Biomedical subjects

Keith D Paulsen

Publications and source records attributed to Keith D Paulsen.

At least 19 recordsLinked to original sources

PubMed2006

In vivo EIS characterization of tumour tissue properties is dominated by excess extracellular fluid.

Electrical impedance spectroscopy (EIS) is a non-ionizing, non-invasive technique which can be used to detect the presence of malignant tumours based on their electrical properties. Although it has been suggested that the edema which accompanies tumours strongly influences EIS tumour characterization, such information has not, until now, been documented in the literature. Growing intramuscular rodent tumours were imaged using magnetic resonance imaging (MRI) and EIS at several time points post-tumour implantation. The amount of edema associated with the tumours was calculated from the MRI images. Electrical parameters (resistivity, permittivity, fluid index ratio and peak frequency) were extracted from the EIS spectra. Taken together, the resulting electrical parameters strongly indicate that edema is the dominating pathological feature in EIS characterization and can at times conceal the presence of the tumour. Receiver operating characteristic analysis supports these findings.

Animals↗

PubMed2006

Imaging breast adipose and fibroglandular tissue molecular signatures by using hybrid MRI-guided near-infrared spectral tomography.

Magnetic resonance (MR)-guided near-infrared spectral tomography was developed and used to image adipose and fibroglandular breast tissue of 11 normal female subjects, recruited under an institutional review board-approved protocol. Images of hemoglobin, oxygen saturation, water fraction, and subcellular scattering were reconstructed and show that fibroglandular fractions of both blood and water are higher than in adipose tissue. Variation in adipose and fibroglandular tissue composition between individuals was not significantly different across the scattered and dense breast categories. Combined MR and near-infrared tomography provides fundamental molecular information about these tissue types with resolution governed by MR T1 images.

Adipose Tissue↗

PubMed2006

Spectrally resolved bioluminescence optical tomography.

Spectrally resolved bioluminescence optical tomography is an approach to recover images of luciferase activity within a volume using multiwavelength emission data from internal bioluminescence sources. The underlying problem of uniqueness associated with nonspectrally resolved intensity-based bioluminescence tomography is highlighted. Reconstructed images of bioluminescence are presented by using as input both simulated and real multiwavelength data from a tissue-simulating phantom. The location of the internal bioluminescence is obtained with 1 mm accuracy. Further, the amplitude of the reconstructed source is proportional to the actual bioluminescence intensity.

Imaging, Three-Dimensional↗

PubMed2006

In vivo hemoglobin and water concentrations, oxygen saturation, and scattering estimates from near-infrared breast tomography using spectral reconstruction.

RATIONALE AND OBJECTIVES: Near-infrared (NIR) imaging has its niche in quantifying and characterizing functional changes in tissue relating to vascularity and metabolic status. Here, NIR tomography was applied to study mammographically normal breast tissue in vivo by evaluating relationships between functional parameters so obtained to clinical representers in an effort to understand factors influencing tissue compositional changes. MATERIALS AND METHODS: A new spectral reconstruction method that is considered to provide the most accurate estimates of hemoglobin level, oxygen saturation, water fraction, scattering power, and amplitude was used to assess healthy breast tissue imaged in vivo by means of NIR tomography. The approach directly recovers functional parameters with inherent inclusion of spectral behavior enforced through the incorporation of a priori model assumptions. Sixty subjects were imaged by using a frequency-domain instrument followed by spectral image reconstruction and statistical analysis for significant correlations. RESULTS: The new analysis shows statistically significant inverse correlations between body mass index and breast total hemoglobin and water fractions. Water fraction also correlated inversely with age and separated certain categories of breast density. Average scatter power was indicative of breast radiographic density composition, whereas scatter amplitude varied inversely with breast diameter. Total hemoglobin correlated with water fraction, whereas water correlated with scatter power. CONCLUSION: The changes observed here are attributable to volume fraction alterations and provide some of the most comprehensive data on breast composition variations with demographic factors.

Adult↗

PubMed2006

Data-guided brain deformation modeling: evaluation of a 3-D adjoint inversion method in porcine studies.

Biomechanical models of brain deformation are useful tools for estimating parenchymal shift that results during open cranial procedures. Intraoperative data is likely to improve model estimates, but incorporation of such data into the model is not trivial. This study tests the adjoint equations method (AEM) for data assimilation as a viable approach for integrating displacement data into a brain deformation model. AEM was applied to two porcine experiments. AEM-based estimates were compared both to measured displacement data [from computed tomography (CT) scans] and to model solutions obtained without the guidance of sparse data, which we term the best prior estimate (BPE). Additionally, the sensitivity of the AEM solution to inverse parameter selection was investigated. The results suggest that it is most important to estimate the size of the variance in the measurement error correctly, make the correlation length long and estimate displacement (over stress) boundary conditions. Application of AEM shows an average 33% improvement over BPE. This paper represents the first evidence of successful use of the AEM technique in three dimensions with experimental data validation. The guidelines established for selection of model parameters are starting points for further optimization of the method under clinical conditions.

