Hemodynamics reflects the interaction between local metabolic needs and systemic regulation, and can be used as a marker for tissue health and metabolic function, with applications from physiology to neuroscience. Despite their importance, processes underlying human hemodynamics remain difficult to investigate due to the invasiveness or high costs of conventional measurement methods. Within this context, diffuse optical techniques, particularly Time-Domain (TD)- Near-Infrared Spectroscopy (NIRS), offer an effective tool for probing human hemodynamics in vivo. This thesis explores how TD-NIRS can be exploited to characterize hemodynamics across different tissues, i.e., the human brain and skeletal muscle, and experimental paradigms, combining numerical simulations, instrumental development, and in-vivo studies. First, this work shows that TD-NIRS can resolve spontaneous cerebral hemodynamic oscillations and their interaction with the systemic physiological rhythms in a depth-resolved manner. Both simulations and experimental measurements demonstrate that oscillatory components associated with multiple mechanisms of vascular control can be detected, and that cerebral and extracerebral contributions can be independently reconstructed from TD-NIRS data. Subsequently, a methodological limitation of depth-resolved TD-NIRS, the dependence of reconstructed signals on the model assumed for the probed tissue, is addressed. Results show that inaccuracies in the tissue model can significantly distort the frequency-domain features of reconstructed hemodynamic signals. To address this limitation, a method is proposed that exploits these distortions to optimize the tissue model geometry directly from TD-NIRS data acquired using specifically designed functional tasks. This thesis then expands frequency-domain analysis frameworks to skeletal muscle applications, demonstrating that TD-NIRS can capture both basal vascular oscillations and adaptations to increased metabolic demand during sustained exercise. Distinctions across groups differing in age and training levels are also noted, indicating different microvascular control strategies. Finally, applications of Continuous-Wave (CW)-NIRS in developmental and social neuroscience were investigated during a secondment abroad. The work carried out during this period is presented in the last chapter, which explores a novel method for modeling atypical and superimposed hemodynamic responses to functional tasks, a common challenge in developmental neuroscience contexts, and discusses the reliability of NIRS-based hyperscanning measurements. As a general outcome, this thesis advances the experimental methodology and applications of diffuse optical spectroscopy and contributes to the state-of-the-art in NIRS studies investigating human hemodynamics across tissues and research domains.
L’emodinamica riflette l’interazione tra le esigenze metaboliche locali e i meccanismi di regolazione sistemica e può essere utilizzata come indicatore dello stato di salute dei tessuti e della loro funzionalità metabolica, con applicazioni che spaziano dalla fisiologia alle neuroscienze. Nonostante la sua rilevanza, lo studio dei processi che regolano l’emodinamica umana è tuttora complesso, a causa dell’invasività o degli elevati costi delle tecniche di misura tradizionali. In questo contesto, le tecniche ottiche diffuse, e in particolare la spettroscopia nel vicino infrarosso tempo-risolta (Time-Domain Near-Infrared Spectroscopy, TD-NIRS), rappresentano uno strumento efficace per indagare in vivo l’emodinamica umana. Questa tesi esplora il potenziale della TD-NIRS per la caratterizzazione dell’emodinamica in diversi tessuti, in particolare cervello e muscolo scheletrico, e in vari contesti sperimentali, combinando simulazioni numeriche, sviluppo strumentale e studi in vivo. In primo luogo, viene dimostrato che la TD-NIRS è in grado di risolvere le oscillazioni emodinamiche cerebrali spontanee e la loro interazione con i ritmi fisiologici sistemici con risoluzione in profondità. Le simulazioni e le misure sperimentali presentate evidenziano che è possibile rilevare componenti oscillatorie associate a diversi meccanismi di controllo vascolare e ricostruire in modo indipendente i contributi cerebrali ed extracerebrali a partire da dati TD-NIRS. Successivamente, viene affrontata una limitazione metodologica della TD-NIRS, ovvero la dipendenza dei segnali ricostruiti dal modello di tessuto assunto. I risultati mostrano che imprecisioni nel modello possono introdurre distorsioni significative nelle caratteristiche in frequenza dei segnali emodinamici ricostruiti e associati ai vari strati che compongono