Antagonistic network signature of motor function in Parkinson's disease revealed by connectome-based predictive modeling

Structural connectome combining DTI features predicts postoperative language decline and its recovery in glioma patients

OBJECTIVES

A decline in language function is a common complication after glioma surgery, affecting patients' quality of life and survival. This study predicts the postoperative decline in language function and whether it can be recovered based on the preoperative white matter structural network.

MATERIALS AND METHODS

Eighty-one right-handed patients with glioma involving the left hemisphere were retrospectively included. Their language function was assessed using the Western Aphasia Battery before and 1 week and 3 months after surgery. Structural connectome combining DTI features was selected to predict postoperative language decline and recovery. Nested cross-validation was used to optimize the models, evaluate the prediction performance of the models, and identify the most predictive features.

RESULTS

Five, seven, and seven features were finally selected as the predictive features in each model and used to establish predictive models for postoperative language decline (1 week after surgery), long-term language decline (3 months after surgery), and language recovery, respectively. The overall accuracy of the three models in nested cross-validation and overall area under the receiver operating characteristic curve were 0.840, 0.790, and 0.867, and 0.841, 0.778, and 0.901, respectively.

CONCLUSION

We used machine learning algorithms to establish models to predict whether the language function of glioma patients will decline after surgery and whether postoperative language deficit can recover, which may help improve the development of treatment strategies. The difference in features in the non-language decline or the language recovery group may reflect the structural basis for the protection and compensation of language function in gliomas.

CLINICAL RELEVANCE STATEMENT

Models can predict the postoperative language decline and whether it can recover in glioma patients, possibly improving the development of treatment strategies. The difference in selected features may reflect the structural basis for the protection and compensation of language function.

KEY POINTS

• Structural connectome combining diffusion tensor imaging features predicted glioma patients' language decline after surgery. • Structural connectome combining diffusion tensor imaging features predicted language recovery of glioma patients with postoperative language disorder. • Diffusion tensor imaging and connectome features related to language function changes imply plastic brain regions and connections.

MRI Assessment of Intrinsic Neural Timescale and Gray Matter Volume in Parkinson's Disease

BACKGROUND

Numerous studies have indicated altered temporal features of the brain function in Parkinson's disease (PD), and the autocorrelation magnitude of intrinsic neural signals, called intrinsic neural timescales, were often applied to estimate how long neural information stored in local brain areas. However, it is unclear whether PD patients at different disease stages exhibit abnormal timescales accompanied with abnormal gray matter volume (GMV).

PURPOSE

To assess the intrinsic timescale and GMV in PD.

STUDY TYPE

Prospective.

POPULATION

74 idiopathic PD patients (44 early stage (PD-ES) and 30 late stage (PD-LS), as determined by the Hoehn and Yahr (HY) severity classification scale), and 73 healthy controls (HC).

FIELD STRENGTH/SEQUENCE

3.0 T MRI scanner; magnetization prepared rapid acquisition gradient echo and echo planar imaging sequences.

ASSESSMENT

The timescales were estimated by using the autocorrelation magnitude of neural signals. Voxel-based morphometry was performed to calculate GMV in the whole brain. Severity of motor symptoms and cognitive impairments were assessed using the unified PD rating scale, the HY scale, the Montreal cognitive assessment, and the mini-mental state examination.

STATISTICAL TEST

Analysis of variance; two-sample t-test; Spearman rank correlation analysis; Mann-Whitney U test; Kruskal-Wallis' H test. A P value <0.05 was considered statistically significant.

RESULTS

The PD group had significantly abnormal intrinsic timescales in the sensorimotor, visual, and cognitive-related areas, which correlated with the symptom severity (ρ = -0.265, P = 0.022) and GMV (ρ = 0.254, P = 0.029). Compared to the HC group, the PD-ES group had significantly longer timescales in anterior cortical regions, whereas the PD-LS group had significantly shorter timescales in posterior cortical regions.

CONCLUSION

This study suggested that PD patients have abnormal timescales in multisystem and distinct patterns of timescales and GMV in cerebral cortex at different disease stages. This may provide new insights for the neural substrate of PD.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY STAGE: 1.

Motor networks, but also non-motor networks predict motor signs in Parkinson's disease

OBJECTIVE

Investigate the brain functional networks associated with motor impairment in people with Parkinson's disease (PD).

BACKGROUND

PD is primarily characterized by motor dysfunction. Resting-state functional connectivity (RsFC) offers a unique opportunity to non-invasively characterize brain function. In this study, we hypothesized that the motor dysfunction observed in people with PD involves atypical connectivity not only in motor but also in higher-level attention networks. Understanding the interaction between motor and non-motor RsFC that are related to the motor signs could provide insights into PD pathophysiology.

METHODS

We used data from 88 people with PD (mean age: 68.2(SD:10), 55 M/33F) coming from 2 cohorts. Motor severity was assessed in practical OFF-medication state, using MDS-UPDRS Part-III motor scores (mean: 49 (SD:10)). RsFC was characterized using an atlas of 384 regions that were grouped into 13 functional networks. Associations between RsFC and motor severity were assessed independently for each RsFC using predictive modeling.

