Brain functional connectivity and activity during motor control in children with cerebral palsy: a pilot cross-sectional fNIRS study

Introduction

Cerebral palsy (CP) is a neurological disorder caused by non-progressive brain injuries in fetuses or infants, affecting approximately 2.0 to 2.5 per 1,000 births worldwide (1). The spastic subtype, the most common form of CP, is characterized by hypertonia, muscle weakness, and impaired motor control (2,3). CP involves both positive motor signs, such as spasticity and hyperreflexia, and negative motor signs, including muscle weakness and impaired selective motor control. Negative signs, in particular, significantly impact daily life activities and participation, often leading to greater functional limitations than positive signs alone (4). Motor control issues in children with CP often present as involuntary movements that accompany voluntary actions, as well as simultaneous muscle activation patterns. Even children with mild to moderate impairments [Gross Motor Function Classification System (GMFCS) levels I–II] often display immature motor control, particularly in lower-limb function during walking, which results in abnormal gait and limited participation in daily activities (5). Damage to the corticospinal tract and motor cortex, often seen in CP, can disrupt neural connectivity and lead to motor area deafferentation, further impairing motor control (6). Strong functional connectivity is essential for neural communication and coordination among brain regions (7). However, current CP research largely focuses on electromyography, gait analysis, and muscle activation patterns, with limited insights into brain activation patterns and functional connectivity during motor tasks in children with CP (5).

Functional magnetic resonance imaging (fMRI) provides valuable data on brain function, but its use with children with CP is constrained by requirements for sedation, high costs, and limited accessibility (8). Functional near-infrared spectroscopy (fNIRS), a non-invasive, portable brain imaging technique with good temporal resolution, offers an alternative by measuring changes in oxygenated hemoglobin (HbO₂) and deoxygenated hemoglobin (HbR) during specific tasks (9). Additionally, fNIRS has been increasingly used to explore neuroplasticity and the effects of interventions, providing real-time insights into cortical activation and connectivity during motor tasks. As a promising neuroimaging tool for rehabilitation research, fNIRS proves particularly advantageous in studying children with CP due to its portability, non-invasive nature, and capacity for motion-tolerant brain monitoring (10). Unlike fMRI, fNIRS allows greater flexibility in positioning, making it better suited for use with children with neurological disorders (11).

Previous fNIRS studies in bilateral CP (BCP) have predominantly examined the sensory-motor cortex (SMC), demonstrating increased activation during ankle dorsiflexion with knee flexion tasks compared to typically developing (TD) children (12). The children with CP exhibit altered functional connectivity, particularly reduced connectivity in motor control regions such as the SMC and premotor cortex (PM), due to disrupted neural pathways from early brain injuries (5). Motor control involves multiple brain regions (13), the PM aids planning, selecting, and preparing motor programs, the dorsolateral prefrontal cortex (DLPFC) provides cognitive control (14), the supplementary motor area (SMA) supports (15), and the somatosensory association cortex (SAC) integrates sensory feedback for precision (16,17). This study introduced an innovative task—ankle dorsiflexion with knee extension—mimicking the swing phase of gait, and linked brain activation and functional connectivity to motor function, assessed via the Selective Control Assessment of the Lower Extremity (SCALE) score, in children with CP, offering insights into neural-motor relationships to guide rehabilitation. Expanding on prior research, this study investigates a wider array of brain regions (DLPFC, SMC, PM, SMA, SAC) and includes both BCP and hemiplegic CP (HCP) alongside a TD control group, capturing the heterogeneity of CP—BCP typically involves bilateral brain damage, while HCP is characterized by unilateral injury.

This study aims to compare brain activity and functional connectivity between children with CP and TD children during motor control tasks. We hypothesize that children with CP exhibit increased brain activation but reduced functional connectivity in specific regions of interest (ROIs) compared to TD children. Furthermore, we anticipate the existence of a correlation between motor control, brain activity, and functional connectivity. We present this article in accordance with the STROBE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-11/rc).

