Accuracy of fluorescent tomography in the presence of heterogeneities: study of the normalized Born ratio

AAPM Task Group Report 311: Guidance for performance evaluation of fluorescence-guided surgery systems

The last decade has seen a large growth in fluorescence-guided surgery (FGS) imaging and interventions. With the increasing number of clinical specialties implementing FGS, the range of systems with radically different physical designs, image processing approaches, and performance requirements is expanding. This variety of systems makes it nearly impossible to specify uniform performance goals, yet at the same time, utilization of different devices in new clinical procedures and trials indicates some need for common knowledge bases and a quality assessment paradigm to ensure that effective translation and use occurs. It is feasible to identify key fundamental image quality characteristics and corresponding objective test methods that should be determined such that there are consistent conventions across a variety of FGS devices. This report outlines test methods, tissue simulating phantoms and suggested guidelines, as well as personnel needs and professional knowledge bases that can be established. This report frames the issues with guidance and feedback from related societies and agencies having vested interest in the outcome, coming from an independent scientific group formed from academics and international federal agencies for the establishment of these professional guidelines.

Deep background-mismodeling-learned reconstruction for high-accuracy fluorescence diffuse optical tomography

We present a deep background-mismodeling-learned reconstruction framework for high-accuracy fluorescence diffuse optical tomography (FDOT). A learnable regularizer incorporating background mismodeling is formulated in the form of certain mathematical constraints. The regularizer is then learned to obtain the background mismodeling automatically using a physics-informed deep network implicitly. Here, a deep-unrolled FIST-Net for optimizing L1-FDOT is specially designed to obtain fewer learning parameters. Experiments show that the accuracy of FDOT is significantly improved via implicitly learning the background mismodeling, which proves the validity of the deep background-mismodeling-learned reconstruction. The proposed framework can also be used as a general method to improve a class of image modalities based on linear inverse problems with unknown background modeling errors.

Selfrec-Net: self-supervised deep learning approach for the reconstruction of Cherenkov-excited luminescence scanned tomography

As an emerging imaging technique, Cherenkov-excited luminescence scanned tomography (CELST) can recover a high-resolution 3D distribution of quantum emission fields within tissue using X-ray excitation for deep penetrance. However, its reconstruction is an ill-posed and under-conditioned inverse problem because of the diffuse optical emission signal. Deep learning based image reconstruction has shown very good potential for solving these types of problems, however they suffer from a lack of ground-truth image data to confirm when used with experimental data. To overcome this, a self-supervised network cascaded by a 3D reconstruction network and the forward model, termed Selfrec-Net, was proposed to perform CELST reconstruction. Under this framework, the boundary measurements are input to the network to reconstruct the distribution of the quantum field and the predicted measurements are subsequently obtained by feeding the reconstructed result to the forward model. The network was trained by minimizing the loss between the input measurements and the predicted measurements rather than the reconstructed distributions and the corresponding ground truths. Comparative experiments were carried out on both numerical simulations and physical phantoms. For singular luminescent targets, the results demonstrate the effectiveness and robustness of the proposed network, and comparable performance can be attained to a state-of-the-art deep supervised learning algorithm, where the accuracy of the emission yield and localization of the objects was far superior to iterative reconstruction methods. Reconstruction of multiple objects is still reasonable with high localization accuracy, although with limits to the emission yield accuracy as the distribution becomes more complex. Overall though the reconstruction of Selfrec-Net provides a self-supervised way to recover the location and emission yield of molecular distributions in murine model tissues.

UHR-DeepFMT: Ultra-High Spatial Resolution Reconstruction of Fluorescence Molecular Tomography Based on 3-D Fusion Dual-Sampling Deep Neural Network

Fluorescence molecular tomography (FMT) is a promising and high sensitivity imaging modality that can reconstruct the three-dimensional (3D) distribution of interior fluorescent sources. However, the spatial resolution of FMT has encountered an insurmountable bottleneck and cannot be substantially improved, due to the simplified forward model and the severely ill-posed inverse problem. In this work, a 3D fusion dual-sampling convolutional neural network, namely UHR-DeepFMT, was proposed to achieve ultra-high spatial resolution reconstruction of FMT. Under this framework, the UHR-DeepFMT does not need to explicitly solve the FMT forward and inverse problems. Instead, it directly establishes an end-to-end mapping model to reconstruct the fluorescent sources, which can enormously eliminate the modeling errors. Besides, a novel fusion mechanism that integrates the dual-sampling strategy and the squeeze-and-excitation (SE) module is introduced into the skip connection of UHR-DeepFMT, which can significantly improve the spatial resolution by greatly alleviating the ill-posedness of the inverse problem. To evaluate the performance of UHR-DeepFMT network model, numerical simulations, physical phantom and in vivo experiments were conducted. The results demonstrated that the proposed UHR-DeepFMT can outperform the cutting-edge methods and achieve ultra-high spatial resolution reconstruction of FMT with the powerful ability to distinguish adjacent targets with a minimal edge-to-edge distance (EED) of 0.5 mm. It is assumed that this research is a significant improvement for FMT in terms of spatial resolution and overall imaging quality, which could promote the precise diagnosis and preclinical application of small animals in the future.

Mapping fluorescence resonance energy transfer parameters of a bifunctional agent using time-domain fluorescence diffuse optical tomography

We present a method to map fluorescence resonance energy transfer (FRET) parameters of a bifunctional photodynamic therapy agent, (2-[1-hexyloxyethyl]-2-devinyl pyropheophorbide-a)-cyanine dye (HPPH-CD) conjugate, which consists of a photosensitizer (HPPH) and a fluorescent agent CD. We utilized time-domain fluorescence diffuse optical tomography, the normalized Born ratio model in the Fourier-domain, and an iterative algorithm to map depth-resolved spatial heterogeneities of FRET parameters. Our results exhibited depth-resolved changes of fluorophore's lifetime and the distance maps due to FRET between HPPH and CD. Our model suggests a potential approach of using FRET parameters to monitor efficacies of multifunctional photodynamic therapy agents in deep tissue.

