Large field-of-view non-invasive imaging through scattering layers using fluctuating random illumination

Abstract

Non-invasive optical imaging techniques are essential diagnostic tools in many fields. Although various recent methods have been proposed to utilize and control light in multiple scattering media, non-invasive optical imaging through and inside scattering layers across a large field of view remains elusive due to the physical limits set by the optical memory effect, especially without wavefront shaping techniques. Here, we demonstrate an approach that enables non-invasive fluorescence imaging behind scattering layers with field-of-views extending well beyond the optical memory effect. The method consists in demixing the speckle patterns emitted by a fluorescent object under variable unknown random illumination, using matrix factorization and a novel fingerprint-based reconstruction. Experimental validation shows the efficiency and robustness of the method with various fluorescent samples, covering a field of view up to three times the optical memory effect range. Our non-invasive imaging technique is simple, neither requires a spatial light modulator nor a guide star, and can be generalized to a wide range of incoherent contrast mechanisms and illumination schemes.

Fig. 1. Schematic of the experimental setup and reconstruction principle.

a Schematic view of experimental setup. A coherent light source illuminates a rotating diffuser in order to excite the fluorescent object through a scattering medium with a random modulated speckle pattern. Once excited, the emitted signal from the fluorescent objects is recorded with a camera. I_fluo is a series of t fluorescent speckles corresponding to different random speckle illuminations. The fingerprints can be recovered from I_fluo by using NMF. Fingerprint-based reconstruction. b Pairwise deconvolution (labeled as ⊛⁻¹) between all the possible pairs of emitter fingerprints is performed. c The result of each deconvolution provides the relative position between one emitter and its neighbors. d By adding the resulting images for each emitter, it is possible to recover a partial image of the object centered at that emitter (see Eq. (4)). e All the partial images can be merged into the final reconstruction according to the relative position between neighboring emitters. Dashed circle indicates the optical memory range. Scale bar sizes are 10 μm. RD: rotating diffuser, DM: dichroic mirror, OB: objective, Scat.: scattering medium, Fluo. Obj.: fluorescent object, SF: spectral filter, TL: tube lens.

Fig. 2. Experimental results of imaging through a scattering medium with fluorescent beads.

a, b Fluorescent images of beads recorded without scattering medium. c, d Reconstruction of the object using NMF + FBR approach. The estimated rank of NMF is ρ = 26 for (c) and ρ = 16 for (d). In both cases, t = 5120 fluorescent speckle patterns are captured. The exposure time of c, d is set to 15 ms and 20 ms, respectively. Dashed circles indicate the optical memory effect range.

Fig. 3. Experimental results of imaging through scattering media with continuous objects.

Fluorescent images of different pollen seed structures (a, b) and different cellulose fiber structures (c, d) recorded without scattering medium. e–h Reconstruction of the objects with the NMF + FBR approach. The estimated rank for the NMF is ρ = 68 for (e), ρ = 85 for (f), ρ = 45 for (g), and ρ = 55 for (h), respectively. In both cases, t = 5120 fluorescent speckle patterns are recorded with an exposure time of 10 ms. Dashed circle indicates the optical memory effect range.