il tessuto in analisi. Per superare questo limite, viene proposto un metodo che sfrutta tali distorsioni per ottimizzare direttamente la geometria del modello tissutale a partire da dati TD-NIRS acquisiti durante compiti funzionali appositamente progettati. La tesi estende poi i metodi di analisi nel dominio della frequenza alle applicazioni sul muscolo scheletrico, dimostrando che la TD-NIRS è in grado di rilevare sia le oscillazioni vascolari basali sia gli adattamenti alla maggiore richiesta metabolica durante l’esercizio prolungato. Vengono inoltre osservate differenze tra gruppi distinti per età e livello di allenamento, suggerendo strategie differenti di controllo microvascolare. Infine, durante un periodo di ricerca all’estero, sono state investigate applicazioni della NIRS in onda continua (Continuous-Wave, CW-NIRS) nell’ambito delle neuroscienze dello sviluppo e sociali. Il lavoro svolto è presentato nell’ultimo capitolo, che introduce un nuovo metodo per modellare risposte emodinamiche funzionali atipiche e/o sovrapposte, problematica frequente negli studi di neuroscienze dello sviluppo, e discute l’affidabilità delle misure di hyperscanning basate su dati NIRS. Nel complesso, questa tesi contribuisce all’avanzamento delle metodologie sperimentali e delle applicazioni della spettroscopia ottica diffusa, apportando nuovi elementi allo stato dell’arte degli studi NIRS sull’emodinamica umana in diversi tessuti e ambiti di ricerca.
Advancing near-infrared spectroscopy for measuring human hemodynamics across domains
Contini, Letizia
2025/2026
Abstract
Hemodynamics reflects the interaction between local metabolic needs and systemic regulation, and can be used as a marker for tissue health and metabolic function, with applications from physiology to neuroscience. Despite their importance, processes underlying human hemodynamics remain difficult to investigate due to the invasiveness or high costs of conventional measurement methods. Within this context, diffuse optical techniques, particularly Time-Domain (TD)- Near-Infrared Spectroscopy (NIRS), offer an effective tool for probing human hemodynamics in vivo. This thesis explores how TD-NIRS can be exploited to characterize hemodynamics across different tissues, i.e., the human brain and skeletal muscle, and experimental paradigms, combining numerical simulations, instrumental development, and in-vivo studies. First, this work shows that TD-NIRS can resolve spontaneous cerebral hemodynamic oscillations and their interaction with the systemic physiological rhythms in a depth-resolved manner. Both simulations and experimental measurements demonstrate that oscillatory components associated with multiple mechanisms of vascular control can be detected, and that cerebral and extracerebral contributions can be independently reconstructed from TD-NIRS data. Subsequently, a methodological limitation of depth-resolved TD-NIRS, the dependence of reconstructed signals on the model assumed for the probed tissue, is addressed. Results show that inaccuracies in the tissue model can significantly distort the frequency-domain features of reconstructed hemodynamic signals. To address this limitation, a method is proposed that exploits these distortions to optimize the tissue model geometry directly from TD-NIRS data acquired using specifically designed functional tasks. This thesis then expands frequency-domain analysis frameworks to skeletal muscle applications, demonstrating that TD-NIRS can capture both basal vascular oscillations and adaptations to increased metabolic demand during sustained exercise. Distinctions across groups differing in age and training levels are also noted, indicating different microvascular control strategies. Finally, applications of Continuous-Wave (CW)-NIRS in developmental and social neuroscience were investigated during a secondment abroad. The work carried out during this period is presented in the last chapter, which explores a novel method for modeling atypical and superimposed hemodynamic responses to functional tasks, a common challenge in developmental neuroscience contexts, and discusses the reliability of NIRS-based hyperscanning measurements. As a general outcome, this thesis advances the experimental methodology and applications of diffuse optical spectroscopy and contributes to the state-of-the-art in NIRS studies investigating human hemodynamics across tissues and research domains.| File | Dimensione | Formato | |
|---|---|---|---|
|
PhD_Thesis_Contini_Letizia.pdf
non accessibile
Descrizione: Tesi di Dottorato
Dimensione
51.42 MB
Formato
Adobe PDF
|
51.42 MB | 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/10589/255438