Resting-State Functional MRI Approaches to Parkinsonisms and Related Dementia

Purpose of the Review In this review, we attempt to summarize the most updated studies that applied resting-state functional magnetic resonance imaging (rs-fMRI) in the field of Parkinsonisms and related dementia. Recent Findings Over the past decades, increasing interest has emerged on investigating the presence and pathophysiology of cognitive symptoms in Parkinsonisms and their possible role as predictive biomarkers of neurodegenerative brain processes. In recent years, evidence has been provided, applying mainly three methodological approaches (i.e. seed-based, network-based and graph-analysis) on rs-fMRI data, with promising results. Summary Neural correlates of cognitive impairment and dementia have been detected in patients with Parkinsonisms along the diseases course. Interestingly, early functional connectivity signatures were proposed to track and predict future progression of neurodegenerative processes. However, longitudinal studies are still sparce and further investigations are needed to overcome this knowledge gap.


Introduction
Cognitive impairment has been increasingly identified as a relevant condition for the majority of patients with Parkinsonism, potentially presenting from the earliest disease phases to the advanced stages [1].
Parkinsonisms are heterogeneous neurodegenerative disorders characterised by clinical parkinsonian features that may be differently associated with other motor and nonmotor symptoms [2].The deposition of aggregated proteins into intracellular inclusion bodies is a common neuropathological denominator for these disorders, with pathological changes typically spreading into the brain over specific anatomical patterns that are characteristic for each disease.
Tau and α-synuclein are the most abundant proteins that may be found in pathological aggregates occurring typically in the presynaptic and axonal portion of neurons, but also in glial cells [1,2].
Abnormal aggregates of α-synuclein, such as Lewy bodies (LB) and Lewy neurites have been indeed implicated in the pathophysiology of Parkinson's disease (PD), Dementia with Lewy bodies (DLB) and Multiple systems atrophy (MSA), thereby leading to the umbrella term of synucleinopathies [1,2].
However, the watershed behind this classification is blurred and mixed pathological forms have been recognized [3].Similarly, overlapping clinical manifestations between synucleinopathies and tauopathies make the differential diagnosis challenging in the spectrum of neurodegenerative diseases [2,3].Cognitive impairment is common and heterogenous in PD, often extending as a continuum from subjective cognitive impairment to mild cognitive impairment (MCI) and dementia (PDD) [4].
The early pattern of PD-related cognitive dysfunction mainly involves executive and attentive/working memory domains with a slow progression over time.Memory, language and visuospatial impairments usually occur later in the disease course and have been associated to an increased risk to develop clinical overt dementia [1,4].This parallels with the "dual syndrome hypothesis", that pictures the dopaminergic failure of frontostriatal connections as occurring early in the disease, leading to executive deficits with mild progression over time, whereas the involvement of the cholinergic system could determine a more "posterior" cognitive dysfunction, with higher risk to convert into dementia [1,4].
Interestingly, along with motor (i.e.postural and gait disturbances) and demographic features (i.e.male sex, older age at onset), the presence of specific nonmotor symptoms (i.e.REM sleep behavioural disorders, neuropsychiatric symptoms, autonomic dysfunction) has been suggested to increase the risk of developing dementia in PD [1,4].
While cognitive impairment in PD occurs usually later in the disease course with respect to motor symptoms, DLB should be diagnosed when dementia occurs before or concurrently with parkinsonism, with prominent hallucinations and visuospatial dysfunction [5].
MSA is clinically characterized by a variable combination of autonomic dysfunction, levodopa-unresponsive parkinsonism, and cerebellar signs [1][2][3]6].Cognitive disturbances were previously considered as a non-supporting feature of MSA.However, according to more recent findings and updated diagnostic criteria, cognitive symptoms have been recognized as not uncommon in this disease [7].
Among tauopathies, PSP is a rapidly progressive neurodegenerative disease with four clinical cornerstones such as ocular motor dysfunction, postural instability, akinesia and cognitive dysfunctions [8].Cognitive deficits are present up to 58% of PSP patients at the disease onset, with an earlier presentation in the Richardson's syndrome than Parkinsonian phenotype [1,8].Executive, memory and visuospatial functions are typically impaired in these patients.[6] CBD is characterised by cortical and extrapyramidal signs.Apraxia, cortical sensory deficits and alien limb phenomena are the most common cortical signs, whereas asymmetrical parkinsonism, dystonia and myoclonus comprise the motor signs [8,9].The prevalence of cognitive impairment in patients with CBD is 52% at disease onset, progressively increasing up to 70% over the disease course, with language as well as visuospatial dysfunctions being the most frequent clinical syndromes [8,9].
Neuroimaging methods have greatly improved the ability to understand the pathophysiology of Parkinsonisms, support the diagnosis of parkinsonian syndromes, and detect and monitor disease progression [10].Among different techniques, resting-state functional magnetic resonance imaging (rs-fMRI) has been widely applied to this purpose in patients with Parkinsonisms.
Rs-fMRI is based on the spontaneous oscillation of the blood oxygen level dependent (BOLD) signals [11,12], that derive from the processing of neuronal information at the synaptic level in specific brain areas according to the paramagnetic properties of blood [11][12][13][14].The temporal coherence of neuronal firing patterns from different brain areas represents the so-called functional connectivity (FC) [11].Different analytic approach may be applied to rs-fMRI data, such as seed-based FC, network-based independent component analysis and graph theory [12].
Seed-based analysis determines the FC patterns as emerging from a predefined seed or region of interest (ROI) to the whole-brain voxels or other seeds/ROIs voxels [11][12][13][14].
Independent component analysis is a data-driven method that can be applied to rs-fMRI data to isolate large-scale spatially distributed FC networks, called resting-state networks (RSNs).This method does not necessarily need a previous assumption [11][12][13][14].
Finally, graph analysis measures and techniques have been used to understand the global topological organization of brain networks.By applying this approach to rs-fMRI data, anatomic brain regions are considered to be nodes, linked by edges, which represent the FC between nodes [11][12][13][14].Such wiring diagram of the brain is called connectome and support an efficient global integration between high-specialized segregated areas.Interestingly, it has been proposed that the connectome architecture could play a direct role in spreading misfolded proteins across the brain, also explaining the stereotypical patterns of neurodegeneration [15].
In this narrative review, we aimed at summarizing the most updated studies that applied rs-fMRI to investigate the neural correlates of cognitive impairment in patients with tau and α-synuclein-based Parkinsonisms.

