Background: Resting state functional magnetic resonance imaging (rs-fMRI) revealed the existence of spatially distributed and temporally correlated brain regions. As one of the many brain functions whose underlying mechanisms remain unclear, the brain ability to integrate polymodal information, store it, and retrieve it under specific circumstances suggests the existence of a functional organization above the modular level. In light of these considerations, the idea of large-scale brain networks nicely accommodate for this need. In this framework, the discovery that several cognitive networks display cerebellar components adds to the growing body of evidence that indicate a cerebellar contribution to cognitive functions (e.g. working memory, planning, and error detection). The current study investigated the changes in resting state networks, specifically their cerebellar components, in response to a continuous verbal/cognitive stimulus. Methods: MRI data acquisition was performed on 30 subjects using a 3T MR Philips Achieva (Philips Healthcare, Best, The Netherlands) with a 32-channel head coil. Four resting state sessions were acquired (Pre, Story, Post1, Post2), each 8.02 minutes long. In the second session (i.e. Story) the audio of an 8-minute-long story was presented binaurally to the subjects. A questionnaire measuring subjects’ attention was handed out at the end of the experiment. Results: An independent component analysis on the four temporally-concatenated sessions was performed. A total of 16 networks were identified. Session-specific mean functional connectivity and mean spatial extension were calculated for each network. With the exception of three networks, the Story session had the highest connectivity and the lowest spatial extension. Cerebellum regional functional connectivity was calculated. The mean functional connectivity, across those voxels in resting state during the Story session, was calculated in all four sessions and then subtracted from their Pre-session level. The difference in functional connectivity between the Story and the Pre-story session was 11.7%. When this global difference was distributed, proportionally to their spatial extension, among all networks with a cerebellar component, the DMN, OVN, RVAN, LN, and ECN were the networks with the greatest changes. These five networks were then added together and mapped onto a 28-region cerebellum atlas. For each of the 28 regions, the residuals in resting state activation between the Story session and the Pre-story session and between the Post1 session and the Pre-story session were calculated. In both contrasts, the greatest changes in resting state activation occurred in regions VI and VII (i.e. Crus I, Crus II, VIIB) of the cerebellum. Regions known for their activation during language processing, working memory tasks, and executive functions. The interpretations of these results along with their implications for future experiments are discussed in the last section of this project.
Background: L’utilizzo di tecniche di imaging di risonanza magnetica funzionale in resting state (rs-fMRI) ha rivelato l’esistenza di regioni cerebrali spazialmente distribuite e temporalmente correlate fra loro. Una fra le tante funzioni del cervello di cui ancora non conosciamo i meccanismi sottostanti è la sua abilità di integrare e gestire il flusso di informazioni polimodali, immagazzinarle, e richiamarle in determinate circostanze. L’esistenza di un’organizzazione funzionale più ampia di quella modulare sembra però un requisito necessario e l’identificazione di diverse network cerebrali che si estendono su larga scala ne è un ulteriore conferma. La scoperta che alcune network cognitive includano anche regioni del cervelletto corrobora l’ipotesi che il cervelletto abbia anche funzioni cognitive (e.g. working memory, planning e error detection). Il presente studio ha analizzato i cambiamenti delle network in resting state, e in particolare delle loro componenti cerebellari in funzione di uno stimolo verbale/cognitivo. Metodi: L’acquisizione dati è stata eseguita su un campione di 30 soggetti con l’utilizzo di un 3T MR Philips Achieva (Philips Healthcare, Best, The Netherlands) con un head coil a 32 canali. Un totale di quattro sessioni in resting state sono state acquisite (Pre, Storia, Post1, Post2), ognuna della durata di 8.02 minuti. Durante la seconda sessione (i.e. Storia) un audio-storia di 8 minuti è stata presentata al soggetto. Un questionario per misurare il livello di attenzione dei soggetti è stato somministrato al termine dell’esperimento. Risultati: Un’analisi delle componenti indipendenti eseguita sulle quattro sessioni concatenate temporalmente ha rivelato 16 network. Per ognuna delle 4 sessioni e per ognuna delle 16 network, la connettività funzionale media e l’estensione spaziale media sono state calcolate rivelando un andamento simile in 13 delle 16 network. Il picco della connettività