Quantitative bioluminescence tomography using spectral derivative data

Bioluminescence imaging (BLI) is a non-contact, optical imaging technique based on measurement of emitted light due to an internal source, which is then often directly related to cellular activity. It is widely used in pre-clinical small animal imaging studies to assess the progression of diseases such as cancer, aiding in the development of new treatments and therapies. For many applications, the quantitative assessment of accurate cellular activity and spatial distribution is desirable as it would enable direct monitoring for prognostic evaluation. This requires quantitative spatially-resolved measurements of bioluminescence source strength inside the animal to be obtained from BLI images. This is the goal of bioluminescence tomography (BLT) in which a model of light propagation through tissue is combined with an optimization algorithm to reconstruct a map of the underlying source distribution. As most models consider only the propagation of light from internal sources to the animal skin surface, an additional challenge is accounting for the light propagation from the skin to the optical detector (e.g. camera). Existing approaches typically use a model of the imaging system optics (e.g. ray-tracing, analytical optical models) or approximate corrections derived from calibration measurements. However, these approaches are typically computationally intensive or of limited accuracy. In this work, a new approach is presented in which, rather than directly using BLI images acquired at several wavelengths, the spectral derivative of that data (difference of BLI images at adjacent wavelengths) is used in BLT. As light at similar wavelengths encounters a near-identical system response (path through the optics etc.) this eliminates the need for additional corrections or system models. This approach is applied to BLT with simulated and experimental phantom data and shown that the error in reconstructed source intensity is reduced from 49% to 4%. Qualitatively, the accuracy of source localization is improved in both simulated and experimental data, as compared to reconstruction using the standard approach. The outlined algorithm can widely be adapted to all commercial systems without any further technological modifications.

Correcting for targeted and control agent signal differences in paired-agent molecular imaging of cancer cell-surface receptors

Paired-agent kinetic modeling protocols provide one means of estimating cancer cell-surface receptors with in vivo molecular imaging. The protocols employ the coadministration of a control imaging agent with one or more targeted imaging agent to account for the nonspecific uptake and retention of the targeted agent. These methods require the targeted and control agent data be converted to equivalent units of concentration, typically requiring specialized equipment and calibration, and/or complex algorithms that raise the barrier to adoption. This work evaluates a kinetic model capable of correcting for targeted and control agent signal differences. This approach was compared with an existing simplified paired-agent model (SPAM), and modified SPAM that accounts for signal differences by early time point normalization of targeted and control signals (SPAMPN). The scaling factor model (SPAMSF) outperformed both SPAM and SPAMPN in terms of accuracy and precision when the scale differences between targeted and imaging agent signals (α) were not equal to 1, and it matched the performance of SPAM for α  =  1. This model could have wide-reaching implications for quantitative cancer receptor imaging using any imaging modalities, or combinations of imaging modalities, capable of concurrent detection of at least two distinct imaging agents (e.g., SPECT, optical, and PET/MR).

Wide-field fluorescence tomography with composited epi-illumination of multi-frequency sinusoidal patterns

We present a spatial-frequency domain (SFD) fluorescence tomography (FT) for acquiring three-dimensional fluorophore distribution in turbid media. The approach uses a composited epi-illumination of multi-frequency sinusoidal patterns on a sample of semi-infinite geometry and demodulates the measured data with a generalized phase shifting scheme to calculate the modulation transfer function (MTF) at each frequency. This method results in a significantly reduced number of the optical field measurements, as compared to those with separate illumination of single-frequency sinusoidal patterns, and, thereby, achieves a fast data acquisition that is desired for a dynamic imaging application. Fluorescence yield images are recovered with the normalized Born formulated inversion of the diffusion model by simultaneously using the multi-frequency MTFs. Simulative and experimental reconstructions are performed in comparison with the single-frequency scheme to validate the proposed algorithm. The results suggest that adoption of the multi-frequency strategy to the SFD-FT can substantially improve the reconstruction quality, as well as its imaging resolution and quantitative accuracy.

Sparsity-promoting Bayesian approximation error method for compensating for the mismodeling of optical properties in fluorescence molecular tomography

In fluorescence molecular tomography, the quality of image reconstruction of fluorophore distribution is heavily reduced by the mismodeling of inaccurately known tissue optical properties. We propose an efficient sparsity-promoting Bayesian approximation error (spBAE) method to compensate for this mismodeling. The spBAE method incorporates sparsity prior into the BAE method by updating the parameters of the Gaussian prior according to its initial solution. In vivo experiments demonstrate that this new method greatly improves the reconstruction quality.

Compactly Supported Radial Basis Function-Based Meshless Method for Photon Propagation Model of Fluorescence Molecular Tomography

Fluorescence Molecular Tomography (FMT) is a powerful imaging modality for the research of cancer diagnosis, disease treatment and drug discovery. Via three-dimensional (3-D) imaging reconstruction, it can quantitatively and noninvasively obtain the distribution of fluorescent probes in biological tissues. Currently, photon propagation of FMT is conventionally described by the Finite Element Method (FEM), and it can obtain acceptable image quality. However, there are still some inherent inadequacies in FEM, such as time consuming, discretization error and inflexibility in mesh generation, which partly limit its imaging accuracy. To further improve the solving accuracy of photon propagation model (PPM), we propose a novel compactly supported radial basis functions (CSRBFs)-based meshless method (MM) to implement the PPM of FMT. We introduced a series of independent nodes and continuous CSRBFs to interpolate the PPM, which can avoid complicated mesh generation. To analyze the performance of the proposed MM, we carried out numerical heterogeneous mouse simulation to validate the simulated surface fluorescent measurement. Then we performed an in vivo experiment to observe the tomographic reconstruction. The experimental results confirmed that our proposed MM could obtain more similar surface fluorescence measurement with the golden standard (Monte-Carlo method), and more accurate reconstruction result was achieved via MM in in vivo application.

Automatic selection of regularization parameters for dynamic fluorescence molecular tomography: a comparison of L-curve and U-curve methods

Dynamic fluorescence molecular tomography (FMT) is a promising technique for the study of the metabolic process of fluorescent agents in the biological body in vivo, and the quality of the parametric images relies heavily on the accuracy of the reconstructed FMT images. In typical dynamic FMT implementations, the imaged object is continuously monitored for more than 50 minutes. During each minute, a set of the fluorescent measurements is acquired and the corresponding FMT image is reconstructed. It is difficult to manually set the regularization parameter in the reconstruction of each FMT image. In this paper, the parametric images obtained with the L-curve and U-curve methods are quantitatively evaluated through numerical simulations, phantom experiments and in vivo experiments. The results illustrate that the U-curve method obtains better accuracy, stronger robustness and higher noise-resistance in parametric imaging. Therefore, it is a promising approach to automatic selection of the regularization parameters for dynamic FMT.