Search Strategy
Articles published in on PubMed in the last 3 years until August 2023 were systematically checked for the purpose of this review, considering only English-written articles published in peer-reviewed journals, with the use of the following words: "Parkinson's disease", "Lewy body dementia", "Multiple system atrophy", "Progressive supranuclear palsy", "Corticobasal degeneration", which were each crossreferenced with "resting state functional magnetic resonance imaging".
Two independent observers (NP and RDM) evaluated the results, excluding duplicates and articles judged irrelevant by title and abstract screening.The same raters performed the quality check of selected studies and the most relevant ones for the topic were finally included in this narrative review (Tables 1, 2 and 3).

Functional MRI Studies in Patients
with Synucleinopathies (Tables 1, 2 and 3)

PD and PDD
Heterogeneous patterns of rs-fMRI FC alterations were found to be associated to PD-related cognitive impairment.This may be potentially due to the inclusion of small samples, patients at different disease stages and/or in different medication states, and to the application of various fMRI approaches [16][17][18].In the last three years, several studies have been performed in patients with PD-MCI, sometimes leading to conflicting results [17,19].A few studies have been performed in patients with PDD.
The most reported large-scale RSN associated with cognition in PD is the default-mode network (DMN) [17,19], mainly involved in self-centered mental imagery, mindwandering and episodic memory retrieval.Several studies reported the presence of decreased FC within the DMN as associated with cognitive processing deterioration in PD [17][18][19].Recent findings have confirmed these results showing that decreased FC in DMN is correlated with episodic memory/working memory, executive/attentive or altered information processing speed, visuospatial dysfunctions and worse general cognition [20][21][22][23].A similar pattern of DMN involvement, particularly its posterior hubs, has been found also in early and/or drug-naive PD-MCI patients and in the absence of grey matter atrophy [20][21][22][23], suggesting that FC rearrangements within this network may reflect the presence of pathological processes (i.e. as misfolded protein accumulation) that not directly lead to neuronal death but may induce synaptic activity disturbances, which have been correlated with rs-fMRI signal strength [13].These correlates may be detected early in the disease and may be potentially tested as biomarkers to monitor disease progression and treatment-response.
On the other hand, also increased FC within DMN has been reported in PD-MCI patients and in cognitive normal patients (PD-CN) in advanced stages [24,25].
Even though decreased FC may be straightforward associated to loss of neural function, it is important to note that "hyper-connectivity" has also pathologic implications.Indeed, increased FC may be compensatory, representing a network response to the local neuronal injury that allows for the maintenance of the same global performance in nondemented PD patients to delay the development of clinical overt dementia.Moreover, what we see as "hyper-connected" may be underlined by the loss of networks dynamic properties to shift between different states (i.e. from a "hyper-" to a "hypo-connected" state), which has been also related to the presence of cognitive impairment in PD [26].This is in line with the finding of decreasing FC within the DMN over 3 years of follow-up in PD patients, along with progressive cognitive decline [27].
Together with DMN, other relevant RSNs such as the frontoparietal (FPN), dorsal and ventral attention (DAN/ VAN), salience (SN), executive-control (ECN) and visual (VN) networks, have been studied to investigate the potential association with cognitive impairment in PD.
The FPN is mainly involved in decision making and executive functioning.Decreased FC within this network has been consistently found in PD-MCI patients, also associated with progressive visuospatial, memory and attention dysfunctions [28][29][30][31][32][33].
The DAN plays a key-role in the top-down processes to control voluntary movements while the VAN is involved in the bottom-up detection of relevant stimuli [19].Decreased FC has been revealed within both DAN and VAN as associated with impaired attention and executive functioning, suggesting the presence of altered high cognitive control in PD-MCI [29,32,34,35].
Two recent studies have focused the amnestic (aMCI) and non-amnestic (naMCI) subtypes of PD-related MCI, showing divergent findings.Indeed, Chung and colleagues [29] performed a multimodal MRI study and found increased FC within the SN in PD-aMCI compared to PD-naMCI patients.Interestingly, in a longitudinal sub-cohort of PD, patients with PD-aMCI patients exhibited a higher risk of conversion to dementia compared to PD-naMCI.Conversely, Kawabata et al. revealed the presences of DMN changes characterising PD patients with aMCI from healthy controls (HC), while VN and cerebellar-brainstem network was found to be specifically altered in naMCI [36].These conflicting results may stem from the inclusion of different PD samples along with the application of heterogenous diagnostic criteria and methodological rs-fMRI approaches.Nonetheless, these findings support previous evidence of different cognitive syndromes in PD patients [1,4] that may not include an Alzheimer's disease (AD)-like phenotype [37], determining a slower but progressive multidomain deterioration of the brain efficiency and functioning that might eventually lead to dementia [38].