Methods

Participants

All tests were conducted at the Department of Rehabilitation of the Guangzhou Women and Children’s Medical Center from September to November 2023. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Guangzhou Women and Children’s Medical Center (No. 2023017A01, date: 6 February 2023), and informed consent was taken from all the patients’ guardians.

The inclusion criteria for this study are as follows: All participants need to be: (I) between the ages of 6–18 years, (II) either diagnosed with spastic CP or exhibiting typical development, (III) able to understand and follow simple directions, and (IV) have informed consent signed by their caregivers, with the additional consent from the child if they are over 8 years old. (V) Children with CP should be classified as level I–II on the GMFCS. The exclusion criteria include (I) having muscle and joint surgery or experiencing a seizure within the past year; (II) having taken antiepileptic or tone-adjusting drugs, or received botulinum toxin injections within the past 6 months, as these could alter brain activity or motor function (18); (III) presence of a joint contracture in the lower limbs, a joint contracture was defined as a fixed limitation in the range of motion of the lower limbs, assessed via physical examination by a trained clinician; (IV) inability to maintain attention during repetitive tasks; and (V) presence of auditory impairments. In this prospective study, participants were categorized into three groups based on their clinical characteristics: (I) TD children (TD group); (II) children with BCP (BCP group); and (III) children with HCP (HCP group) (19). Figure 1 presents a flow chart of the study.

Figure 1 Flow chart of the overall research.

Eighteen children were initially enrolled in this cross-sectional study. However, due to difficulties in cooperating during the collection of fNIRS data, one child was excluded from the study. Among the participants, eleven individuals were tested on the left side. The final analysis included 6 participants with BCP classified as GMFCS level II, with an average age of 9.02 years [standard deviation (SD) 1.85 years]; 6 participants with HCP classified as GMFCS level I, with an average age of 9.88 years (SD 1.71 years); and 5 age-matched TD children, with an average age of 8.02 years (SD 1.82 years). One participant’s fNIRS data were missing, which could introduce bias. However, sensitivity analyses indicated that excluding this participant did not significantly impact the overall findings, suggesting that the results remain robust despite the missing data. Participants’ characteristics and clinical features are summarized in Table 1 and Table S1. The study sample size was determined based on practical recruitment feasibility and prior studies utilizing fNIRS in children with CP. Due to the small sample, this study serves as a pilot investigation to inform future larger-scale research with formal power calculations.

Clinical assessments

Participants with CP underwent a series of assessments to evaluate their gross motor function and motor control. The GMFCS is a five-level clinical classification system that describes the gross motor function of people with CP on the basis of self-initiated movement ability (20). The SCALE evaluates selective motor control in five joints of each lower limb: the hip, knee, ankle, subtalar joint, and toes. Each joint is scored from 0 (no selective control) to 2 (full selective control), yielding a total score of 10 per leg, with higher scores indicating better selective motor control. This tool is clinically significant in CP for assessing motor control deficits and guiding treatment (21).

In this study, children in the HCP group were tested on their affected side, while those in the TD and BCP groups were assessed on their non-dominant side using the SCALE (12). The non-dominant leg was determined through a two-step process using subjective and objective methods. First, participants and their parents identified the perceived weaker leg. This was then confirmed with a single-leg standing time test, designating the leg with the shorter stable standing duration as non-dominant. For those unable to perform the standard test due to balance or strength issues, a modified version allowed chair support, and the leg with the shortest standing time in this condition was identified as non-dominant. Subsequent to these evaluations, the fNIRS measurements were conducted.

fNIRS procedure

The fNIRS data were gathered using the NirSmart-6000A (Danyang Huichuang Medical Equipment Co., Ltd., China), which has 24 sources and 16 detectors (a total of 48 channels) with interoptode distances ranging from 2.0 to 3.3 cm. This range of distances was chosen to optimize signal quality and ensure appropriate penetration depth for capturing cortical hemodynamic responses while controlling for superficial hemodynamic changes. The system operates at a sampling rate of 11 Hz in a quiet and dimly lit room and uses 2 wavelengths of near-infrared light (730 and 850 nm) to detect changes in hemoglobin concentration in the cerebral cortex between the rest and task states (22).