Reconstruction for Limited-Projection Fluorescence Molecular Tomography Based on a Double-Mesh Strategy

Limited-projection fluorescence molecular tomography (FMT) has short data acquisition time that allows fast resolving of the three-dimensional visualization of fluorophore within small animal in vivo. However, limited-projection FMT reconstruction suffers from severe ill-posedness because only limited projections are used for reconstruction. To alleviate the ill-posedness, a feasible region extraction strategy based on a double mesh is presented for limited-projection FMT. First, an initial result is rapidly recovered using a coarse discretization mesh. Then, the reconstructed fluorophore area in the initial result is selected as a feasible region to guide the reconstruction using a fine discretization mesh. Simulation experiments on a digital mouse and small animal experiment in vivo are performed to validate the proposed strategy. It demonstrates that the presented strategy provides a good distribution of fluorophore with limited projections of fluorescence measurements. Hence, it is suitable for reconstruction of limited-projection FMT.

Non-stationary reconstruction for dynamic fluorescence molecular tomography with extended kalman filter

Dynamic fluorescence molecular tomography (FMT) plays an important role in drug delivery research. However, the majority of current reconstruction methods focus on solving the stationary FMT problems. If the stationary reconstruction methods are applied to the time-varying fluorescence measurements, the reconstructed results may suffer from a high level of artifacts. In addition, based on the stationary methods, only one tomographic image can be obtained after scanning one circle projection data. As a result, the movement of fluorophore in imaged object may not be detected due to the relative long data acquisition time (typically >1 min). In this paper, we apply extended kalman filter (EKF) technique to solve the non-stationary fluorescence tomography problem. Especially, to improve the EKF reconstruction performance, the generalized inverse of kalman gain is calculated by a second-order iterative method. The numerical simulation, phantom, and in vivo experiments are performed to evaluate the performance of the method. The experimental results indicate that by using the proposed EKF-based second-order iterative (EKF-SOI) method, we cannot only clearly resolve the time-varying distributions of fluorophore within imaged object, but also greatly improve the reconstruction time resolution (~2.5 sec/frame) which makes it possible to detect the movement of fluorophore during the imaging processes.

High-dynamic-range fluorescence molecular tomography for imaging of fluorescent targets with large concentration differences

When CCD-based free-space fluorescence molecular tomography (FMT) is used for imaging of fluorescent targets with a large concentration difference, the limited dynamic range of the CCD diminishes the localization and quantitative accuracy of FMT. To overcome this, we present a high-dynamic-range FMT (HDR-FMT) method. Under the multiple-exposure scheme, HDR fluorescence projection images are constructed using the recovered CCD response curve. Image reconstruction is implemented using iterative reweighted L1 regularization which can reduce artifacts by using fewer HDR fluorescence projection images. Phantom and in vivo animal studies indicate that localization of fluorescent targets with a large concentration difference is effectively improved with HDR-FMT and with good quantitative accuracy.

Quantitative performance characterization of three-dimensional noncontact fluorescence molecular tomography

Fluorescent proteins and dyes are routine tools for biological research to describe the behavior of genes, proteins, and cells, as well as more complex physiological dynamics such as vessel permeability and pharmacokinetics. The use of these probes in whole body in vivo imaging would allow extending the range and scope of current biomedical applications and would be of great interest. In order to comply with a wide variety of application demands, in vivo imaging platform requirements span from wide spectral coverage to precise quantification capabilities. Fluorescence molecular tomography (FMT) detects and reconstructs in three dimensions the distribution of a fluorophore in vivo. Noncontact FMT allows fast scanning of an excitation source and noninvasive measurement of emitted fluorescent light using a virtual array detector operating in free space. Here, a rigorous process is defined that fully characterizes the performance of a custom-built horizontal noncontact FMT setup. Dynamic range, sensitivity, and quantitative accuracy across the visible spectrum were evaluated using fluorophores with emissions between 520 and 660 nm. These results demonstrate that high-performance quantitative three-dimensional visible light FMT allowed the detection of challenging mesenteric lymph nodes in vivo and the comparison of spectrally distinct fluorescent reporters in cell culture.

An Inversion Scheme for Hybrid Fluorescence Molecular Tomography Using a Fuzzy Inference System

The imaging performance of fluorescence molecular tomography (FMT) improves when information from the underlying anatomy is incorporated into the inversion scheme, in the form of priors. The requirement for incorporation of priors has recently driven the development of hybrid FMT systems coupled to other modalities, such as X-ray CT and MRI. A critical methodological aspect in this modality relates to the particular method selected to incorporate prior information obtained from the anatomical imaging modality into the FMT inversion. We propose herein a new approach for utilizing prior information, which preferentially minimizes residual errors associated with measurements that better describe the anatomical segments considered. This preferential minimization was realized using a weighted least square (WLS) approach, where the weights were optimized using a Mamdani-type fuzzy inference system. The method of priors introduced herein was deployed as a two-step structured regularization approach and was verified with experimental measurements from phantoms as well as ex vivo and in vivo animal studies. The results demonstrate accurate performance and minimization of reconstruction bias, without requiring user input for setting the regularization parameters. As such, the proposed method offers significant progress in incorporation of anatomical priors in FMT and, as a result, in realization of the full potential of hybrid FMT.