Beyond FC alterations within each network, the interaction between different RSNs has been found to be associated with cognitive impairment in PD patients [17,20,28,34,39,40].Overall, a functional coupling/decoupling patterns have been reported between the RSNs, with specific directions that seem critical to generate and maintain an efficient behavioral and cognitive performance, with several evidence in PD patients [17,20,28,34,[39][40][41][42].Graph-analysis MRI studies also provided intriguing findings to support the pathophysiology of PD-related cognitive dysfunction.Overall, altered connectomic metrics within the DMN, FPN and VN have been found to be associated with cognitive impairment in PD patients, suggesting the presence of both local and global information processing changes of the brain architecture [24,30,43,44].Indeed, a progressive shift from a global integrated to a small-world highly segregated organisation has been proposed to determine the development of MCI in PD.This would allow PD-MCI patients to confine affected nodes which are direct targets of early degenerative processes, while potentiating still-functioning networks to maintain an efficient brain functioning at a clinical level.
Despite the heterogeneity in rs-fMRI methodology, compelling evidence suggests the influence of the dopaminergic replacement therapy (DRT) on these neural patterns [45,46].A recent study [40] in early PD patients showed that along with the normalization of FC within motor networks, acute levodopa administration also enhances FC metrics in nonmotor areas (i.e.frontal and limbic regions).Further studies are needed to clarify whether this effect may be detrimental for brain functioning, likely leading to the development of cognitive and behavioural disturbances over time.
The presence of specific nonmotor symptoms, such as REM sleep behavioural disorders (RBD), depression, anxiety, impulsive compulsive disorders (ICD) and visual hallucinations (VHs), have been associated with an increased risk of dementia in PD patients.Interestingly, rs-fMRI studies allowed to highlight specific changes over cognitiverelated brain regions and networks in PD patients with these symptoms, even in the absence of clinical overt cognitive impairment.
RBD is a parasomnia characterized by loss of atonia during REM sleep [4,47].RBD may present as a prodromal symptom of PD, occurring even 10 years before the onset of motor features [4,47], and in the absence of overt signs of neurodegeneration (idiopathic RBD).Epidemiological studies recognized RBD as a risk factor for future development of cognitive impairment in PD patients [48,49].Previous studies have reported decreased FC within striato-frontal and temporo-parietal areas [50][51][52] as well as increased FC between basal ganglia and occipital cortex in patients with idiopathic RBD [53].Widespread FC disruption and potential maladaptive compensatory changes involving mainly the dorsolateral prefrontal cortex (PFC) and posterior cortical regions have been found in cognitively unimpaired PD patients with RBD, correlated with attention and memory cognitive outcomes [52,[54][55][56].The presence of such FC rearrangements may be tested as a potential future biomarker of an early vulnerability to develop dementia [54], as often seen in PD patients with RBD.Neuropsychiatric symptoms including depression and anxiety have a great impact on PD patients' quality of life, motor disability and caregiver burden.Their association with cognitive impairment in PD patients has been already reported, also suggesting a predictive role for the future development of dementia [57,58].
Epidemiological studies have found a higher prevalence of anxiety symptoms in PD patients with cognitive impairment [57,58].Both decreased and increased striato-limbic and posterior FC was found in PD patients with anxiety correlated with severity of anxiety symptoms [58,63,69,70].Taken together, these findings support the hypothesis that decreased top-down emotion regulation, limbic-prefrontal disconnection and abnormal cognitive control may represent potential pathophysiological mechanisms underlying neuropsychiatric symptoms in PD that may potentially lead to the development of cognitive impairment over time.
Clinical and neuroimaging evidence support the presence of a link between cognition and ICD in PD patients [17].Increased FC within striato-limbic and posterior regions has been associated with impulsivity in PD patients with ICD [71][72][73], potentially linked to increased perception of reward stimuli and internally-oriented ruminating compulsion.Similarly, decreased FC within the fronto-striatal network has been associated with stronger impulsivity and weaker behavioural control on decision making, leading to risky or inappropriate actions [71][72][73][74].These findings support the crucial role of cognitive functioning to maintain efficient risk/ benefit evaluation and decision-making, which may explain the presence of FC changes over cognitive-related regions.
Finally, VHs are common nonmotor symptoms in PD, regarding up to 40% of patients [75,76].Their presence is commonly linked to a higher risk of cognitive decline and dementia over time [77].In PD patients with VHs, previous rs-fMRI studies have demonstrated the presence of decreased FC within temporo-occipital areas and thalamicvisual network as well as increased FC in VN and between thalamus and frontal areas.Overall, these data suggest that altered integration of high order visual and cognitive processing may underlie the development of VHs in these patients [78][79][80].

