1,721,014 research outputs found
Towards Brain Tissue Microstructure Characterization using Diffusion MRI
No full textLa risonanza magnetica in diffusione è una delle uniche tecniche di imaging in grado di dare informazioni sulla struttura del cervello umano in-vivo. Tramite tecniche matematiche chiamate modelli di ricostruzione, dal segnale di diffusione, \`e possibile ricavare la funzione di distribuzione di probabilità dello spostamento delle molecole d'acqua in ogni voxel, detto propagatore. Dal propagatore è possibile ricavare informazioni riguardanti l'orientazione dei fasci di fibre neuronali, la densità di tali fibre e il calibro assonale, tramite il calcolo di particolari indici.Lo scopo principale di questa tesi è la caratterizzazione di tali indici, e in particolare la loro validazione tramite, sia simulazioni al computer, sia dati in-vivo. In particolare ci siamo concentrati sugli indici calcolati utilizzando tre modelli di ricostruzione: il DTI, il 3D-SHORE e il MAPMRI. Il primo contributo di questa tesi riguarda il calcolo e il confronto dei valori degli indici del propagatore per tali modelli. Il secondo contributo della tesi è lo studio della variazione degli indici rispetto alle principali variazioni microstrutturali, tipiche della materia bianca. Nella terza parte della tesi abbiamo proposto un nuovo modello di ricostruzione in grado di dare risultati pi\`u accurati nel caso di incrocio di fibre. Un'accurata rappresentazione degli incroci è fondamentale visto che rappresentano la maggior parte della materia bianca del cervello umano.L'ultimo contributo della tesi è lo sviluppo di un nuovo modello di ricostruzione del propagatore in grado di modellare acquisizioni con gradienti a tempi di diffusione multipli. I risultati mostrano come gli indici del propagatore siano estremamente sensibili a variazioni microstrutturali della densità e dell'orientazione degli assoni della materia bianca. Gli indici sono risultati invece insensibili a variazioni del diametro assonale, a causa del lento decadimento del segnale, che richiederebbe acquisizioni effettuate con campi magnetici estremamente elevati. I nuovi modelli di ricostruzione proposti hanno dato eccellenti risultati sia nella modellazione degli incroci di fibre che del segnale a tempi di diffusione multipli.Diffusion magnetic resonance imaging is one of the only non-invasive imaging technique which is able to provide information on the human brain structure in-vivo.From the diffusion signal, using mathematical techniques called reconstruction models, it is possible to retrieve the probability density function of the water molecules displacement, or ensemble average propagator (EAP). From the EAP, it is possible to calculate a series of indices which provide information regarding the fiber orientation, fiber density, and the average diameter of the axons. The main aim of this thesis is the characterization of these indices, and, in particular, their validation. In order to characterize the indices, we take advantage of computer simulation of diffusion in different media, as well as human brain acquisition. In particular, we focused on the EAP indices calculated using three EAP models: the DTI, the 3D-SHORE, and the MAPMRI. The first contribution of this thesis is the developing and the comparison of the values of the indices for the different models.The second contribution of the thesis is the study of the variation of the indices with respect to the principal microstructural parameters which characterize the white matter. The third contribution of the thesis is the proposal of a new reconstruction model designed to reconstruct accurately the EAP in the case of crossing fibers. The fourth contributions is the developing of a new tensor model, which is able to capture the dependence on the timing parameters of the diffusion signal.Results show the sensibility of the EAP-derived indices to microstructural variations such as the orientation dispersion of the axons and the density of the fibers. Diameter axons variation, on the contrary, are not measurable by the EAP indices because of the slow signal decay, which would require extremely high magnetic fields to be measured. The new reconstruction models proposed provide excellent results in the modeling of crossing fibers and multiple diffusion times, respectively
A new tensor model for the measurement of diffusional anisotropy due to restricted diffusion