Patch-based anisotropic diffusion scheme for fluorescence diffuse optical tomography--part 2: image reconstruction

Fluorescence diffuse optical tomography (fDOT) provides 3D images of fluorescence distributions in biological tissue, which represent molecular and cellular processes. The image reconstruction problem is highly ill-posed and requires regularisation techniques to stabilise and find meaningful solutions. Quadratic regularisation tends to either oversmooth or generate very noisy reconstructions, depending on the regularisation strength. Edge preserving methods, such as anisotropic diffusion regularisation (AD), can preserve important features in the fluorescence image and smooth out noise. However, AD has limited ability to distinguish an edge from noise. We propose a patch-based anisotropic diffusion regularisation (PAD), where regularisation strength is determined by a weighted average according to the similarity between patches around voxels within a search window, instead of a simple local neighbourhood strategy. However, this method has higher computational complexity and, hence, we wavelet compress the patches (PAD-WT) to speed it up, while simultaneously taking advantage of the denoising properties of wavelet thresholding. Furthermore, structural information can be incorporated into the image reconstruction with PAD-WT to improve image quality and resolution. In this case, the weights used to average voxels in the image are calculated using the structural image, instead of the fluorescence image. The regularisation strength depends on both structural and fluorescence images, which guarantees that the method can preserve fluorescence information even when it is not structurally visible in the anatomical images. In part 1, we tested the method using a denoising problem. Here, we use simulated and in vivo mouse fDOT data to assess the algorithm performance. Our results show that the proposed PAD-WT method provides high quality and noise free images, superior to those obtained using AD.

Multiplexed fluorescence tomography with spectral and temporal data: demixing with intrinsic regularization

We consider the joint use of spectral and temporal data for multiplexed fluorescence molecular tomography to enable high-throughput imaging of multiple fluorescent targets in bulk tissue. This is a challenging problem due to the narrow near-infrared diagnostic window and relatively broad emission spectra of common fluorophores, and the distortion ("redshift") that the fluorophore signals undergo as they propagate through tissue. We show through a Cramér-Rao lower bound analysis that demixing with spectral-temporal data could result in an order of magnitude improvement in performance over either modality alone. To cope with the resulting large data set, we propose a novel two-stage algorithm that decouples the demixing and tomographic reconstruction operations. In this work we concentrate on the demixing stage. We introduce an approach which incorporates ideas from sparse subspace clustering and compressed sensing and does not require a regularization parameter. We report on simulations in which we simultaneously demixed four fluorophores with closely overlapping spectral and temporal profiles in a 25 mm diameter cross-sectional area with a root-mean-square error of less than 3% per fluorophore, as well as on studies of sensitivity of the method to model mismatch.

Adaptively Alternative Light-Transport-Model-Based Three-Dimensional Optical Imaging for Longitudinal and Quantitative Monitoring of Gastric Cancer in Live Animal

OBJECTIVE

The existence of void regions and the complexity of structural heterogeneity and optical specificity are two challenges encountered in three-dimensional (3-D) optical imaging of in situ gastric cancer. An adaptively alternative light-transport-model-based 3-D optical imaging method was developed for addressing these challenges.

METHODS

In this newly developed imaging method, both the anatomical structure and optical properties are considered as a priori information, which makes the full use of the specificity of the biological tissues and improves both the quality and efficiency of the reconstructed images. The soul of the adaptively alternative technique is technique is configured to first classify the tissues based on the anatomical structure and optical properties and then select various equations to specifically characterize the light transport in different categories of tissues.

RESULTS

A series of rigorous simulations were conducted to demonstrate the performance of the hybrid light transport model, and the quality of the reconstructed images was then evaluated using a digital-mouse-based gastric cancer mimicing simulation, which showed that the localization error was less than 1 mm and the quantification error was approximately 10%. Finally, the applicability of the proposed method for in vivo imaging of gastric cancer was illustrated using groups of gastric cancer-bearing mice, which demonstrated the ability of the proposed method for longitudinal and quantitative monitoring of gastric cancer.

CONCLUSION

The developed method offers improved probability and great potential in the applications of earlier detecting in situ gastric cancer and longitudinal and quantitative observation of its development.

Noncontact full-angle fluorescence molecular tomography system based on rotary mirrors

We propose a novel noncontact fluorescence molecular tomography system that achieves full-angle capacity with the use of a new rotary-mirrors-based imaging head. In the imaging head, four plane mirrors are mounted on a rotating gantry to enable illumination and detection over 360°. In comparison with existing full-angle systems, our system does not require rotation of the specimen animal, a large and heavy light source (with scanning head), or a bulky camera (with filters and lens). The system design and implementation are described in detail. Both physical phantom and in vivo experiments are performed to verify the performance of the proposed system.

X-ray luminescence computed tomography imaging based on X-ray distribution model and adaptively split Bregman method

X-ray luminescence computed tomography (XLCT) has become a promising imaging technology for biological application based on phosphor nanoparticles. There are mainly three kinds of XLCT imaging systems: pencil beam XLCT, narrow beam XLCT and cone beam XLCT. Narrow beam XLCT can be regarded as a balance between the pencil beam mode and the cone-beam mode in terms of imaging efficiency and image quality. The collimated X-ray beams are assumed to be parallel ones in the traditional narrow beam XLCT. However, we observe that the cone beam X-rays are collimated into X-ray beams with fan-shaped broadening instead of parallel ones in our prototype narrow beam XLCT. Hence we incorporate the distribution of the X-ray beams in the physical model and collected the optical data from only two perpendicular directions to further speed up the scanning time. Meanwhile we propose a depth related adaptive regularized split Bregman (DARSB) method in reconstruction. The simulation experiments show that the proposed physical model and method can achieve better results in the location error, dice coefficient, mean square error and the intensity error than the traditional split Bregman method and validate the feasibility of method. The phantom experiment can obtain the location error less than 1.1 mm and validate that the incorporation of fan-shaped X-ray beams in our model can achieve better results than the parallel X-rays.

Fluorescence molecular tomography reconstruction via discrete cosine transform-based regularization

Fluorescence molecular tomography (FMT) as a noninvasive imaging modality has been widely used for biomedical preclinical applications. However, FMT reconstruction suffers from severe ill-posedness, especially when a limited number of projections are used. In order to improve the quality of FMT reconstruction results, a discrete cosine transform (DCT) based reweighted L1-norm regularization algorithm is proposed. In each iteration of the reconstruction process, different reweighted regularization parameters are adaptively assigned according to the values of DCT coefficients to suppress the reconstruction noise. In addition, the permission region of the reconstructed fluorophores is adaptively constructed to increase the convergence speed. In order to evaluate the performance of the proposed algorithm, physical phantom and in vivo mouse experiments with a limited number of projections are carried out. For comparison, different L1-norm regularization strategies are employed. By quantifying the signal-to-noise ratio (SNR) of the reconstruction results in the phantom and in vivo mouse experiments with four projections, the proposed DCT-based reweighted L1-norm regularization shows higher SNR than other L1-norm regularizations employed in this work.