DLB
DLB is the primary diagnosis of approximately 5% of patients with dementia [81].Clinically, patients with DLB present with dementia which may be variously associated with hallucinations, cognitive fluctuations, parkinsonism, and RBD.
In the last 3 years, only a few studies have been published that investigated the potential FC changes associated to DLB, also reporting controversial results with respect to previous evidence [81,82].
Two studies reported a similar pattern of FC alterations within the posterior DMN in DLB and AD patients compared to HC, even in the absence of clinical overt dementia [83,84].When compared to HC, PDD and AD patients, DLB patients showed disrupted FC within the DMN (mainly the anterior portion), FPN, SN, SMN, VAN as well as within striato-frontal, striato-temporal, striatooccipital, temporo-occipital and parieto-occipital areas and cerebellar regions [81,82,[85][86][87][88][89][90][91].Moreover, the presence of functional decoupling within the most reported neurocognitive networks has also been reported [81, 82, [85][86][87][88][89][90][91].Overall, a mismatch between the bottom-up inputs from the visual network and the top-down processing of visual stimuli from the prefrontal areas has been proposed to underlie the development of DLB core symptoms [81,82].On the other hand, other studies reported increased FC within the DMN (mainly the posterior portion), the VAN as well as between basal ganglia and posterior cortices, fronto-parietal and visuoperceptual areas in DLB patients relative to HC and AD [81,86,90], that have been hypothesized to exert a potential compensatory role.