Algorithms↗

PubMed2006

The multidimensional phase unwrapping integral and applications to microwave tomographical image reconstruction.

Spatial unwrapping of the phase component of time varying electromagnetic fields has important implications in a range of disciplines including synthetic aperture radar (SAR) interferometry, MRI, optical confocal microscopy, and microwave tomography. This paper presents a fundamental framework based on the phase unwrapping integral, especially in the complex case where phase singularities are enclosed within the closed path integral. With respect to the phase unwrapping required when utilized in Gauss-Newton iterative microwave image reconstruction, the concept of dynamic phase unwrapping is introduced where the singularity location varies as a function of the iteratively modified property distributions. Strategies for dynamic phase unwrapping in the microwave problem were developed and successfully tested in simulations and clinical experiments utilizing large, high contrast targets to validate the approach.

Algorithms↗

PubMed2006

Finite element implementation of Maxwell's equations for image reconstruction in electrical impedance tomography.

Traditionally, image reconstruction in electrical impedance tomography (EIT) has been based on Laplace's equation. However, at high frequencies the coupling between electric and magnetic fields requires solution of the full Maxwell equations. In this paper, a formulation is presented in terms of the Maxwell equations expressed in scalar and vector potentials. The approach leads to boundary conditions that naturally align with the quantities measured by EIT instrumentation. A two-dimensional implementation for image reconstruction from EIT data is realized. The effect of frequency on the field distribution is illustrated using the high-frequency model and is compared with Laplace solutions. Numerical simulations and experimental results are also presented to illustrate image reconstruction over a range of frequencies using the new implementation. The results show that scalar/vector potential reconstruction produces images which are essentially indistinguishable from a Laplace algorithm for frequencies below 1 MHz but superior at frequencies reaching 10 MHz.

Algorithms↗

PubMed2005

Spectral priors improve near-infrared diffuse tomography more than spatial priors.

We compare the benefits of spatial and spectral priors in near-infrared diffuse tomography image reconstruction. Although previous studies that incorporated anatomical spatial priors have shown improvement in algorithm convergence and resolution, our results indicate that functional parameter quantification by this approach can be suboptimal. The incorporation of a priori spectral information significantly improves the accuracy observed in recovered images. Specifically, phantom results show that the maximum total hemoglobin concentration ([Hb(T)]) in a region of heterogeneity reached 91% of the true value compared to 63% using spatial priors. The combination of both priors produced results accurate to 98% of the true [Hb(T)]. When both spatial and spectral priors were applied in a healthy volunteer, glandular tissue showed a higher [Hb(T)], water fraction, and scattering power compared to adipose tissue.

Algorithms↗

PubMed2005

Multiepitope HER2 targeting enhances photoimmunotherapy of HER2-overexpressing cancer cells with pyropheophorbide-a immunoconjugates.

Multi-targeting strategies improve the efficacy of antibody and immunotoxin therapies but have not yet been thoroughly explored for HER2-based cancer treatments. We investigated multi-epitope HER2 targeting to boost photosensitizer immunoconjugate uptake as a way of enhancing photoimmunotherapy. Photoimmunotherapy may allow targeted photodynamic destruction of malignancies and may also potentiate anticancer antibodies. However, one obstacle preventing its clinical use is the delivery of enough photosensitizer immunoconjugates to target cells. Anti-HER2 photosensitizer immunoconjugates were constructed from two monoclonal antibodies (mAb), HER50 and HER66, using a novel method originally developed to label photosensitizer immunoconjugates with the photosensitizer, benzoporphyrin derivative verteporfin. Photosensitizer immunoconjugates were labeled instead with a promising alternative photosensitizer, pyropheophorbide-a (PPa), which required only minor changes to the conjugation procedure. Uptake and phototoxicity experiments using human cancer cells were conducted with the photosensitizer immunoconjugates and, for comparison, with free PPa. SK-BR-3 and SK-OV-3 cells served as HER2-overexpressing target cells. MDA-MB-468 cells served as HER2-nonexpressing control cells. Photosensitizer immunoconjugates with PPa/mAb molar ratios up to approximately 10 specifically targeted and photodynamically killed HER2-overexpressing cells. On a per mole basis, photosensitizer immunoconjugates were less phototoxic than free PPa, but photosensitizer immunoconjugates were selective for target cells whereas free PPa was not. Multiepitope targeted photoimmunotherapy with a HER50 and HER66 photosensitizer immunoconjugate mixture was significantly more effective than single-epitope targeted photoimmunotherapy with a single anti-HER2 photosensitizer immunoconjugate, provided photosensitizer immunoconjugate binding was saturated. This study shows that multiepitope targeting enhances HER2-targeted photoimmunotherapy and maintains a high degree of specificity. Consequently, it seems that multitargeted photoimmunotherapy should also be useful against cancers that overexpress other receptors.