The probes were placed according to the 10–20 international system for EEG electrode placement, ensuring standardized and reproducible positioning (23). The fNIRS channels were identified using 3D spatial registration techniques, allowing for accurate localization of the ROIs, which were defined as the DLPFC, PM & SMA, SMC, and SAC (Figure 2A,2B). These ROIs were selected due to their critical roles in motor planning (DLPFC, PM&SMA), execution (SMC), and sensory integration (SAC), which are essential for coordinated motor function.

Figure 2 The arrangement and positioning of fNIRS channels and their corresponding ROIs on the brain. (A) The arrangements of fNIRS channels. (B) fNIRS emitters and detectors for the ROIs on a brain in top view. This figure was drawn using Adobe Photoshop (version 23.5.1). fNIRS, functional near-infrared spectroscopy; ROIs, regions of interest.

Participants were seated on a wooden bench with back support, maintaining a hip flexion of approximately 90°. Both lower extremities were supported with thighs and calves resting on the bench surface to ensure assisted knee extension (Figure 3). Auditory cues standardized the timing protocol: participants performed dorsiflexion with knee extension of the non-dominant ankle at a frequency of 1 Hz initiated by a tone (adhering to a dorsiflexion-relaxation-dorsiflexion-relaxation cycle), maintaining movement for precisely 15 seconds followed by 30-second rest intervals. This sequence was repeated across five cycles under direct supervision to ensure adherence to the movement parameters. This approach ensured consistent timing and execution across trials, allowing for accurate synchronization with the fNIRS data collection. All participants followed a structured motor task, guided by auditory cues, and supervised by an experienced investigator to ensure uniform execution.

Figure 3 The position of participants during the execution of the motor control task.

Data analysis

fNIRS data processing

Data processing was conducted using the NirSpark software package. Firstly, motion artifacts were automatically removed six times using selected channels and spline interpolation. Secondly, to differentiate brain hemodynamic changes from superficial signals, we applied a bandpass filter at 0.01–0.20 Hz to eliminate physiological noise, such as heart rate and respiration (24). These preprocessing steps ensured the accurate isolation of brain-specific HbO₂ concentration changes. Finally, using a modified version of the Beer-law, Lambert’s optical density was derived from the light intensity signal to track changes in HbO₂ concentration (10).

Analysis of cortical functional connectivity

The channels were reversed across the midline in participants whose non-dominant lower limb tested was on the right so that all results were reported as if all participants were performing at their left-side. Using the NirSpark software’s network module, In the network module of the NirSpark software, changes in HbO₂ concentrations at each time point, recorded by the subject throughout task blocks, were extracted. The Pearson correlation coefficients for HbO₂ levels of each ROIs were examined on a time series (25). The transformed values were defined as functional connectivity strengths between the ROIs, after the Fisher-R transformation was completed. Besides, self-connectivity refers to the functional connectivity within the same ROI, indicating the strength of intra-regional communication.

Analysis of brain activation

We utilized the General Linear Model as a statistical approach to quantify task-related changes in HbO₂ concentrations (26,27). By using a one-sample t-test and FDR correction, it was possible to determine which channels were significantly active during the task state compared to the resting state (27). The total HbO₂ values of the active channels refer to the cumulative sum of HbO₂ concentration changes across all channels identified as significantly active during the motor task. This metric provides an aggregate measure of brain activation in response to the task. The number of significant channels refers to the count of fNIRS channels that show statistically significant changes in HbO₂ concentrations between the task and rest conditions. These significant channels indicate areas of the brain that are actively engaged during the motor task.