MSA
MSA is a rare synucleinopathy characterized primarily by the presence of autonomic dysfunction associated with parkinsonian (i.e.MSA parkinsonian variant, MSA-P) or/and cerebellar (i.e.MSA, cerebellar variant, MSA-C) symptoms [6,92].Moreover, approximately 30% of MSA patients presents MCI, with executive/attentive, visuospatial and verbal functions being the most impaired cognitive domains [6].Most MRI studies in MSA patients have been focused on brain structural alterations, while a few reports have explored functional neural correlates in these patients.
Rs-fMRI studies have compared MSA patients with PD patients and HC, considering the two variants together and also separately, leading to controversial results.Many studies demonstrated the presence of disrupted FC within the DMN, SMN, VN and cerebellar regions, and between the cerebellum and the neurocognitive networks in MSA patients, particularly in MSA-C, relative to HC and PD patients [93][94][95][96][97][98][99].FC changes within the cerebellum-striato-cortical network have been also associated with the presence of cognitive dysfunction in these patients.Interestingly, in patients with MSA, FC changes within the cerebello-prefrontal network have been associated with verbal fluency and memory deficits whereas disconnection within the cerebello-limbic/temporal loop has been involved in language and visuospatial impairment [96,100].
Functional rearrangements within cognitive-related brain areas have been found in both MSA clinical phenotypes.Indeed, an increased FC between the dentate nucleus and posterior cingulate cortex has been demonstrated in MSA-P relative to PD patients and HC, while an increased FC between cerebellar and temporo-parietal regions and within the ponto-cerebellar network was found in MSA-C patients relative to HC [93,95,99], with a potential compensatory role.
Finally, increased FC between cerebellum and frontal areas/anterior cingulate cortex was also reported in MSA patients and supposed to be related to neuropsychiatric symptoms such as anxiety and/or depression [95].
Functional MRI Studies in Patients with Tauopathies (Table 4) PSP Among the tauopathies, PSP is a heterogeneous neurodegenerative disorder [8], with several phenotypic variants.PSP-Richardson's syndrome (PSP-RS) and PSP parkinsonian variant (PSP-P) are the most frequent [8].Beyond postural instability and oculomotor dysfunction, the great majority of PSP patients present also with early cognitive impairment.Frontal executive and verbal fluency dysfunction are the most characteristic and early deficits in PSP patients [8].However, memory, naming, visuospatial and social cognition deficits may also develop in these patients over the disease course [8].Growing and consistent structural MRI findings have increased the interest on neuroimaging correlates that may support clinical and pathological features in PSP [8,101].However, very little is known about rs-fMRI neural patterns that may be associated to the clinical spectrum of cognitive deficits in PSP.
Most studies demonstrated the presence of decreased FC within the most reported rs-fMRI networks, particularly in prefrontal areas and basal ganglia, that are associated with worse cognitive performances.Decreased FC within the midbrain has also been found and related to worse executive functions and vertical gaze impairment.Finally, rs-fMRI alterations within the thalamus and cerebellum have been also reported in PSP patients [102][103][104][105][106][107][108][109].Contrarily, some studies demonstrated the presence of increased FC within the DMN and the thalamo-cerebello-midbrain pathway, that correlated with worse cognitive status.Thus, this pattern could potentially underlie the presence of maladaptive FC between these areas, that eventually exert a detrimental effect on cognitive processing.
The presence of FC changes within the neurocognitive networks was also recently investigated in studies comparing PSP-RS and PSP-P patients, with PSP-RS showing more diffuse alterations within the DMN, SN, FPN networks as well as motor, limbic, cerebellar, occipito-temporal and thalamic areas potentially linked to increased neuropathological changes [108,110].

