Diffusion Magnetic Resonance Imaging (dMRI) is a non-invasive modality that probes the microstructural properties of biological tissues, providing valuable insights into cellular architecture and tissue integrity. This thesis investigates the application of advanced diffusion encoding schemes for the detection of early microstructural changes associated with Alzheimer’s Disease (AD), with a particular focus on axonal pathology. An ex vivo protocol was implemented, combining Triple Diffusion Encoding (TDE) and Single Diffusion Encoding (SDE) at high b-values, and results were compared with those obtained using Diffusion Tensor Imaging (DTI) fitted on Pulsed Gradient Spin Echo (PGSE) data. The TDE–SDE approach enables estimation of intra-axonal diffusivity, axonal signal fraction, and microscopic anisotropy with fewer modeling assumptions. It uses high b-values to suppress extracellular signals and applies the spherical mean technique to eliminate confounding effects from orientation dispersion and multiple fiber orientations—common sources of error in DTI metrics. Experiments were conducted on TgF344-AD rats, a transgenic model of the disease, at two progression stages (4 and 17 months). The extracted microstructural parameters were statistically analyzed to assess their sensitivity to pathological changes. Results show that the TDE–SDE protocol outperforms DTI in detecting intra-axonal alterations, revealing significant differences in axonal diffusivity as early as 4 months. These findings support the use of advanced diffusion encoding techniques as non-invasive, sensitive, and potentially specific biomarkers for the early diagnosis of Alzheimer’s disease in preclinical models.
La Risonanza Magnetica di Diffusione (dMRI) è una tecnica non invasiva che esplora le proprietà microstrutturali dei tessuti biologici, fornendo informazioni sull’architettura cellulare e sull’integrità tissutale. Questa tesi analizza l’impiego di schemi avanzati di codifica della diffusione per l’individuazione precoce di alterazioni microstrutturali associate alla Malattia di Alzheimer (AD), con particolare attenzione alla patologia assonale. È stato utilizzato un protocollo ex vivo combinando Codifica Tripla di Diffusione (TDE) e Codifica Singola di Diffusione (SDE) a elevati valori di b, confrontando i risultati con quelli ottenuti tramite Diffusion Tensor Imaging (DTI) basato su sequenze Pulsed Gradient Spin Echo (PGSE). L'approccio TDE–SDE consente di stimare la diffusività intra-assonale, la frazione di segnale assonale e l'anisotropia microscopica con un numero ridotto di assunzioni modellistiche. Utilizza alti valori di b per sopprimere il segnale extracellulare e applica la tecnica della media sferica per eliminare gli effetti confondenti dovuti alla dispersione dell'orientamento e alla presenza di più orientazioni delle fibre, che spesso compromettono l'accuratezza delle metriche della DTI. Gli esperimenti sono stati condotti su ratti TgF344-AD, un modello transgenico della malattia, in due fasi di progressione (4 e 17 mesi). I parametri microstrutturali estratti sono stati analizzati statisticamente per valutarne la sensibilità ai cambiamenti patologici. I risultati mostrano che il protocollo TDE–SDE supera il DTI nel rilevare alterazioni intra-assonali, evidenziando differenze significative nella diffusività assonale già a 4 mesi. Questi dati supportano l’uso di tecniche avanzate di codifica della diffusione come biomarcatori non invasivi, sensibili e potenzialmente specifici per la diagnosi precoce dell’Alzheimer in modelli preclinici.
Directly encoding morphological features of microstructure anatomy into the diffusion MRI signal
CAPPOZZO, VITTORIA
2024/2025
Abstract
Diffusion Magnetic Resonance Imaging (dMRI) is a non-invasive modality that probes the microstructural properties of biological tissues, providing valuable insights into cellular architecture and tissue integrity. This thesis investigates the application of advanced diffusion encoding schemes for the detection of early microstructural changes associated with Alzheimer’s Disease (AD), with a particular focus on axonal pathology. An ex vivo protocol was implemented, combining Triple Diffusion Encoding (TDE) and Single Diffusion Encoding (SDE) at high b-values, and results were compared with those obtained using Diffusion Tensor Imaging (DTI) fitted on Pulsed Gradient Spin Echo (PGSE) data. The TDE–SDE approach enables estimation of intra-axonal diffusivity, axonal signal fraction, and microscopic anisotropy with fewer modeling assumptions. It uses high b-values to suppress extracellular signals and applies the spherical mean technique to eliminate confounding effects from orientation dispersion and multiple fiber orientations—common sources of error in DTI metrics. Experiments were conducted on TgF344-AD rats, a transgenic model of the disease, at two progression stages (4 and 17 months). The extracted microstructural parameters were statistically analyzed to assess their sensitivity to pathological changes. Results show that the TDE–SDE protocol outperforms DTI in detecting intra-axonal alterations, revealing significant differences in axonal diffusivity as early as 4 months. These findings support the use of advanced diffusion encoding techniques as non-invasive, sensitive, and potentially specific biomarkers for the early diagnosis of Alzheimer’s disease in preclinical models.File | Dimensione | Formato | |
---|---|---|---|
2025_07_Cappozzo_Tesi.pdf
accessibile in internet per tutti
Descrizione: Testo tesi
Dimensione
12.28 MB
Formato
Adobe PDF
|
12.28 MB | Adobe PDF | Visualizza/Apri |
2025_07_Cappozzo_Executive Summary.pdf
accessibile in internet per tutti
Descrizione: Executive summary
Dimensione
632.22 kB
Formato
Adobe PDF
|
632.22 kB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/240405