Título
Image reconstruction in functional optical neuroimaging: the modelling and separation of the scalp blood flow
Autor
Javier Herrera-Vega
Colaborador
FELIPE ORIHUELA ESPINA (Asesor de tesis)
Nivel de Acceso
Acceso Abierto
Materias
Resumen o descripción
The reconstruction of brain haemodynamics from functional diffuse optical neuroimaging is severely affected by scalp blood flow. Correction of this artefact has received wide attention from the community with proposed solutions including variants of multidistance spectroscopy, spatial filtering, and concurrent physiological measurements. Here we investigate the less explored path of explicit parameterization of the forward modelling with the haemoglobin in the scalp, and face the challenge of inverting such model. It is hypothesized that incorporating scalp haemoglobin in the reconstruction process increases the sensitivity to gray matter haemoglobin (as compared to not doing so). This isn’t obvious since increasing the number of parameters of the model affects the conditioning of the problem as it is shown here. To ensure reconstruction remains feasible, we aim to put forth a solution based on the combination of a more detailed forward model, followed by a hybrid reconstruction algorithm. This hybrid reconstruction combines the precalculation of a colour map within a regularized Newton-based approximation in which the error estimation is based on Shampine’s optimized error function for algorithm stability. A realistic anatomical model of the adult human head is developed with finer detail considering the sublayers of the scalp. Optical properties of the head tissue are set from literatura or modelled according to existing theory when unavailable. Radiation transport is resolved using Monte Carlo method. The forward model is validated by matching and reconstructing broadband experimental spectra whilst checking that recovered parameters are within plausible physiological ranges. The results showed vectors of absolute concentrations of haemoglobins within plausible ranges. Model sensitivity was estimated and compared with existing models. Regarding inversion, we start by showing that naively upgrading the model number of parameters, impedes an otherwise monotonic behaviour. Then, the contribution of the haemoglobin in scalp is quantified in terms of contrast to noise ratio, and it is shown that the system becomes less stable when blood in the scalp is present. Finally, we carried out a study over the Cytochrome-c-oxidase as a marker of neural activity less affected by the blood in the scalp.
Editor
Instituto Nacional de Astrofísica, Óptica y Electrónica
Fecha de publicación
abril de 2018
Tipo de publicación
Tesis de doctorado
Versión de la publicación
Versión aceptada
Recurso de información
Formato
application/pdf
Idioma
Inglés
Audiencia
Estudiantes
Investigadores
Público en general
Sugerencia de citación
Herrera Vega, Javier., (2018). Image reconstruction in functional optical neuroimaging: the modelling and separation of the scalp blood flow. Tesis de Doctorado, Instituto Nacional de Astrofísica, Óptica y Electrónica
Repositorio Orígen
Repositorio Institucional del INAOE
Descargas
886