Breast Neoplasms↗

PubMed2005

Magnetic-resonance-imaging-coupled broadband near-infrared tomography system for small animal brain studies.

A novel magnetic-resonance-coupled broadband near-infrared (NIR) tomography system for small animal brain studies is described. Several features of the image formation approach are new in NIR tomography and represent major advances in the path to recovering high-resolution hemoglobin and oxygen saturation images of tissue. The NIR data were broadband and continuous wave and were used along with a second-derivative-based estimation of the path length from water absorption. The path length estimation from water was then used along with the attenuation spectrum to recover absorption and reduced scattering coefficient images at multiple wavelengths and then to recover images of total hemoglobin and oxygen saturation. Going beyond these basics of NIR tomography, software has been developed to allow inclusion of structures derived from MR imaging (MRI) for the external and internal tissue boundaries, thereby improving the accuracy and spatial resolution of the properties in each tissue type. The system has been validated in both tissue-simulating phantoms, with 10% accuracy observed, and in a rat cranium imaging experiment. The latter experiment used variation in inspired oxygen (FiO2) to vary the observed hemoglobin and oxygen saturation images. Quantitative agreement was observed between the changes in deoxyhemoglobin values derived from NIR and the changes predicted with blood-oxygen-level-dependent (BOLD) MRI. This system represents the initial stage in what will likely be a larger role for NIR tomography, coupled to MRI, and illustrates that the technological challenges of using continuous-wave broadband data and inclusion of a priori structural information can be met with careful phantom studies.

Algorithms↗

PubMed2005

Spectrally constrained chromophore and scattering near-infrared tomography provides quantitative and robust reconstruction.

A multispectral direct chromophore and scattering reconstruction technique has been implemented for near-infrared frequency-domain tomography in recovering images of total hemoglobin, oxygen saturation, water, and scatter parameters. The method applies the spectral constraint of the chromophores and scattering spectra directly in the reconstruction algorithm, thereby reducing the parameter space of the inversion process. This new method was validated by use of simulated and experimental data, and results show better robustness and stability in the presence of higher levels of noise. The method suppresses artifacts, especially those significant in water and scatter power images, and reduces cross talk between chromophore and scatter parameters. Variation in scattering was followed by this spectral approach successfully in experimental data from 90-mm-diameter cylindrical phantoms, and results show linear variation in scatter amplitude and reduced scattering coefficient (micro(s)'), with total hemoglobin, oxygen saturation, and water remaining constant and quantitatively accurate. Similar experiments were carried out for varying oxygen saturation and total hemoglobin. Accurate quantification was obtained with a mean error of 7.7% for oxygen saturation and 6.2% for total hemoglobin, with minimal cross talk between different parameters.

Algorithms↗

PubMed2005

Effects of refractive index on near-infrared tomography of the breast.

Near infrared (NIR) optical tomography is an imaging technique in which internal images of optical properties are reconstructed with the boundary measurements of light propagation through the medium. Recent advances in instrumentation and theory have led to the use of this method for the detection and characterization of tumors within the female breast tissue. Most image reconstruction approaches have used the diffusion approximation and have assumed that the refractive index of the breast is constant, with a bulk value of approximately 1.4. We have applied a previously reported modified diffusion approximation, in which the refractive index for different tissues can be modeled. The model was used to generate NIR data from a realistic breast geometry containing a localized anomaly. Using this simulated data, we have reconstructed optical images, both with and without correct knowledge of the refractive-index distribution to show that the modified diffusion approximation can accurately recover the anomaly given a priori knowledge of refractive index. But using a reconstruction algorithm without the use of correct a priori information regarding the refractive-index distribution is shown as recovering the anomaly but with a degraded quality, depending on the degree of refractive index mismatch. The results suggest that provided the refractive index of breast tissue is approximately 1.3-1.4, their exclusion will have minimal effect on the reconstructed images.

Algorithms↗

PubMed2005

Excitation patterns in three-dimensional electrical impedance tomography.

Electrical impedance tomography (EIT) is a non-invasive technique that aims to reconstruct images of internal electrical properties of a domain, based on electrical measurements on the periphery. Improvements in instrumentation and numerical modeling have led to three-dimensional (3D) imaging. The availability of 3D modeling and imaging raises the question of identifying the best possible excitation patterns that will yield to data, which can be used to produce the best image reconstruction of internal properties. In this work, we describe our 3D finite element model of EIT. Through singular value decomposition as well as examples of reconstructed images, we show that for a homogenous female breast model with four layers of electrodes, a driving pattern where each excitation plane is a sinusoidal pattern out-of-phase with its neighboring plane produces better qualitative images. However, in terms of quantitative imaging an excitation pattern where all electrode layers are in phase produces better results.

