-plane phase retrieval
WO2021159084A1 (fr) Nappe de lumière en réseau et holographie de fresnel à corrélation incohérente
US12446780B2 (en) Spectral domain-optical nonlinearity tomography device
JP7591424B2 (ja) インコヒーレントホログラム撮像装置
Samanta et al. Improving resolution in two orthogonal orientations from a single-shot digital holographic microscopy
KR101332984B1 (ko) 홀로그램 촬영 장치 및 이를 이용한 홀로그램 촬영 방법
Shevkunov et al. CNN-assisted quantitative phase microscopy for biological cell imaging.