Recent advances in optical imaging through deep tissue: imaging probes and techniques

Abstract

Optical imaging has been essential for scientific observations to date, however its biomedical applications has been restricted due to its poor penetration through tissues. In living tissue, signal attenuation and limited imaging depth caused by the wave distortion occur because of scattering and absorption of light by various molecules including hemoglobin, pigments, and water. To overcome this, methodologies have been proposed in the various fields, which can be mainly categorized into two stategies: developing new imaging probes and optical techniques. For example, imaging probes with long wavelength like NIR-II region are advantageous in tissue penetration. Bioluminescence and chemiluminescence can generate light without excitation, minimizing background signals. Afterglow imaging also has high a signal-to-background ratio because excitation light is off during imaging. Methodologies of adaptive optics (AO) and studies of complex media have been established and have produced various techniques such as direct wavefront sensing to rapidly measure and correct the wave distortion and indirect wavefront sensing involving modal and zonal methods to correct complex aberrations. Matrix-based approaches have been used to correct the high-order optical modes by numerical post-processing without any hardware feedback. These newly developed imaging probes and optical techniques enable successful optical imaging through deep tissue. In this review, we discuss recent advances for multi-scale optical imaging within deep tissue, which can provide reseachers multi-disciplinary understanding and broad perspectives in diverse fields including biophotonics for the purpose of translational medicine and convergence science. Methodologies for multi-scale optical imaging within deep tissues are discussed in diverse fields including biophotonics for the purpose of translational medicine and convergence science. Recent imaging probes have tried deep tissue imaging by NIR-II imaging, bioluminescence, chemiluminescence, and afterglow imaging. Optical techniques including direct/indirect and coherence-gated wavefront sensing also can increase imaging depth.

Keywords: Adaptive optics; Bioluminescence; Chemiluminescence; Deep tissue imaging; Imaging probe; NIR-II; Optical imaging; Wavefront sensing; Wavefront shaping.

Scheme 1

Schematic illustration of deep tissue imaging strategies using imaging probes and optical techniques

Fig. 1

NIR-II fluorescent imaging probe for immune cell-mediated tumor imaging. a Schematic illustration of the strategy for tumor accumulation and imaging. b NIR-II imaging in tumor tissue using mice. Adapted with permission from Ref. [11]

Fig. 2

NIR-II phosphorescent probe for tumor imaging with superior penetration depth. a Schematic illustration of the assembled nanotubes. b Imaging signal attenuation test in tumor-bearing mice. Adapted with permission from Ref. [12]

Fig. 3

Bioluminescent imaging probe for tracking of the injected nanoparticles in cancer tissue. a Schematic illustration of the ferritin nanocage conjugated with luciferin. b Time-dependent bioluminescence signals in mice after intravenous injection of the probe. Adapted with permission from Ref. [15]

Fig. 4

Chemiluminescent imaging probe for diagnosis of acute kidney injury. a Schematic illustration of the chemiluminescence probe with three components. b Chemiluminescence/NIRF dual imaging of acute kidney injury in mice. Adapted with permission from Ref. [17]

Fig. 5

Afterglow imaging probe for in vivo imaging of drug-induced hepatotoxicity. a Schematic illustration of the persistent luminescence nanoparticle and luminescence signal data. b Persistent luminescence imaging in liver of the mice treated with acetaminophen. Adapted with permission from Ref. [20]

Fig. 6

Diagrams of AO techniques. a Direct wavefront sensing method using a wavefront sensor to directly measure the wavefront. b Indirect wavefront sensing method to optimize the image-based metric

Fig. 7

Direct wavefront sensing with focus scanning microscopy. a The three-dimensional volumetric images of oligodendrocytes (magenta) and neuronal nuclei (green) in a zebrafish brain in vivo. Adapted with permission from Ref. [33]. b Thy1-YFPH neurons at 600–620 μm in a mouse brain in vivo. Adapted with permission from Ref. [34]. c-e mRuby2-labeled layer 5 pyramidal neurons in a 150 × 150 × 810 μm³ of vS1 mouse brain cortex in vivo. Adapted with permission from Ref. [35]. Scale bar in b, 20 μm. Scale bar in d, 10 μm. Scale bar in e, 2 μm

Fig. 8

Direct wavefront sensing with LLSM and SIM. a A live human stem cell-derived organoid with four steps of corrections. Adapted with permission from Ref. [36]. b The eye of a zebrafish embryo 24 hpf. The separated cells are indicated in color. Adapted with permission from Ref. [36]. c Cortical neurons of a Thy1-GFP mouse brain in vivo. Adapted with permission from Ref. [37]. d C. elegans in vivo images expressed by the adherens junction marker ajm-1::GFP. Adapted with permission from Ref. [38]. Scale bar in a, 5 μm. Scale bar in b, 30 μm. Scale bar in c, 10 μm; inset, 3 μm

Fig. 9

Indirect wavefront sensing with focus scanning microscopy. a The visual cortex and hippocampus at the depth-of-field 700–750 μm, 860–900 μm, and 1000–1200 μm for a EGFP–Thy1(M) mouse brain in vivo. Adapted with permission from Ref. [39]. b The synaptic structures in the deep cortical region of a mouse (Thy1-GFP-M) brain. Adapted with permission from Ref. [40]. c High and basal dendritic spines of a mouse V1 neuron in vivo. Adapted with permission from Ref. [41]. Scale bar in a, 20 μm. Scale bar in b is 10 μm

Fig. 10

Indirect wavefront sensing with SIM and holographic method. a GFP expressed microtubule of Drosophila larval macrophage by SIM and IsoSense correction. Adapted with permission from Ref. [42]. b GFP-expressed pyramidal neuron below 630 μm in a living mouse brain measured by conventional and F-SHARP. Adapted with permission from Ref. [43]. c The single microglia cell from 530 μm thick hippocampus tissue of a mouse brain measured by conventional and DASH. Adapted with permission from Ref. [44]. Scale bar in b, 5 μm

Fig. 11

AO correction based on CGWS. a Schematics of low-coherence interferometer for CGWS. Adapted with permission from Ref. [77]. b CGWS-based AO imaging of the developing olfactory bulb in transgenic zebrafish larvae. Adapted with permission from Ref. [45]. Wavefront aberration map as measured by CGWS (top-right). Two-photon fluorescence images at a depth of 50 μm before aberration correction (bottom-left) and after aberration correction (bottom-right)

Fig. 12

CLASS microscopy for imaging the hyphae of Aspergillus cells in a rabbit’s cornea. a Layout of imaging geometry. b,c Incoherent addition of time-gated reflection images. d-i, Same as (b,c) respectively, after aberration correction by CLASS algorithm. f,g Identified sample-induced wavefront aberrations for input and output directions, respectively. Adapted with permission from Ref. [46]

Fig. 13

High-speed time-gated reflection matrix approaches for in vivo imaging. a Label-free volumetric images of the hindbrain of a 10-dpf larval zebrafish taken by AO-SASM. Adapted with permission from Ref [47]. b In vivo through-skull mouse brain imaging with LS-RMM. Adapted with permission from Ref. [48]. OCM: in vivo through-skull optical coherence microscopy image. LS-RMM: LS-RMM image corresponding to OCM image. Before HC: ex vivo through-skull two-photon images showing neuronal dendrites in the brain of a Thy1-EGFP line M mouse before hardware wavefront correction. After HC: two-photon images after hardware wavefront correction, corresponding to before HC images. Imaging depths: z₁ = 113 ± 1.5 μm, z₂ = 122 ± 1.5 μm. Adapted with permission from Ref. [48]