Improvement of transcutaneous fluorescent images with a depth-dependent point-spread function

Spatial Resolution for X-ray Excited Luminescence Chemical Imaging (XELCI)

Measuring chemical concentrations at the surface of implanted medical devices is important for elucidating the local biochemical environment, especially during implant infection. Although chemical indicator dyes enable chemical measurements in vitro, they are usually ineffective when measuring through tissue because the background obscures the dye signal and scattering dramatically reduces the spatial resolution. X-ray excited luminescent chemical imaging (XELCI) is a recent imaging modality which overcomes these limitations using a focused X-ray beam to excite a small spot of red light on scintillator-coated medical implants with well-defined location (because X-rays are minimally scattered) and low background. A spectrochemical indicator film placed over the scintillator layer, e.g., a polymer film containing pH-indicator dyes, absorbs some of the luminescence according to the local chemical environment, and this absorption is then detected by measuring the light intensity/spectrum passing through the tissue. A focused X-ray beam is used to scan point-by-point with a spatial resolution mainly limited by the X-ray beam width with minimum increase from X-ray absorption and scattering in the tissue. X-ray resolution, implant surface specificity, and chemical sensitivity are the three key features of XELCI. Here, we study spatial resolution using optically absorptive targets. For imaging a series of lines, the 20-80% knife-edge resolution was ∼285 (±15) μm with no tissue and 475 ± 18 and 520 ± 34 μm, respectively, through 5 and 10 mm thick tissue. Thus, doubling the tissue depth did not appreciably change the spatial resolution recorded through the tissue. This shows the promise of XELCI for submillimeter chemical imaging through tissue.

Imaging through thick scattering media based on envelope-informed learning with a simulated training dataset

Computational imaging faces significant challenges in dealing with multiple scattering through thick complex media. While deep learning has addressed some ill-posed problems in scattering imaging, its practical application is limited by the acquisition of the training dataset. In this study, the Gaussian-distributed envelope of the speckle image is employed to simulate the point spread function (PSF), and the training dataset is obtained by the convolution of the handwritten digits with the PSF. This approach reduces the requirement of time and conditions for constructing the training dataset and enables a neural network trained on this dataset to reconstruct objects obscured by an unknown scattering medium in real experiments. The quality of reconstructed objects is negatively correlated with the thickness of the scattering medium. Our proposed method provides a new way, to the best of our knowledge, to apply deep learning in scattering imaging by reducing the time needed for constructing the training dataset.

Imaging objects hidden inside the strongly scattering media based on bidirectional ghost imaging

We demonstrate a novel, to the best of our knowledge, method for imaging objects hidden inside the strongly scattering media based on bidirectional ghost imaging (GI). In this method, GI is performed separately on both sides of the object, resulting in two GI results. Through an autocorrelation operation to the two GI results, the convolution between the autocorrelation of the object and the point spread function (PSF) of the strongly scattering media can be recovered. Therefore, the object can be recovered by obtaining the PSF of the strongly scattering media through noninvasive measurement or numerical calculation. Simulation and experimental results show that bidirectional ghost imaging (BGI) can reconstruct high-quality images, particularly when the thickness of the strongly scattering media greatly exceeds the scattering mean free path.

Near-Infrared Transillumination for Macroscopic Functional Imaging of Animal Bodies

The classical transillumination technique has been revitalized through recent advancements in optical technology, enhancing its applicability in the realm of biomedical research. With a new perspective on near-axis scattered light, we have harnessed near-infrared (NIR) light to visualize intricate internal light-absorbing structures within animal bodies. By leveraging the principle of differentiation, we have extended the applicability of the Beer-Lambert law even in cases of scattering-dominant media, such as animal body tissues. This approach facilitates the visualization of dynamic physiological changes occurring within animal bodies, thereby enabling noninvasive, real-time imaging of macroscopic functionality in vivo. An important challenge inherent to transillumination imaging lies in the image blur caused by pronounced light scattering within body tissues. By extracting near-axis scattered components from the predominant diffusely scattered light, we have achieved cross-sectional imaging of animal bodies. Furthermore, we have introduced software-based techniques encompassing deconvolution using the point spread function and the application of deep learning principles to counteract the scattering effect. Finally, transillumination imaging has been elevated from two-dimensional to three-dimensional imaging. The effectiveness and applicability of these proposed techniques have been validated through comprehensive simulations and experiments involving human and animal subjects. As demonstrated through these studies, transillumination imaging coupled with emerging technologies offers a promising avenue for future biomedical applications.