Next-generation Raman tomography instrument for non-invasive in vivo bone imaging

Combining diffuse optical tomography methods with Raman spectroscopy of tissue provides the ability for in vivo measurements of chemical and molecular characteristics, which have the potential for being useful in diagnostic imaging. In this study a system for Raman tomography was developed and tested. A third generation microCT coupled system was developed to combine 10 detection fibers and 5 excitation fibers with laser line filtering and a Cytop reference signal. Phantom measurements of hydroxyapatite concentrations from 50 to 300 mg/ml had a linear response. Fiber placement and experiment design was optimized using cadaver animals with live animal measurements acquired to validate the systems capabilities. Promising results from the initial animal experiments presented here, pave the way for a study of longitudinal measurements during fracture healing and the scaling of the Raman tomography system towards human measurements.

Optical in vivo imaging of the alarmin S100A9 in tumor lesions allows for estimation of the individual malignant potential by evaluation of tumor-host cell interaction

Tumors recruit and reprogram immune cells to support tumor development and spread, the most prominent among them being of monocytic origin such as tumor-associated macrophages (TAM) or myeloid-derived suppressor cells (MDSC). The alarmin S100A8/A9 has been implicated in the induction of TAM and MDSC. We assessed S100A9 as a molecular imaging marker for the activity of tumor-associated immune cells in a syngeneic murine breast cancer model. S100A9 could serve as a surrogate marker for tumor immune crosstalk as a function of malignancy, providing a tool with the potential for both basic research in tumor immunology and clinical stratification of patients.

METHODS

BALB/c mice were inoculated with murine breast cancer cells of common origin but different metastatic capability. At different times during tumor development, optical imaging was performed using a S100A9-specific probe to visualize activated monocytes. To further explore the impact of tumor-educated monocytes, splenic myeloid cells were isolated from either healthy or tumor-bearing animals and injected into tumor-bearing mice. We analyzed the effect of the cell transfer on immune cell activity and tumor development.

RESULTS

We could prove S100A9-driven imaging to sensitively and specifically reflect monocyte activity in primary tumor lesions. The imaging results were corroborated by histology and fluorescence-activated cell sorting analyses. In a prospective experiment, S100A9 imaging proved indicative of the individual tumor growth, with excellent correlation. Moreover, we could show that the monocyte activity as depicted by S100A9 activity in the primary tumor lesion mirrored the tumor's metastatic behavior. Treatment with tumor-primed splenic monocytes induced increased tumor growth, accompanied by an augmented infiltration of activated myeloid cells (MDSC and TAM) into the tumor. The consecutive S100A9 expression as depicted by in vivo imaging was significantly increased.

CONCLUSION

S100A9 proved to be a sensitive and specific marker for the activity of tumor-associated immune cells. To our knowledge, S100A9 imaging represents a first in vivo imaging approach for the estimation of recruitment and activity of tumor-associated myeloid immune cells. We demonstrated the potential value of this imaging approach for prediction of local and systemic tumor development.

Fluorescent probes concentration estimation in vitro and ex vivo as a model for early detection of Alzheimer's disease

The pathogenic process of Alzheimer's disease (AD) begins years before clinical diagnosis. Here, we suggest a method that may detect AD several years earlier than current exams. The method is based on previous reports that relate the concentration ratio of biomarkers (amyloid-beta and tau) in the cerebrospinal fluid (CSF) to the development of AD. Our method replaces the lumbar puncture process required for CSF drawing by using fluorescence measurements. The system uses an optical fiber coupled to a laser source and a detector. The laser radiation excites two fluorescent probes which may bond to the CSF biomarkers. Their concentration ratio is extracted from the fluorescence intensities and can be used for future AD detection. First, we present a theoretical model for fluorescence concentration ratio estimation. The method's feasibility was validated using Monte Carlo simulations. Its accuracy was then tested using multilayered tissue phantoms simulating the epidural fat, CSF, and bone. These phantoms have various optical properties, thicknesses, and fluorescence concentrations in order to simulate human anatomy variations and different fiber locations. The method was further tested using ex vivo chicken tissue. The average errors of the estimated concentration ratios were low both in vitro (4.4%) and ex vivo (10.9%), demonstrating high accuracy.

An adaptive support driven reweighted L1-regularization algorithm for fluorescence molecular tomography

Fluorescence molecular tomography (FMT) is a promising in vivo functional imaging modality in preclinical study. When solving the ill-posed FMT inverse problem, L1 regularization can preserve the details and reduce the noise in the reconstruction results effectively. Moreover, compared with the regular L1 regularization, reweighted L1 regularization is recently reported to improve the performance. In order to realize the reweighted L1 regularization for FMT, an adaptive support driven reweighted L1-regularization (ASDR-L1) algorithm is proposed in this work. This algorithm has two integral parts: an adaptive support estimate and the iteratively updated weights. In the iteratively reweighted L1-minimization sub-problem, different weights are equivalent to different regularization parameters at different locations. Thus, ASDR-L1 can be considered as a kind of spatially variant regularization methods for FMT. Physical phantom and in vivo mouse experiments were performed to validate the proposed algorithm. The results demonstrate that the proposed reweighted L1-reguarization algorithm can significantly improve the performance in terms of relative quantitation and spatial resolution.

An adaptive support driven reweighted L1-regularization algorithm for fluorescence molecular tomography

Fluorescence molecular tomography (FMT) is a promising in vivo functional imaging modality in preclinical study. When solving the ill-posed FMT inverse problem, L1 regularization can preserve the details and reduce the noise in the reconstruction results effectively. Moreover, compared with the regular L1 regularization, reweighted L1 regularization is recently reported to improve the performance. In order to realize the reweighted L1 regularization for FMT, an adaptive support driven reweighted L1-regularization (ASDR-L1) algorithm is proposed in this work. This algorithm has two integral parts: an adaptive support estimate and the iteratively updated weights. In the iteratively reweighted L1-minimization sub-problem, different weights are equivalent to different regularization parameters at different locations. Thus, ASDR-L1 can be considered as a kind of spatially variant regularization methods for FMT. Physical phantom and in vivo mouse experiments were performed to validate the proposed algorithm. The results demonstrate that the proposed reweighted L1-reguarization algorithm can significantly improve the performance in terms of relative quantitation and spatial resolution.