No full textDiffusion Tensor Imaging (DTI, Basser et al 1994) is the most widely used technique to measure the diffusion properties of the human brain tissues in-vivo. Although datasets comprising 30 samples, all but one of which acquired at a single b-value, are enough to estimate the diffusion tensor, in recent yearsnumerous acquisition schemes featuring more general diffusion encodings have been proposed. In order to vary the b-value, it is necessary to change either the gradient strength (G) or the duration or separation of the diffusion pulses. In the case of unrestricted diffusion (e.g. Gaussian), the b-value is theonly parameter that determines the diffusion-weighting. Thus, the same level of diffusion sensitization, hence the same signal attenuation, can be obtained by varying the gradient strength or the effective diffusion time (tau). In more realistic cases of diffusion, e.g., within restricted media (such as cells)changing G and tau have different effects on the signal. In this work, we compared the effect of changing the tau and the gradient strength G on the reconstruction error, using two tensor models: DTI, which is appropriate only for unrestricted diffusion, and the Diffusion Imaging with Confinement Tensor(DICT, Yolcu et al., Afzali et al 2015), which is applicable to both restricted and unrestricted diffusion scenarios. singthe either model, it is possible to estimate clinically important features, such as the Mean Diffusivity, and the Fractional Anisotropy (Afzali et al 2015, Pierpaoli and Basser 1996)
Comparison between discrete and continuous propagator indices from Cartesian q-space DSI sampling
No full textDSI1 is often considered the state-of-the-art technique to analyze q-space measurements sampled from a Cartesian grid. The 3D fastFourier transform is used to directly obtain a discrete version of the EAP (Ensemble Average Propagator). DSI was one of the first techniques used toinfer complex fiber configurations as it allows resolving crossings. In principle, DSI also captures some radial information which, in theory, can be used toextract diffusion features of the EAP. However, a discrete propagator representation suffers from a limited frequency band, which makes infiniteintegration impossible. Hence, EAP derived indices2,3 are problematic and quantitatively questionable, as one needs to artificially normalize andapproximate the infinite integrals. Combined with the recent popularity of DSI in the Human Connectome Project4, it is important to investigate thedifferent angular and EAP indices that can be computed from these DSI datasets. In this work, we investigate alternatives to the discrete model-freeapproach of DSI and investigate the Simple Harmonic Oscillator based Reconstruction and Estimation3 (SHORE) models based on the evaluation of (i)the orientation distribution function (ODF) ; (ii) the return to the origin probability2,3 (RTOP) and (iii) the mean square displacement3 (MSD)
Human brain tissue microstructure characterization using 3D-SHORE on the HCP data
No full textThe Human Connectome Project (HCP)1 data is composed of high resolution multi-shell diffusion weighted imaging acquisitions.
Continuous analytical reconstruction models, such as the 3D Simple Harmonic Oscillator based Reconstruction and Estimation2 (3D-SHORE), are able
to process multi-shell acquisition natively and give an analytical representation of the Ensemble Average Propagator (EAP). From the EAP, it is then
possible to retrieve information about the water molecules displacement like the Orientation Distribution Function (ODF), necessary to perform brain
tractography, but also other scalar indices that can provide accurate estimation of microstructural properties of the brain tissues. A large number of
studies have been conducted on the HCP data targeting the extraction of fibers orientation information, while the assessment of tissue microstructural
properties on this data is still unexplored. In this work, microstructural features are inferred from the reconstruction of the HCP data using the 3DSHORE
model. The related numerical measures3 are: i) Return To the Origin Probability (RTOP), ii) Return To the Axis Probability (RTAP) and iii)
Return To the Plane Probability (RTPP), which are calculated and used for assessing the potential of multi-shell acquisitions for characterizing human
brain tissues
Detection of fiber crossings and fannings using MultiTensor Distribution Model
No full textPURPOSE Tractography algorithms ground on the Orientation Distribution Function 1 (ODF) estimation in each voxel, in order to locally extract the principal directions of diffusion. Although ODF peaks correspond to the white matter bundles main orientation, the peaks alone are not able to represent complex fiber patterns such as fanning. In this work, we propose a new method called MultiTensor Distribution Model (MTDM) which, starting from a single fiber response modeled as a diffusion tensor and an initial estimation of the principal diffusion direction, it is able to estimate the dispersion of the white matter bundles in each voxel