CBD/FTD
CBD is a rare and progressive tauopathy with different clinical presentations [9,111].The most common presentation of CBD is the corticobasal syndrome (CBS).Together with parkinsonian symptoms, CBS patients present with altered Seed-based analysis (SMA, primary motor area, putamen, globus pallidus and thalamus) 12 FTD-Park + , 18 FTD-Park-and 30 HC Decreased FC between SMA and putamen in FTD-Park + relative to FTD-Park-patients and HC high cortical functioning and behavioral changes [9].To date, rs-fMRI evidence in these patients is scarce.
Decreased FC between thalamus and fronto-striatal and cerebellar regions as well as between lateral VN and auditory networks were found in CBS patients relative to HC [106,109].Decreased FC within the cortico-subcortical-thalamic network may reflect the presence of specific neuropathological changes in these regions, that may in turn explain the presence of parkinsonian signs/symptoms.Similarly, the functional disconnection between auditory and visual networks may be related to altered high order multisensory inputs integration that may be related to the presence of cortical signs/ symptoms in CBS patients [106,109,111].On the other hand, an increased FC has been demonstrated in CBS patients relative to HC within the DMN, SMN, ECN, cerebellar, frontocerebellar and insular networks, and have been supposed to be linked to altered motor planning-preparation and executive/ emotional control on movements [106,109,111,112].
To the best of our knowledge, only one study has explored rs-fMRI changes in patients with Fronto-temporal dementia and parkinsonism (FTD-P), demonstrating the presence of decreased FC between the striatum and the supplementary motor area compared to HC, potentially responsible for failure in cognitive and motor processing in these patients [113].

Conclusion
This review resumes the most updated findings of rs-fMRI studies in patients with Parkinsonisms and related dementia.
Overall, in patients with Parkinsonisms and cognitive impairment, both synucleinopathies and tauopathies, aberrant increased and/or decreased FC alterations within the most reported neurocognitive networks have been found, further supporting a potential role for rs-fMRI as a surrogate biomarker of cognitive outcome.
Interestingly, FC changes within cognitive-related areas have been also found in PD patients with specific nonmotor symptoms which are known to be associated with high risk of dementia, even in the absence of cognitive impairment.This could potentially support the identification of early neural correlates that may be used to predict future conversion to dementia in PD patients.
While consistent results have been collected in patients with PD, a more intricate scenario emerged in other Parkinsonisms, with some studies showing also conflicting findings.More complex neuropathological patterns, also including the coexistence of different types of protein aggregates, as well as heterogenous clinical presentations compared to PD, may potentially explain this issue, still limiting the use of rs-fMRI as a reliable tool to help clinicians in the differential diagnosis.