Pixel-based absorption correction for dual-tracer fluorescence imaging of receptor binding potential

Ratiometric approaches to quantifying molecular concentrations have been used for decades in microscopy, but have rarely been exploited in vivo until recently. One dual-tracer approach can utilize an untargeted reference tracer to account for non-specific uptake of a receptor-targeted tracer, and ultimately estimate receptor binding potential quantitatively. However, interpretation of the relative dynamic distribution kinetics is confounded by differences in local tissue absorption at the wavelengths used for each tracer. This study simulated the influence of absorption on fluorescence emission intensity and depth sensitivity at typical near-infrared fluorophore wavelength bands near 700 and 800 nm in mouse skin in order to correct for these tissue optical differences in signal detection. Changes in blood volume [1-3%] and hemoglobin oxygen saturation [0-100%] were demonstrated to introduce substantial distortions to receptor binding estimates (error > 30%), whereas sampled depth was relatively insensitive to wavelength (error < 6%). In response, a pixel-by-pixel normalization of tracer inputs immediately post-injection was found to account for spatial heterogeneities in local absorption properties. Application of the pixel-based normalization method to an in vivo imaging study demonstrated significant improvement, as compared with a reference tissue normalization approach.

Fast reconstruction of fluorescence molecular tomography via a permissible region extraction strategy

In order to obtain precise reconstruction results in fluorescence molecular tomography (FMT), large-scale matrix equations would be solved in the inverse problem generally. Thus, much time and memory needs to be consumed. In this paper, a permissible region extraction strategy is proposed to solve this problem. First, a preliminary result is rapidly reconstructed using the weight matrix compressed by principal component analysis or uniform sampling. And then the reconstructed target area in this preliminary result is considered as the a priori permissible region to guide the final reconstruction. Phantom experiments with double fluorescent targets are performed to test the performance of the strategy. The results illustrate that the proposed strategy can significantly accelerate the image reconstruction in FMT almost without quality degradation.

Absorption reconstruction improves biodistribution assessment of fluorescent nanoprobes using hybrid fluorescence-mediated tomography

Aim: Fluorescence-mediated tomography (FMT) holds potential for accelerating diagnostic and theranostic drug development. However, for proper quantitative fluorescence reconstruction, knowledge on optical scattering and absorption, which are highly heterogeneous in different (mouse) tissues, is required. We here describe methods to assess these parameters using co-registered micro Computed Tomography (µCT) data and nonlinear whole-animal absorption reconstruction, and evaluate their importance for assessment of the biodistribution and target site accumulation of fluorophore-labeled drug delivery systems.

Methods: Besides phantoms with varying degrees of absorption, mice bearing A431 tumors were imaged 15 min and 48 h after i.v. injection of a fluorophore-labeled polymeric drug carrier (pHPMA-Dy750) using µCT-FMT. The outer shape of mice and a scattering map were derived using automated segmentation of the µCT data. Furthermore, a 3D absorption map was reconstructed from the trans-illumination data. We determined the absorption of five interactively segmented regions (heart, liver, kidney, muscle, tumor). Since blood is the main near-infrared absorber in vivo, the absorption was also estimated from the relative blood volume (rBV), determined by contrast-enhanced µCT. We compared the reconstructed absorption with the rBV-based values and analyzed the effect of using the absorption map on the fluorescence reconstruction.

Results: Phantom experiments demonstrated that absorption reconstruction is possible and necessary for quantitative fluorescence reconstruction. In vivo, the reconstructed absorption showed high values in strongly blood-perfused organs such as the heart, liver and kidney. The absorption values correlated strongly with the rBV-based absorption values, confirming the accuracy of the absorption reconstruction. Usage of homogenous absorption instead of the reconstructed absorption map resulted in reduced values in the heart, liver and kidney, by factors of 3.5, 2.1 and 1.4, respectively. For muscle and subcutaneous tumors, which have a much lower rBV and absorption, absorption reconstruction was less important.

Conclusion: Quantitative whole-animal absorption reconstruction is possible and can be validated in vivo using the rBV. Usage of an absorption map is important when quantitatively assessing the biodistribution of fluorescently labeled drugs and drug delivery systems, to avoid a systematic underestimation of fluorescence in strongly absorbing organs, such as the heart, liver and kidney.

Reweighted L1 regularization for restraining artifacts in FMT reconstruction images with limited measurements

In fluorescence molecular tomography (FMT), many artifacts exist in the reconstructed images because of the inherently ill-posed nature of the FMT inverse problem, especially with limited measurements. A new method based on iterative reweighted L1 (IRL1) regularization is proposed for reducing artifacts with limited measurements. Phantom experiments demonstrate that the reconstructed images have fewer artifacts even with very limited measurements. This indicates that FMT based on IRL1 can obtain high-quality images and thus has the potential to observe dynamic changes in fluorescence-targeted molecules.

Ultrasound guided fluorescence molecular tomography with improved quantification by an attenuation compensated Born-normalization and in vivo preclinical study of cancer

Ultrasound imaging, having the advantages of low-cost and non-invasiveness over MRI and X-ray CT, was reported by several studies as an adequate complement to fluorescence molecular tomography with the perspective of improving localization and quantification of fluorescent molecular targets in vivo. Based on the previous work, an improved dual-modality Fluorescence-Ultrasound imaging system was developed and then validated in imaging study with preclinical tumor model. Ultrasound imaging and a profilometer were used to obtain the anatomical prior information and 3D surface, separately, to precisely extract the tissue boundary on both sides of sample in order to achieve improved fluorescence reconstruction. Furthermore, a pattern-based fluorescence reconstruction on the detection side was incorporated to enable dimensional reduction of the dataset while keeping the useful information for reconstruction. Due to its putative role in the current imaging geometry and the chosen reconstruction technique, we developed an attenuation compensated Born-normalization method to reduce the attenuation effects and cancel off experimental factors when collecting quantitative fluorescence datasets over large area. Results of both simulation and phantom study demonstrated that fluorescent targets could be recovered accurately and quantitatively using this reconstruction mechanism. Finally, in vivo experiment confirms that the imaging system associated with the proposed image reconstruction approach was able to extract both functional and anatomical information, thereby improving quantification and localization of molecular targets.