An evolutionary procedure for inferring MP systems regulation functions of biological networks
No full textMetabolic P systems are a modeling framework for metabolic, regulatory and signaling processes. The key point of MP systems are flux regulation functions, which determine the evolution of a system from a given initial state. This paper presents important improvements to a technique, based on genetic algorithms and multiple linear regression, for inferring regulation functions that reproduce observed behaviors (time series datasets). An accurate analysis of three case studies, namely the mitotic oscillator in early amphibian embryos, the Lodka–Volterra predator-prey model and the chaotic logistic map show that this methodology can provide, from observed data, significant knowledge about the regulation mechanisms underlying biological processes
Waveform decoding and detection in hdEEG
No full textBrain oscillations are very powerful descriptors of both physiological and pathological brain states. In general, EEG signals consist of complex mixtures of components whose characterization provides reliable information about the neuronal activity. This study is inspired to the {\em consensus matching pursuit} (CMP) representation and proposes an effective method for the detection and modeling of interictal prototypical signal patterns in temporal lobe epilepsy. CMP allows accounting for inter-trial variability in temporal jitter, frequency and number of oscillations. In this work, we propose to generalize the approach and exploit the resulting spike representation for automatic interictal spike detection. Performance was evaluated on both synthetic and real high density EEG signals. Results show high sensitivity and specificty in spike detection as well as an accurate separation in the transient and oscillation components
Characterization of diffusion MRI signal non Gaussianity using MAPMRI
No full textPURPOSE Diffusion in restricted media, such as the neuron axons, is nonGaussian 1. Mean Apparent Propagator Magnetic Resonance Imaging (MAPMRI 2,3 ) is a reconstruction model for diffusion MRI which is able to estimating the nonGaussianity 2 (NG) of the diffusion signal. This study aims at investigating the minimum requirements of a diffusion weighted acquisition, in terms of Signal to Noise Ratio (SNR) and maximum bvalue, for MAPMRI to capture the NG of the signal. METHODS We used Camino (http://camino.cs.ucl.ac.uk/) MonteCarlo to simulate the diffusion signal inside a pack of parallel cylinders oriented along the z axis (100000 spins 4 , 1000 timesteps 4, radius 0.5μm, and 0.1μm of space between the cylinders). We acquired the diffusion signal using 10 different sampling schemes, Δ=57.9ms, δ=13.8ms, TE=91.3ms, and considering 10 b 0, 60 directions at bvalue= 700s/mm 2 ( first shell), and 60 directions at bvalues 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, and 10000 s/mm 2, r espectively, per sampling scheme (second shell). We fit both Diffusion Tensor Imaging 5 (DTI) and MAPMRI to the simulated data and the NG index was computed using MAPMRI. Successively, the Mean Squared Error (MSE) between ground truth and fitted signal was calculated for both DTI and MAPMRI. The same analyses were performed also adding Rician Noise to the diffusion signal with SNR=[40, 30, 20] considering 100 different instances of noise per sampling scheme. The MSE in this case was calculated with respect to the noiseless ground truth signal
From time series to biological network regulations: an evolutionary approach
No full textIn this paper we present a new methodology, based on genetic algorithms and multiple linearregression, for discovering regulation mechanisms responsible for observed time series inbiological networks. The modeling framework employed is called Metabolic P systems; they aredeterministic and time-discrete dynamical systems proposed as an effective alternative to ordinarydifferential equations for modeling biochemical systems. Our methodology is here successfullyapplied to the mitotic oscillator in early amphibian embryos. Starting from the time series ofsubstances involved in this system, we are able to reconstruct an MP system reproducing theobserved dynamics, where the regulatory components were discovered by our evolutionarymethodology. In particular, genetic algorithms are used as a variable selection technique toidentify the best representation of any regulation function in terms of some given primitivefunctions
- …
