29 February 2012 Construction of the Jacobian matrix for fluorescence diffuse optical tomography using a perturbation Monte Carlo method
Author Affiliations +
Proceedings Volume 8216, Multimodal Biomedical Imaging VII; 82160O (2012) https://doi.org/10.1117/12.906439
Event: SPIE BiOS, 2012, San Francisco, California, United States
Image formation in fluorescence diffuse optical tomography is critically dependent on construction of the Jacobian matrix. For clinical and preclinical applications, because of the highly heterogeneous characteristics of the medium, Monte Carlo methods are frequently adopted to construct the Jacobian. Conventional adjoint Monte Carlo method typically compute the Jacobian by multiplying the photon density fields radiated from the source at the excitation wavelength and from the detector at the emission wavelength. Nonetheless, this approach assumes that the source and the detector in Green's function are reciprocal, which is invalid in general. This assumption is particularly questionable in small animal imaging, where the mean free path length of photons is typically only one order of magnitude smaller than the representative dimension of the medium. We propose a new method that does not rely on the reciprocity of the source and the detector by tracing photon propagation entirely from the source to the detector. This method relies on the perturbation Monte Carlo theory to account for the differences in optical properties of the medium at the excitation and the emission wavelengths. Compared to the adjoint methods, the proposed method is more valid in reflecting the physical process of photon transport in diffusive media and is more efficient in constructing the Jacobian matrix for densely sampled configurations.
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiaofeng Zhang, Xiaofeng Zhang, } "Construction of the Jacobian matrix for fluorescence diffuse optical tomography using a perturbation Monte Carlo method", Proc. SPIE 8216, Multimodal Biomedical Imaging VII, 82160O (29 February 2012); doi: 10.1117/12.906439; https://doi.org/10.1117/12.906439

Back to Top