Algorithms↗

PubMed2005

Assimilating intraoperative data with brain shift modeling using the adjoint equations.

Biomechanical models of brain deformation are increasingly being used to nonrigidly register preoperative MR (pMR) images of the brain to the surgical scene. These model estimates can potentially be improved by incorporating sparse displacement data available in the operating room (OR), but integrating the intraoperative information with model calculations is a nontrivial problem. We present an inverse method to estimate the unknown boundary and volumetric forces necessary to achieve a least-squares fit between the model and the data that is formulated in terms of the adjoint equations, which are solved directly by the method of representers. The scheme is illustrated in a 2D simulation and in a 2D approximation based on a patient case using actual OR data.

Algorithms↗

PubMed2005

On optimal current patterns for electrical impedance tomography.

We develop a statistical criterion for optimal patterns in planar circular electrical impedance tomography. These patterns minimize the total variance of the estimation for the resistance or conductance matrix. It is shown that trigonometric patterns (Isaacson, 1986), originally derived from the concept of distinguishability, are a special case of our optimal statistical patterns. New optimal random patterns are introduced. Recovering the electrical properties of the measured body is greatly simplified when optimal patterns are used. The Neumann-to-Dirichlet map and the optimal patterns are derived for a homogeneous medium with an arbitrary distribution of the electrodes on the periphery. As a special case, optimal patterns are developed for a practical EIT system with a finite number of electrodes. For a general nonhomogeneous medium, with no a priori restriction, the optimal patterns for the resistance and conductance matrix are the same. However, for a homogeneous medium, the best current pattern is the worst voltage pattern and vice versa. We study the effect of the number and the width of the electrodes on the estimate of resistivity and conductivity in a homogeneous medium. We confirm experimentally that the optimal patterns produce minimum conductivity variance in a homogeneous medium. Our statistical model is able to discriminate between a homogenous agar phantom and one with a 2 mm air hole with error probability (p-value) 1/1000.

Algorithms↗

PubMed2005

Adaptive model initialization and deformation for automatic segmentation of T1-weighted brain MRI data.

A fully automatic, two-step, T1-weighted brain magnetic resonance imaging (MRI) segmentation method is presented. A preliminary mask of parenchyma is first estimated through adaptive image intensity analysis and mathematical morphological operations. It serves as the initial model and probability reference for a level-set algorithm in the second step, which finalizes the segmentation based on both image intensity and geometric information. The Dice coefficient and Euclidean distance between boundaries of automatic results and the corresponding references are reported for both phantom and clinical MR data. For the 28 patient scans acquired at our institution, the average Dice coefficient was 98.2% and the mean Euclidean surface distance measure was 0.074 mm. The entire segmentation for either a simulated or a clinical image volume finishes within 2 min on a modern PC system. The accuracy and speed of this technique allow us to automatically create patient-specific finite element models within the operating room on a timely basis for application in image-guided updating of preoperative scans.

Algorithms↗

PubMed2005

A three-parameter mechanical property reconstruction method for MR-based elastic property imaging.

A reconstruction process featuring full parameterization of the three dimensional, time-harmonic equations of linear elasticity is developed and reconstructed property images are presented from simulation-based investigation. While interesting in its own right through the potential for increased adaptability of these reconstructive elastic imaging techniques, this study also presents a set of analysis tools used to study the poor convergence behavior found in the case of tissue like conditions (i.e. nearly incompressible materials). The choice of elastic properties for imaging in elastography research remains an open question at this point; the use of the stability and sensitivity-based analytical methods described here will help to predict and understand the value and reliability of different parameterizations of elasticity imaging. Additionally, though results indicate significant work needs to be done to achieve effective multiparameter reconstructive imaging, the methods detailed here offer the promise of increased flexibility and sophistication in elastographic imaging techniques.

Algorithms↗

PubMed2005

Stereopsis-guided brain shift compensation.

Brain deformation models have proven to be a powerful tool in compensating for soft tissue deformation during image-guided neurosurgery. The accuracy of these models can be improved by incorporating intraoperative measurements of brain motion. We have designed and implemented a passive intraoperative stereo vision system capable of estimating the three-dimensional shape of the surgical scene in near real-time. This intraoperative shape is compared with the cortical surface in the co-registered preoperative magnetic resonance (MR) volume for the estimation of the cortical motion resulting from the open cranial surgery. The estimated cortical motion is then used to guide a full brain model, which updates a preoperative MR volume. We have found that the stereo vision system is accurate to within approximately 1 mm. Based on data from two representative clinical cases, we show that stereopsis guidance improves the accuracy of brain shift compensation both at and below the cortical surface.

Algorithms↗