si è registrato durante la storia in concomitanza con un drastico calo dell’estensione spaziale. La connettività funzionale media del cervelletto è stata calcolata sommando tutte le componenti cerebellari di ogni network. La connettività funzionale media, esclusivamente di quei voxel che erano in resting state durante la seconda sessione, è stata calcolata in tutte le quattro sessioni e poi sottratta alla connettività funzionale media della prima sessione. La differenza maggiore in connettività funzionale si è registrata nel contrasto Storia vs Pre-storia, con una differenza di connettività del 11,7%. Quando questa differenza è stata distribuita, in modo proporzionale all’estensione spaziale, fra tutte le network aventi una componente cerebellare. Il DMN, OVN, RVAN, LN, e l’ECN hanno mostrato il cambiamento maggiore. Le componenti cerebellari di queste cinque network sono state sommate e mappate su un atlas del cervelletto suddiviso in 28 regioni. Per ognuna delle 28 regioni, sono stati calcolati i residui di attivazione in resting state fra la sessione Storia e la sessione Pre-Storia e i residui fra la sessione Post1 e la sessione Pre-Storia. In entrambi i contrasti, i maggiori cambiamenti si sono verificati nelle regioni VI e VII (i.e. Crus I, Crus II, e VIIB) del cervelletto. Regioni note in letteratura per la loro attivazione durante la comprensione del linguaggio, task di working memory, e di funzioni esecutive. Le interpretazioni di questi risultati e le loro implicazioni per esperimenti futuri vengono discusse nella parte finale di questo lavoro.
Cerebellar involvement in dynamical cognitive circuitries: a resting state fMRI study
GRASSI, EMILIANO
2014/2015
Abstract
Background: Resting state functional magnetic resonance imaging (rs-fMRI) revealed the existence of spatially distributed and temporally correlated brain regions. As one of the many brain functions whose underlying mechanisms remain unclear, the brain ability to integrate polymodal information, store it, and retrieve it under specific circumstances suggests the existence of a functional organization above the modular level. In light of these considerations, the idea of large-scale brain networks nicely accommodate for this need. In this framework, the discovery that several cognitive networks display cerebellar components adds to the growing body of evidence that indicate a cerebellar contribution to cognitive functions (e.g. working memory, planning, and error detection). The current study investigated the changes in resting state networks, specifically their cerebellar components, in response to a continuous verbal/cognitive stimulus. Methods: MRI data acquisition was performed on 30 subjects using a 3T MR Philips Achieva (Philips Healthcare, Best, The Netherlands) with a 32-channel head coil. Four resting state sessions were acquired (Pre, Story, Post1, Post2), each 8.02 minutes long. In the second session (i.e. Story) the audio of an 8-minute-long story was presented binaurally to the subjects. A questionnaire measuring subjects’ attention was handed out at the end of the experiment. Results: An independent component analysis on the four temporally-concatenated sessions was performed. A total of 16 networks were identified. Session-specific mean functional connectivity and mean spatial extension were calculated for each network. With the exception of three networks, the Story session had the highest connectivity and the lowest spatial extension. Cerebellum regional functional connectivity was calculated. The mean functional connectivity, across those voxels in resting state during the Story session, was calculated in all four sessions and then subtracted from their Pre-session level. The difference in functional connectivity between the Story and the Pre-story session was 11.7%. When this global difference was distributed, proportionally to their spatial extension, among all networks with a cerebellar component, the DMN, OVN, RVAN, LN, and ECN were the networks with the greatest changes. These five networks were then added together and mapped onto a 28-region cerebellum atlas. For each of the 28 regions, the residuals in resting state activation between the Story session and the Pre-story session and between the Post1 session and the Pre-story session were calculated. In both contrasts, the greatest changes in resting state activation occurred in regions VI and VII (i.e. Crus I, Crus II, VIIB) of the cerebellum. Regions known for their activation during language processing, working memory tasks, and executive functions. The interpretations of these results along with their implications for future experiments are discussed in the last section of this project.È consentito all'utente scaricare e condividere i documenti disponibili a testo pieno in UNITESI UNIPV nel rispetto della licenza Creative Commons del tipo CC BY NC ND.
Per contatti: unitesi@unipv.it
https://hdl.handle.net/20.500.14239/8565