High-performance fluorescence molecular tomography through shape-based reconstruction using spherical harmonics parameterization

Fluorescence molecular tomography in the near-infrared region is becoming a powerful modality for mapping the three-dimensional quantitative distributions of fluorochromes in live small animals. However, wider application of fluorescence molecular tomography still requires more accurate and stable reconstruction tools. We propose a shape-based reconstruction method that uses spherical harmonics parameterization, where fluorophores are assumed to be distributed as piecewise constants inside disjointed subdomains and the remaining background. The inverse problem is then formulated as a constrained nonlinear least-squares problem with respect to shape parameters, which decreases ill-posedness because of the significantly reduced number of unknowns. Since different shape parameters contribute differently to the boundary measurements, a two-step and modified block coordinate descent optimization algorithm is introduced to stabilize the reconstruction. We first evaluated our method using numerical simulations under various conditions for the noise level and fluorescent background; it showed significant superiority over conventional voxel-based methods in terms of the spatial resolution, reconstruction accuracy with regard to the morphology and intensity, and robustness against the initial estimated distribution. In our phantom experiment, our method again showed better spatial resolution and more accurate intensity reconstruction. Finally, the results of an in vivo experiment demonstrated its applicability to the imaging of mice.

Enhanced fluorescence diffuse optical tomography with indocyanine green-encapsulating liposomes targeted to receptors for vascular endothelial growth factor in tumor vasculature

To develop an indocyanine green (ICG) tracer with slower clearance kinetics, we explored ICG-encapsulating liposomes (Lip) in three different formulations: untargeted (Lip/ICG), targeted to vascular endothelial growth factor (VEGF) receptors (scVEGF-Lip/ICG) by the receptor-binding moiety single-chain VEGF (scVEGF), or decorated with inactivated scVEGF (inactive-Lip/ICG) that does not bind to VEGF receptors. Experiments were conducted with tumor-bearing mice that were placed in a scattering medium with tumors located at imaging depths of either 1.5 or 2.0 cm. Near-infrared fluorescence diffuse optical tomography that provides depth-resolved spatial distributions of fluorescence in tumor was used for the detection of postinjection fluorescent signals. All liposome-based tracers, as well as free ICG, were injected intravenously into mice in the amounts corresponding to 5 nmol of ICG/mouse, and the kinetics of increase and decrease of fluorescent signals in tumors were monitored. A signal from free ICG reached maximum at 15-min postinjection and then rapidly declined with t1/2 of ~20 min. The signals from untargeted Lip/ICG and inactive-Lip/ICG also reached maximum at 15-min postinjection, however, declined somewhat slower than free ICG with t1/2 of ~30 min. By contrast, a signal from targeted scVEGF-Lip/ICG grew slower than that of all other tracers, reaching maximum at 30-min postinjection and declined much slower than that of other tracers with t1/2 of ~90 min, providing a more extended observation window. Higher scVEGF-Lip/ICG tumor accumulation was further confirmed by the analysis of fluorescence on cryosections of tumors that were harvested from animals at 400 min after injection with different tracers.

Efficient L1 regularization-based reconstruction for fluorescent molecular tomography using restarted nonlinear conjugate gradient

For the ill-posed fluorescent molecular tomography (FMT) inverse problem, the L1 regularization can protect the high-frequency information like edges while effectively reduce the image noise. However, the state-of-the-art L1 regularization-based algorithms for FMT reconstruction are expensive in memory, especially for large-scale problems. An efficient L1 regularization-based reconstruction algorithm based on nonlinear conjugate gradient with restarted strategy is proposed to increase the computational speed with low memory consumption. The reconstruction results from phantom experiments demonstrate that the proposed algorithm can obtain high spatial resolution and high signal-to-noise ratio, as well as high localization accuracy for fluorescence targets.

Targeting tumor hypoxia with 2-nitroimidazole-indocyanine green dye conjugates

Tumor hypoxia is a major indicator of treatment resistance to chemotherapeutic drugs, and fluorescence optical tomography has tremendous potential to provide clinically useful, functional information by identifying tumor hypoxia. The synthesis of a 2-nitroimidazole-indocyanine green conjugate using a piperazine linker (piperazine-2-nitroimidazole-ICG) capable of robust fluorescent imaging of tumor hypoxia is described. In vivo mouse tumor imaging studies were completed and demonstrate an improved imaging capability of the new dye relative to an earlier version of the dye that was synthesized with an ethanolamine linker (ethanolamine-2-nitroimidazole-ICG). Mouse tumors located at imaging depths of 1.5 and 2.0 cm in a turbid medium were imaged at various time points after intravenous injection of the dyes. On average, the reconstructed maximum fluorescence concentration of the tumors injected with piperazine-2-nitroimidazole-ICG was twofold higher than that injected with ethanolamine-2-nitroimidazole-ICG within 3 h postinjection period and 1.6 to 1.7 times higher beyond 3 h postinjection. The untargeted bis-carboxylic acid ICG completely washed out after 3 h postinjection. Thus, the optimal window to assess tumor hypoxia is beyond 3 h postinjection. These findings were supported with fluorescence images of histological sections of tumor samples and an immunohistochemistry technique for identifying tumor hypoxia.

Quantitative fluorescence diffuse optical tomography in the presence of heterogeneities

In fluorescence diffuse optical tomography (fDOT), the accuracy of reconstructed fluorescence distributions highly depends on the knowledge of the tissue optical heterogeneities for correct modeling of light propagation. Common approaches are to assume homogeneous optical properties or, when structural information is available, assign optical properties to various segmented organs, which is likely to result in inaccurate reconstructions. Furthermore, DOT based only on intensity (continuous wave-DOT) is a nonunique inverse problem, and hence, cannot be used to retrieve simultaneously maps of absorption and diffusion coefficients. We propose a method that reconstructs a single parameter from the excitation measurements, which is used in the fDOT problem to accurately recover fluorescence distribution.