RESULTS

The top 5 % models that predicted the MDS-UPDRS-III motor scores with effect size >0.5 were the connectivity between (1) the somatomotor and Subcortical-Basal-ganglia, (2) somatomotor and Visual and (3) CinguloOpercular (CiO) and language/Ventral attention (Lan/VeA) network pairs.

DISCUSSION

Our findings suggest that, along with motor networks, visual- and attention-related cortical networks are also associated with the motor symptoms of PD. Non-motor networks may be involved indirectly in motor-coordination. When people with PD have deficits in motor networks, more attention may be needed to carry out formerly automatic motor functions, consistent with compensatory mechanisms in parkinsonian movement disorders.

Connectome-based fingerprint of motor impairment is stable along the course of Parkinson's disease

Functional alterations in brain connectivity have previously been described in Parkinson's disease, but it is not clear whether individual differences in connectivity profiles might be also linked to severity of motor-symptom manifestation. Here we investigated the relevance of individual functional connectivity patterns measured with resting-state fMRI with respect to motor-symptom severity in Parkinson's disease, through a whole-brain, data-driven approach (connectome-based predictive modeling). Neuroimaging and clinical data of Parkinson's disease patients from the Parkinson's Progression Markers Initiative were derived at baseline (session 1, n = 81) and at follow-up (session 2, n = 53). Connectome-based predictive modeling protocol was implemented to predict levels of motor impairment from individual connectivity profiles. The resulting predictive model comprised a network mainly involving functional connections between regions located in the cerebellum, and in the motor and frontoparietal networks. The predictive power of the model was stable along disease progression, as the connectivity within the same network could predict levels of motor impairment, even at a later stage of the disease. Finally, connectivity profiles within this network could be identified at the individual level, suggesting the presence of individual fingerprints within resting-state fMRI connectivity associated with motor manifestations in Parkinson's disease.

Characterization of Hemodynamic Alteration in Parkinson's Disease and Effect on Resting-State Connectivity

Functional magnetic resonance imaging (fMRI) is a convolution of latent neural activity and the hemodynamic response function (HRF). According to prior studies, the neurodegenerative process in idiopathic Parkinson's Disease (PD) interacts significantly with neuromuscular abnormalities. Although these underlying neuromuscular changes might influence the temporal characteristics of HRF and fMRI signals, relatively few studies have explored this possibility. We hypothesized that such alterations would engender changes in estimated functional connectivity (FC) in fMRI space compared to latent neural space. To test these theories, we calculated voxel-level HRFs by deconvolving resting-state fMRI data from PD patients (n = 61) and healthy controls (HC) (n = 47). Significant group differences in HRF (P < 0.05, Gaussian random field-corrected) were observed in several regions previously associated with PD. Subsequently, we focused on putamen-seed-based FC differences between the PD and HC groups using fMRI and latent neural signals. The results suggested that neglecting HRF variability may cultivate false-positive and false-negative FC group differences. Furthermore, HRF was related to dopamine receptor type 2 (DRD2) gene expression (P < 0.001, t = -7.06, false discover rate-corrected). Taken together, these findings reveal HRF variation and its possible underlying molecular mechanism in PD, and suggest that deconvolution could reduce the impact of HRF variation on FC group differences.

Attenuated brain white matter functional network interactions in Parkinson's disease

Parkinson's disease (PD) is a neurodegenerative disorder characterized by extensive structural abnormalities in cortical and subcortical brain areas. However, an association between changes in the functional networks in brain white matter (BWM) and Parkinson's symptoms remains unclear. With confirming evidence that resting‐state functional magnetic resonance imaging (rs‐fMRI) of BWM signals can effectively describe neuronal activity, this study investigated the interactions among BWM functional networks in PD relative to healthy controls (HC). Sixty‐eight patients with PD and sixty‐three HC underwent rs‐fMRI. Twelve BWM functional networks were identified by K‐means clustering algorithm, which were further classified as deep, middle, and superficial layers. Network‐level interactions were examined via coefficient Granger causality analysis. Compared with the HC, the patients with PD displayed significantly weaker functional interaction strength within the BWM networks, particularly excitatory influences from the superficial to deep networks. The patients also showed significantly weaker inhibitory influences from the deep to superficial networks. Additionally, the sum of the absolutely positive/negative regression coefficients of the tri‐layered networks in the patients was lower relative to HC (p < .05, corrected for false discovery rate). Moreover, we found the functional interactions involving the deep BWM networks negatively correlated with part III of the Unified Parkinson's Disease Rating Scales and Hamilton Depression Scales. Taken together, we demonstrated attenuated BWM interactions in PD and these abnormalities were associated with clinical motor and nonmotor symptoms. These findings may aid understanding of the neuropathology of PD and its progression throughout the nervous system from the perspective of BWM function.