Reconstructing a Deblurred 3D Structure in a Turbid Medium from a Single Blurred 2D Image—For Near-Infrared Transillumination Imaging of a Human Body

To provide another modality for three-dimensional (3D) medical imaging, new techniques were developed to reconstruct a 3D structure in a turbid medium from a single blurred 2D image obtained using near-infrared transillumination imaging. One technique uses 1D information of a curvilinear absorber, or the intensity profile across the absorber image. Profiles in different conditions are calculated by convolution with the depth-dependent point spread function (PSF) of the transillumination image. In databanks, profiles are stored as lookup tables to connect the contrast and spread of the profile to the absorber depth. One-to-one correspondence from the contrast and spread to the absorber depth and thickness were newly found. Another technique uses 2D information of the transillumination image of a volumetric absorber. A blurred 2D image is deconvolved with the depth-dependent PSF, thereby producing many images with points of focus on different parts. The depth of the image part can be estimated by searching the deconvolved images for the image part in the best focus. To suppress difficulties of high-spatial-frequency noise, we applied a noise-robust focus stacking method. Experimentation verified the feasibility of the proposed techniques, and suggested their applicability to curvilinear and volumetric absorbers such as blood vessel networks and cancerous lesions in tissues.

Three-dimensional imaging through turbid media using deep learning: NIR transillumination imaging of animal bodies

Using near-infrared (NIR) light with 700-1200 nm wavelength, transillumination images of small animals and thin parts of a human body such as a hand or foot can be obtained. They are two-dimensional (2D) images of internal absorbing structures in a turbid medium. A three-dimensional (3D) see-through image is obtainable if one can identify the depth of each part of the structure in the 2D image. Nevertheless, the obtained transillumination images are blurred severely because of the strong scattering in the turbid medium. Moreover, ascertaining the structure depth from a 2D transillumination image is difficult. To overcome these shortcomings, we have developed a new technique using deep learning principles. A fully convolutional network (FCN) was trained with 5,000 training pairs of clear and blurred images. Also, a convolutional neural network (CNN) was trained with 42,000 training pairs of blurred images and corresponding depths in a turbid medium. Numerous training images were provided by the convolution with a point spread function derived from diffusion approximation to the radiative transport equation. The validity of the proposed technique was confirmed through simulation. Experiments demonstrated its applicability. This technique can provide a new tool for the NIR imaging of animal bodies and biometric authentication of a human body.

Conformal Coating of Orthopedic Plates with X-ray Scintillators and pH Indicators for X-ray Excited Luminescence Chemical Imaging through Tissue

We describe a pH-indicating material that can be directly implanted or coated on orthopedic implant surfaces to provide high-spatial-resolution pH mapping through tissue by X-ray excited luminescence chemical imaging (XELCI). This is especially useful for detecting local pH changes during treatment of implant-associated infections. The material has two layers: an X-ray scintillator layer with Gd₂O₂S:Eu in epoxy, which emits 620 and 700 nm light when irradiated with X-rays, and a pH indicator dye layer, which absorbs some of the 620 nm light in a pH-dependent fashion. To acquire each pixel in the image, a focused X-ray beam irradiates a small region of scintillators and the ratio of 620 to 700 nm light is acquired through the tissue. Scanning the X-ray beam across the implant surface generates high-spatial-resolution chemical measurements. Two associated challenges are (1) to make robust sensors that can be implanted in tissue to measure local chemical concentrations specifically for metal orthopedic implants and (2) to conformally coat the implant surface with scintillators and pH indicator dyes in order to make measurements over a large area. Previously, we have physically pressed or glued a pH-sensitive hydrogel sensor onto the surface of an implant, but this is impractical for imaging over large irregular device areas such as an orthopedic plate with holes and edges. Herein, we describe a chemically sensitive and biocompatible XELCI sensor material that can conformally coat the implant surface. A two-part commercial-grade epoxy resin was mixed with Gd₂O₂S:Eu and adhered to the titanium surface. Sugar and salt particles were added to the surface of the epoxy as it cured to create a roughened surface and increase the surface area. On this roughened surface, a secondary layer of diacrylated polyethylene glycol (PEG) hydrogel, containing a pH sensitive dye, was polymerized. This combination of epoxy-PEG layers was found to adhere well to the metal implant unlike other previously tested polymer surfaces, which delaminated when exposed to water or humidity. The focused X-ray beam enabled 0.5 mm spatial resolution through 1 cm-thick tissue. The pH sensor-coated orthopedic plate was imaged with XELCI, through tissue, with different pH levels to acquire a calibration curve. The plates were also imaged through tissue, with a low pH region on one section due to growth of a Staphylococcus aureus biofilm. A pH sensor-coated stainless-steel rod with two distinct pH regions was inserted in a rabbit tibia specimen, and the pH was imaged through both bone and soft tissue. These studies demonstrate the use of pH sensor-coated orthopedic plates and rods for mapping the local pH through tissue during biofilm formation by XELCI.

