Implementation of a computationally efficient least-squares algorithm for highly under-determined three-dimensional diffuse optical tomography problems.

Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire 03755, USA.
Medical Physics (Impact Factor: 2.91). 06/2008; 35(5):1682-97. DOI: 10.1118/1.2889778
Source: PubMed

ABSTRACT Three-dimensional (3D) diffuse optical tomography is known to be a nonlinear, ill-posed and sometimes under-determined problem, where regularization is added to the minimization to allow convergence to a unique solution. In this work, a generalized least-squares (GLS) minimization method was implemented, which employs weight matrices for both data-model misfit and optical properties to include their variances and covariances, using a computationally efficient scheme. This allows inversion of a matrix that is of a dimension dictated by the number of measurements, instead of by the number of imaging parameters. This increases the computation speed up to four times per iteration in most of the under-determined 3D imaging problems. An analytic derivation, using the Sherman-Morrison-Woodbury identity, is shown for this efficient alternative form and it is proven to be equivalent, not only analytically, but also numerically. Equivalent alternative forms for other minimization methods, like Levenberg-Marquardt (LM) and Tikhonov, are also derived. Three-dimensional reconstruction results indicate that the poor recovery of quantitatively accurate values in 3D optical images can also be a characteristic of the reconstruction algorithm, along with the target size. Interestingly, usage of GLS reconstruction methods reduces error in the periphery of the image, as expected, and improves by 20% the ability to quantify local interior regions in terms of the recovered optical contrast, as compared to LM methods. Characterization of detector photo-multiplier tubes noise has enabled the use of the GLS method for reconstructing experimental data and showed a promise for better quantification of target in 3D optical imaging. Use of these new alternative forms becomes effective when the ratio of the number of imaging property parameters exceeds the number of measurements by a factor greater than 2.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We evaluated the potential of mesh-based Monte Carlo (MC) method for widefield time-gated fluores-cence molecular tomography, aiming to improve accuracy in both shape discretization and photon transport mod-eling in preclinical settings. An optimized software platform was developed utilizing multithreading and distributed parallel computing to achieve efficient calculation. We validated the proposed algorithm and software by both simulations and in vivo studies. The results establish that the optimized mesh-based Monte Carlo (mMC) method is a computationally efficient solution for optical tomography studies in terms of both calculation time and memory utilization.
    Journal of Biomedical Optics 10/2012; 17(10):106009. · 2.75 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To optimize the data-collection strategy for diffuse optical tomography and to obtain a set of independent measurements among the total measurements using the model based data-resolution matrix characteristics. The data-resolution matrix is computed based on the sensitivity matrix and the regularization scheme used in the reconstruction procedure by matching the predicted data with the actual one. The diagonal values of data-resolution matrix show the importance of a particular measurement and the magnitude of off-diagonal entries shows the dependence among measurements. Based on the closeness of diagonal value magnitude to off-diagonal entries, the independent measurements choice is made. The reconstruction results obtained using all measurements were compared to the ones obtained using only independent measurements in both numerical and experimental phantom cases. The traditional singular value analysis was also performed to compare the results obtained using the proposed method. The results indicate that choosing only independent measurements based on data-resolution matrix characteristics for the image reconstruction does not compromise the reconstructed image quality significantly, in turn reduces the data-collection time associated with the procedure. When the same number of measurements (equivalent to independent ones) are chosen at random, the reconstruction results were having poor quality with major boundary artifacts. The number of independent measurements obtained using data-resolution matrix analysis is much higher compared to that obtained using the singular value analysis. The data-resolution matrix analysis is able to provide the high level of optimization needed for effective data-collection in diffuse optical imaging. The analysis itself is independent of noise characteristics in the data, resulting in an universal framework to characterize and optimize a given data-collection strategy.
    Medical Physics 08/2012; 39(8):4715-25. · 2.91 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: To improve object depth-localization for diffuse optical tomography (DOT) in a circular-array outward-imaging geometry that is subjected to strong sensitivity variation with respect to imaging depth.Methods: The authors introduce an alternative DOT image reconstruction approach that optimizes the data-model fit based on the paired measurements corresponding to two pairs of source-detector that share either the source or the detector, in comparison to the conventional method that optimizes the data-model fit based on the unpaired measurements corresponding to individual pairs of source-detector. This alternative approach, namely, geometric-sensitivity-difference (GSD) method, effectively reduces the variation of the reconstruction sensitivity with respect to imaging depth. The DOT image reconstruction based on GSD-scheme applied to same-source source-detector pairs is demonstrated using simulated and experimental continuous-wave measurements in a circular-array outward-imaging geometry, of which the native sensitivity varies strongly with respect to the depth. The outcomes of GSD-based image reconstruction are compared to those of two other methods: one is the conventional baseline method that utilizes the native sensitivity but does not involve depth-compensating scheme; and the other is a reference-compensation approach that employs active and depth-adapted compensation scheme to counteract the dependence of the reconstruction sensitivity with respect to imaging depth.Results: The GSD method generally outperforms the other two methods in localizing the depth of single object, resolving two objects that are azimuthally separated, and estimating the optical property of single object or azimuthally separated dual objects. The GSD method, however, demands more computations due to an increase of the element size of the resulted sensitivity matrix and more matrix multiplications.Conclusions: The GSD method improves the depth localization in the circular-array outward-imaging geometry, by taking advantage of the paired measurements of two source-sharing source-detector-pairs to passively and effectively homogenize the sensitivity of the reconstruction with respect to imaging depth.
    Medical Physics 01/2013; 40(1):013101. · 2.91 Impact Factor


1 Download
Available from