Acceleration of early-photon fluorescence molecular tomography with graphics processing units

Fluorescence molecular tomography (FMT) with early-photons can improve the spatial resolution and fidelity of the reconstructed results. However, its computing scale is always large which limits its applications. In this paper, we introduced an acceleration strategy for the early-photon FMT with graphics processing units (GPUs). According to the procedure, the whole solution of FMT was divided into several modules and the time consumption for each module is studied. In this strategy, two most time consuming modules (G_d and W modules) were accelerated with GPU, respectively, while the other modules remained coded in the Matlab. Several simulation studies with a heterogeneous digital mouse atlas were performed to confirm the performance of the acceleration strategy. The results confirmed the feasibility of the strategy and showed that the processing speed was improved significantly.

Adaptive wide-field optical tomography

We describe a wide-field optical tomography technique, which allows the measurement-guided optimization of illumination patterns for enhanced reconstruction performances. The iterative optimization of the excitation pattern aims at reducing the dynamic range in photons transmitted through biological tissue. It increases the number of measurements collected with high photon counts resulting in a dataset with improved tomographic information. Herein, this imaging technique is applied to time-resolved fluorescence molecular tomography for preclinical studies. First, the merit of this approach is tested by in silico studies in a synthetic small animal model for typical illumination patterns. Second, the applicability of this approach in tomographic imaging is validated in vitro using a small animal phantom with two fluorescent capillaries occluded by a highly absorbing inclusion. The simulation study demonstrates an improvement of signal transmitted (∼2 orders of magnitude) through the central portion of the small animal model for all patterns considered. A corresponding improvement in the signal at the emission wavelength by 1.6 orders of magnitude demonstrates the applicability of this technique for fluorescence molecular tomography. The successful discrimination and localization (∼1  mm error) of the two objects with higher resolution using the optimized patterns compared with nonoptimized illumination establishes the improvement in reconstruction performance when using this technique.

Separating structures of different fluorophore concentrations by principal component analysis on multispectral excitation-resolved fluorescence tomography images

Multispectral excitation-resolved fluorescence tomography (MEFT) uses excitation light of different wavelengths to illuminate the fluorophores and obtains the reconstruction image frame which is fluorescence yield at each corresponding wavelength. For structures containing fluorophores of different concentrations, fluorescence yields show different variation trends with the excitation spectrum. In this study, principal component analysis (PCA) is used to analyze the MEFT reconstructed image frames. By taking advantage of the different variation trends of fluorescence yields, PCA can provide a set of principal components (PCs) in which structures containing different concentrations of fluorophores are shown separately. Simulations and experiments are both performed to test the performance of the proposed algorithm. The results suggest that the location and structure of fluorophores with different concentrations can be obtained and the contrast of fluorophores can be improved further by using this algorithm.

Dual-tracer background subtraction approach for fluorescent molecular tomography

Diffuse fluorescence tomography requires high contrast-to-background ratios to accurately reconstruct inclusions of interest. This is a problem when imaging the uptake of fluorescently labeled molecularly targeted tracers in tissue, which can result in high levels of heterogeneously distributed background uptake. We present a dual-tracer background subtraction approach, wherein signal from the uptake of an untargeted tracer is subtracted from targeted tracer signal prior to image reconstruction, resulting in maps of targeted tracer binding. The approach is demonstrated in simulations, a phantom study, and in a mouse glioma imaging study, demonstrating substantial improvement over conventional and homogenous background subtraction image reconstruction approaches.

MAP estimation with structural priors for fluorescence molecular tomography

Fluorescence molecular tomography (FMT) is an attractive imaging tool for quantitatively and three-dimensionally resolving fluorophore distributions in small animals, but it suffers from low spatial resolution due to its inherent ill-posed nature. Structural priors obtained from a secondary modality system such as x-ray computed tomography or magnetic resonance imaging can help to improve FMT reconstruction results. However, challenge remains in how to fully take advantage of the structural priors while effectively avoid undesirable influence caused by an immoderate usage. In this paper, we propose a new method to resolve the FMT inverse problem based on maximum a posteriori (MAP) estimation with structural priors (MAP-SP) in a Bayesian framework. Instead of imposing the structural priors directly on the reconstruction results, the MAP-SP method utilizes them to constrain the unknown hyperparameters of the prior information model which is essential for the Bayesian framework. Then, a low dimensional inverse problem and an alternating optimization scheme are used to automatically calculate the unknown hyperparameters, which make the FMT reconstruction process self-adaptive. Simulation and phantom results show that the proposed MAP-SP method can effectively make use of the structural priors and leads to improvements in reconstruction quality as compared with traditional regularization methods.

Weighted depth compensation algorithm for fluorescence molecular tomography reconstruction

In fluorescence molecular tomography (FMT), diffuse-light measurements are obtained from a series of source-detector pairs placed on the boundary of the medium. The sensitivity of measurements deteriorates quickly with increased distance from the sources and detectors and therefore yields poor depth quantitative recovery. A depth compensation algorithm is presented in this paper to reconstruct fluorescent inclusions in deep tissues. Two weight matrixes are employed to level off sensitivity differences and enhance prominent elements of the solution. Results of numerical and phantom experiments demonstrate that both relative quantitation and spatial resolution of FMT are improved for inclusions at different depths.

Fluorescence molecular tomography using a two-step three-dimensional shape-based reconstruction with graphics processing unit acceleration

In fluorescence molecular tomography, the accurate and stable reconstruction of fluorescence-labeled targets remains a challenge for wide application of this imaging modality. Here we propose a two-step three-dimensional shape-based reconstruction method using graphics processing unit (GPU) acceleration. In this method, the fluorophore distribution is assumed as the sum of ellipsoids with piecewise-constant fluorescence intensities. The inverse problem is formulated as a constrained nonlinear least-squares problem with respect to shape parameters, leading to much less ill-posedness as the number of unknowns is greatly reduced. Considering that various shape parameters contribute differently to the boundary measurements, we use a two-step optimization algorithm to handle them in a distinctive way and also stabilize the reconstruction. Additionally, the GPU acceleration is employed for finite-element-method-based calculation of the objective function value and the Jacobian matrix, which reduces the total optimization time from around 10 min to less than 1 min. The numerical simulations show that our method can accurately reconstruct multiple targets of various shapes while the conventional voxel-based reconstruction cannot separate the nearby targets. Moreover, the two-step optimization can tolerate different initial values in the existence of noises, even when the number of targets is not known a priori. A physical phantom experiment further demonstrates the method's potential in practical applications.