Noninvasively Imaging pH at the Surface of Implanted Orthopedic Devices with X-ray Excited Luminescence Chemical Imaging

Implanted medical device-associated infections are a leading cause of fixation failure, and early diagnosis is the key to successful treatment. During infection, acidosis near the implant plays a role in antibiotic resistance and low pH is a potential infection indicator. Herein, we describe a pH sensor which attaches to the implants to noninvasively image local pH with high spatial resolution. The sensor has two layers: a scintillator layer which emits 620 and 700 nm light upon X-ray irradiation and a pH indicator layer containing bromocresol green dye that absorbs 620 nm luminescence in neutral/basic pH and passes 700 nm light at all pHs. We also developed a dedicated imaging system capable of scanning relatively large specimens through thick tissues. A focused X-ray beam irradiates one spot on the sensor, and the 620 to 700 nm peak ratio is measured to determine the local pH; images are acquired by scanning the X-ray beam across the surface and measuring the pH point-by-point. The sensor was covered with varying thickness slices of chicken breast tissue (0-19 mm) to evaluate how the tissue affects the peak intensity and ratio. Thick tissues attenuated both 620 and 700 nm light, with more attenuation at 620 nm than 700 nm. Although this spectral distortion shifted the pH calibration curve, the effect could be corrected for using a scintillator film region with no pH indicator layer as a spectral reference. The sensor was attached to an orthopedic plate affixed to a human cadaveric tibia and imaged through tissue. This approach provides both high spatial resolution from focused X-ray excitation and surface chemical specificity from the indicator dye, providing a tool for imaging local pH through tissue.

Bright X-ray and up-conversion nanophosphors annealed using encapsulated sintering agents for bioimaging applications

Nanophosphors are promising contrast agents for deep tissue optical imaging applications because they can be excited by X-ray and near infrared light that penetrates deeply through tissue and generates almost no autofluorescence background in the tissue. For these bioimaging applications, the nanophosophors should ideally be small, monodispersed and brightly luminescent. However, most methods used to improve luminescence yield by annealing the particles to reduce crystal and surface defects (e.g. using flux or sintering agents) also cause particle fusion or require multiple component core-shell structures. Here, we report a novel method to prepare bright, uniformly sized X-ray nanophosphors (Gd2O2S:Eu or Tb) and upconversion nanophosphors (Y2O2S: Yb/Er, or Yb/Tm) with large crystal domain size without causing aggregation. A core-shell nanoparticle is formed, with NaF only in the core. We observe that increasing the NaF sintering agent concentration up to 7.6 mol% increases both crystal domain size and luminescence intensity (up to 40% of commercial microphosphors) without affecting the physical particticle diameter. Above 7.6 mol%, particle fusion is observed. The annealing is insensitive to the cation (Na+ or K+) but varies strongly with anion, with F->Cl->CO32->Br->I-. The luminescence depends strongly on crystal domain size. The data agree reasonably well with a simple domain surface quenching model, although the size-dependence suggests additional quenching mechanisms within small domains. The prepared bright nanophosphors were subsequently functionalized with PEG-folic acid to target MCF-7 breast cancer cells which overexpress folic acid receptors. Both X-ray and upconversion nanophosphors provided low background and bright luminescence which was imaged through 1 cm chicken breast tissue at a low dose of nanophosphors 200 µL (0.1 mg/mL). We anticipate these highly monodispersed and bright X-ray and upconversion nanophosphors will have significant potential for tumor targeted imaging.

Recovering Inner Slices of Layered Translucent Objects by Multi-Frequency Illumination

This paper describes a method for recovering appearance of inner slices of translucent objects. The appearance of a layered translucent object is the summed appearance of all layers, where each layer is blurred by a depth-dependent point spread function (PSF). By exploiting the difference of low-pass characteristics of depth-dependent PSFs, we develop a multi-frequency illumination method for obtaining the appearance of individual inner slices. Specifically, by observing the target object with varying the spatial frequency of checker-pattern illumination, our method recovers the appearance of inner slices via computation. We study the effect of non-uniform transmission due to inhomogeneity of translucent objects and develop a method for recovering clear inner slices based on the pixel-wise PSF estimates under the assumption of spatial smoothness of inner slice appearances. We quantitatively evaluate the accuracy of the proposed method by simulations and qualitatively show faithful recovery using real-world scenes.