Noemi
Piramide and Rosa De Micco contributed equally to this work.Supported by the Italian Ministry of Health (grant RF-2018-12366746).

Table 1
Summary of the most updated MRI seed-based functional connectivity studies in patients with synucleinopathies PFC and PCC in PD-MCI relative to HC Cascone et al., Commun Biol 2021 Seed-based analysis (FPN ROIs) 37 PD-MCI, 22 PD-CN and 21 HC Reduced topological brain-network resilience of FPN in PD-MCI patients associated with cognitive decline Ruppert et al., Hum Brain Mapp 2021 ICA-and seed-based analysis (DMN, FPN, DAN, SMN, VN) 12 PD-MCI, 36 PD-CN and 16 HC Increased DMN connectivity in PD-MCI compared both to PD-CN and HC.FC within DMN correlated with global cognitive, executive, attentive and visuospatial functions in PD-MCI Wang et al., Int J Gen Med.2021 Seed-based analysis (PCC as DMN ROI) 20 PD-MCI, 13 PD-CN and 13 HC Decreased FC in PCC in PD-MCI patients relative to HC, correlated with worse performances information processing speed, episodic memory and general cognition Chen et al., Front Neurosci 2022 Seed-based analysis (PCC as DMN ROI) 50 PD and 50 HC Increased FC between the PCC and the right precuneus, left cuneus, and right angular gyrus in PD patients relative to HC supposed to be responsible for cognitive decline Byun et al., Sleep Med 2020 Seed-based analysis (thalamus) 37 RBD and 15 HC Increased FC between thalamus and occipital regions in RBD patients relative to HC, correlated with cognitive dysfunctions, particularly memory Wakasugi et al., Parkinsonism Relat Disord 2021 ICA-and seed-based analysis (DMN, ECN, SMN, BGN) 50 RBD and 70 HC Decreased FC in ECN (fronto-striatal), SMN (pre and post-central regions) and BGN in RBD patients relative to HC Jia et al., Front Aging Neurosci 2022 Seed-based analysis (PCC as DMN ROI) 18 PD-pRBD, 28 PD-no-pRBD and 22 HC Decreased FC in frontal regions in PD-pRBD patients, correlated with executive dysfunctions that get worse over 3 years of follow-up Huang et al., Front Psychiatry 2020 Seed-based analysis (anterior/posterior insula) 17 PD-Dep, 17 PD-noDep and 17 HC Decreased FC between insula and frontoparietal regions in PD-Dep patients relative to PD-noDep patients and HC, correlated with worse global cognitive functions Liao et al., Front Neurosci.2020 (da controlare) Seed-based analysis (precentral, paracentral and medial frontal gyri) 33 PD-Dep, 60 PD-noDep and 47 HC

Table 2
Summary of the most updated MRI ICA-based functional connectivity studies in patients with synucleinopathies

Table 3
Summary of the most updated MRI graph-analysis functional connectivity studies in patients with synucleinopathies

Table 4
Summary of the most updated MRI functional connectivity studies in patients with tauopathies Abbreviations: Rs-fMRI, resting-state functional magnetic resonance imaging, AD, RS, 10 PSP-noRS, 13 no-AD MCI Decreased FC in DMN, SN, motor and FPN, and between cerebellum and subcortical and limbic structures, as well as increased FC between secondary visual cortex and inferior temporal gyrus, and between the cerebellum and cortical associative nodes in PSP-RS relative to non-AD MCI patients Whiteside et al., Human Brain Mapp 2023 ICA-based analysis (DMN, SMN, FPN, DAN and