X-ray excited luminescent chemical imaging (XELCI) for non-invasive imaging of implant infections

X-ray excited luminescent chemical imaging (XELCI) uses a combination of X-ray excitation to provide high resolution and optical detection to provide chemical sensing. A key application is to detect and study implant-associated infection. The implant is coated with a layer of X-ray scintillators which generate visible near infrared light when irradiated with an X-ray beam. This light first passes through a pH indicator dye-loaded film placed over the scintillator film in order to modulate the luminescence spectrum according to pH. The light then passes through tissue is collected and the spectral ratio measured to determine pH. A focused X-ray beam irradiates a point in the scintillator film, and a pH image is formed point-by-point by scanning the beam across the sample. The sensor and scanning system are described along with preliminary results showing images in rabbit models.

Method for In-Vivo Fluorescence Imaging Contrast Enhancement through Light Modulation

Early diagnosis is one of the most important factors that increase the therapeutic potential of the disease. Diagnoses conducted by conventional equipment are expensive, time-consuming, burdensome to patients, and do not have high success rates. Diagnostic methods have also been investigated using nanoparticles. However, there have been no significant improvements in the early diagnosis of disease. The diagnosis technique proposed in this paper consumes less time, is more cost-effective, and more accurate. It uses a new concept-a low-intensity fluorescence molecular imaging system with a lock-in technique. This study applied the lock-in technique to basic research in contrast enhancement and optimization. This improved fluorescence distribution analysis, resulting in increased resolution of optical molecular imaging for early diagnosis of disease. An experimental lock-in fluorescence imaging system, which used a variety of fluorescent dyes, achieved signal amplification 100 times greater than that of a conventional fluorescence imaging system. The results of this study demonstrate that the lock-in technique could significantly improve optical molecular imaging technology, making it possible to achieve early diagnosis of disease.

Detecting De-gelation through Tissue Using Magnetically Modulated Optical Nanoprobes (MagMOONs)

Alginate gels are widely used for drug delivery and implanted devices. The rate at which these gels break down is important for controlling drug release. Since the de-gelation may be different in vivo, monitoring this process in situ is essential. However, it is challenging to monitor the gel through tissue due to optical scattering and tissue autofluorescence. Herein we describe a method to detect through tissue the chemically-induced changes in viscosity and de-gelation process of alginate gels using magnetically modulated optical nanoprobes (MagMOONs). The MagMOONs are fluorescent magnetic microspheres coated with a thin layer of opaque metal on one hemisphere. The metal layer prevents excitation and emission light from passing through one side of the MagMOONs, which creates orientation-dependent fluorescence intensity. The magnetic particles also align in an external magnetic field and give blinking signals when they rotate to follow an external modulated magnetic field. The blinking signals from these MagMOONs are distinguished from background autofluorescence and can be tracked on a single particle level in the absence of tissue, or for an ensemble average of particles blinking through tissue. When these MagMOONs are dispersed in calcium alginate gel, they become sensors for detecting gel degradation upon addition of either ammonium ion or alginate lyase. Our results show MagMOONs start blinking approximately 10 minutes after 2 mg/mL alginate lyase addition and this blinking is clearly detected even through up to 4 mm chicken breast. This approach can potentially be employed to detect bacterial biofilm formation on medical implants by sensing specific proteases that either activate a related function or regulate biofilm formation. It can also be applied to other biosensors and drug delivery systems based on enzyme-catalyzed breakdown of gel components.

Multifunctional yolk-in-shell nanoparticles for pH-triggered drug release and imaging

Multifunctional nanoparticles are synthesized for both pH-triggered drug release and imaging with radioluminescence, upconversion luminescent, and magnetic resonance imaging (MRI). The particles have a yolk-in-shell morphology, with a radioluminescent core, an upconverting shell, and a hollow region between the core and shell for loading drugs. They are synthesized by controlled encapsulation of a radioluminescent nanophosphor yolk in a silica shell, partial etching of the yolk in acid, and encapsulation of the silica with an upconverting luminescent shell. Metroxantrone, a chemotherapy drug, was loaded into the hollow space between X-ray phosphor yolk and up-conversion phosphor shell through pores in the shell. To encapsulate the drug and control the release rate, the nanoparticles are coated with pH-responsive biocompatible polyelectrolyte layers of charged hyaluronic acid sodium salt and chitosan. The nanophosphors display bright luminescence under X-ray, blue light (480 nm), and near infrared light (980 nm). They also served as T1 and T2 MRI contrast agents with relaxivities of 3.5 mM(-1) s(-1) (r1 ) and 64 mM(-1) s(-1) (r2 ). These multifunctional nanocapsules have applications in controlled drug delivery and multimodal imaging.

Three-dimensional transillumination image reconstruction for small animal with new scattering suppression technique

To realize three-dimensional (3D) optical imaging of the internal structure of an animal body, we have developed a new technique to reconstruct optical computed tomography (optical CT) images from two-dimensional (2D) transillumination images. In transillumination imaging of an animal body using near-infrared light, the image is blurred because of the strong scattering in the tissue. To overcome this problem, we propose a novel technique to apply the point spread function (PSF) for a light source located inside the medium to the transilluminated image of light-absorbing structure. The problem of the depth-dependence of PSF was solved in the calculation of the projection image in the filtered back-projection method. The effectiveness of the proposed technique was assessed in the experiments with a model phantom and a mouse. These analyses verified the feasibility of the practical 3D imaging of the internal light-absorbing structure of a small animal.

Application of fluorescent PSF for 3D reconstruction of absorbing structure using slab transillumination images

In transillumination imaging of an animal body using near-infrared light, the image is blurred due to the strong scattering in the tissue. We have devised the depth-dependent point spread function (PSF) to suppress the scattering effect in fluorescent imaging. In this study, we applied this principle and developed a technique to reconstruct the absorbing structure in turbid medium without using fluorescent material. In experiments, the feasibility and effectiveness of the proposed technique were verified.

Depth estimation and image improvement of fluorescent objects in scattering medium with unknown optical parameters

The depth of fluorescent source is required to suppress the scattering effect in transcutaneous fluorescent imaging. We have developed a technique to estimate the depth using two light sources. However, with this technique, the reduced scattering coefficient (μs') and the absorption coefficient (μa) of the scattering medium have to be known a priori. To eliminate this requirement, we devised a technique that uses three excitation sources. The feasibility of the proposed technique was verified in both computer simulation and experiment. It was demonstrated that we could estimate the depth of the fluorescent source and improve the fluorescent image even with μs' and μa variation.

CW transillumination imaging by extracting weakly scattered light from strongly diffused light

Transmitted light through a diffuse scattering medium includes strongly diffused light (SDL) and weakly scattered light (WSL). To realize clear transillumination imaging through thick body tissue, which is typically more than 10 mm, we developed a technique to extract the WSL component from diffused light. In experiments using a 15-mm-thick scattering medium (mu(s)' = 1.0/mm), the cross-section of the light propagation area at the center of the medium was confined to a 50% area. This method's usefulness was demonstrated by transillumination imaging through a 40-mm-thick piece of chicken meat. The possibility of depth evaluation was also verified.

Diameter measurement from images of fluorescent cylinders embedded in tissue

Absorption and scattering of light by tissue as well as limitations in the resolution of the optical system influence the appearance of tissue embedded objects in fluorescence images and may reduce the accuracy of measurements from these images. Although the principles of light scattering in tissue and optical resolution are well known, the interplay between the two in fluorescence imaging in an imaging cryomicrotome is not well understood. In this paper we present and investigate an image formation model in a reflection geometry, like an imaging cryomicrotome, that takes both light scattering by tissue as well as the point spread function of the lens into account. The validity of the model was investigated by comparison of diameter estimates of fluorescent cylinders as obtained from images acquired in a reflection geometry with those estimated from simulated images. The results reveal that diameter values estimated from the simulated images are in excellent agreement with the experimental estimates. Our approach in modeling the image formation process of embedded fluorescent structures allows for the prediction of accuracy of quantitative estimates from fluorescence images. The relationship between imaging parameters and bias can be applied to arrive at accurate diameter estimates of near cylindrical structures like blood vessels.

The power of single and multibeam two-photon microscopy for high-resolution and high-speed deep tissue and intravital imaging

Two-photon microscopy is indispensable for deep tissue and intravital imaging. However, current technology based on single-beam point scanning has reached sensitivity and speed limits because higher performance requires higher laser power leading to sample degradation. We utilize a multifocal scanhead splitting a laser beam into a line of 64 foci, allowing sample illumination in real time at full laser power. This technology requires charge-coupled device field detection in contrast to conventional detection by photomultipliers. A comparison of the optical performance of both setups shows functional equivalence in every measurable parameter down to penetration depths of 200 microm, where most actual experiments are executed. The advantage of photomultiplier detection materializes at imaging depths >300 microm because of their better signal/noise ratio, whereas only charge-coupled devices allow real-time detection of rapid processes (here blood flow). We also find that the point-spread function of both devices strongly depends on tissue constitution and penetration depth. However, employment of a depth-corrected point-spread function allows three-dimensional deconvolution of deep-tissue data up to an image quality resembling surface detection.