Statistical analysis

All statistical tests performed outside of the NirSpark software were conducted using IBM SPSS (version 23). Distribution of the data was tested for normality using the Shapiro-Wilk test. Normally distributed data are presented as mean ± SD. Nonparametric data are presented as median (range). The Kruskal-Wallis test, with a degree of freedom (df) of 2, was used to assess differences in brain activity and functional connectivity among three groups while participants engaged in a motor control task. The dependent variables include: the number of active fNIRS channels, the sum of HbO₂ values in active fNIRS channels, and the functional connectivity of ROIs. Spearman’s correlation coefficients were used to determine the correlations between SCALE scores, brain activity, and functional connectivity. Bonferroni test was chosen to correct for multiple comparisons. P value less than 0.05 was taken as the level of significance.

Results

Cortical functional connectivity

Figure 4 presents the functional connectivity diagram based on HbO₂ concentrations, while Table 2 shows the comparison of functional connectivity across the three groups. We observed universally positive connectivity strengths amongst all ROIs during motor control tasks. Specifically, significant differences were found in the connectivity between the dominant PM and SMA and the SAC, as well as between the non-dominant SMC and SAC (see Table 2). These differences were most pronounced between the TD and BCP groups. Notably, four connections displayed significant variations among the groups: Significant differences in functional connectivity were observed between the dominant PM and SMA, with both the dominant (H=8.685, P=0.03) and nondominant SAC (H=7.46, P=0.02). Similarly, functional connectivity between the nondominant sensory motor cortex and both the dominant (H=7.319, P=0.02) and nondominant SAC (H=6.214, P=0.04) also showed significant differences. In addition, our findings indicated significant main effects in the self-connectivity of two areas: the nondominant SMC (H=6.942, P=0.03) and the dominant SAC (H=6.727, P=0.04). These findings suggest that the intra-regional connectivity, or self-connectivity, in these areas is differentially impacted across the groups.

Figure 4 Functional connectivity among ROI across three groups, with the strength of connections represented using a color gradient: red indicates stronger connectivity, while blue signifies weaker connections. Specifically, (A) TD children, illustrating the baseline connectivity patterns; (B) children with HCP, highlighting altered connectivity in this group; (C) children with BCP, where the connectivity patterns may differ significantly from TD groups. This figure was drawn using Adobe Photoshop (version 23.5.1). D, dominant; DLPFC, dorsolateral prefrontal cortex; ND, nondominant; PM & SMA, premotor and supplementary motor area; ROI, regions of interest; SAC, somatosensory association cortex; SMC, sensory-motor cortex; TD, typically developing.

These results suggest that TD children may have more efficient communication during motor control tasks. In the HCP group, the dominant DLPFC showed a tendency for stronger connectivity with other ROIs, and connections in other ROIs appeared weaker compared to the TD group, these differences did not reach statistical significance. Table 2 shows a possible reduction in interhemispheric connectivity in HCP, which may result from unilateral brain injury disrupting the corpus callosum during early developmental stages, weakening interhemispheric connectivity. The lack of statistical significance may be due to the small sample size, as no formal power calculation was conducted.

Correlation between motor control and functional connectivity

Figure 5 details the correlation coefficients between SCALE scores and functional connectivity. A strong correlation is defined as being above 0.75, and a moderate correlation falls between 0.5 and 0.75 (28).

Figure 5 Scatter plots show the significant correlation trends and strength between motor control (as indicated by SCALE scores) and functional connectivity for each ROI pair in children with CP. These figures were drawn using GraphPad Prism (version 9.5.0). CP, cerebral palsy; D, dominant; DLPFC, dorsolateral prefrontal cortex; ND, nondominant; PM & SMA, premotor and supplementary motor area; ROI, regions of interest; SAC, somatosensory association cortex; SCALE, Selective Control Assessment of the Lower Extremity; SMC, sensory-motor cortex.

Our findings demonstrate notable correlations in different brain regions, with strong correlations observed particularly between the dominant DLPFC and the SAC (r=0.834–0.873, P<0.001). Additionally, moderate correlations are evident between the dominant DLPFC and other areas, including the premotor and SMA, and SMC (r=0.594–0.716, P<0.05). These results underscore a significant link between enhanced motor control and increased functional connectivity among crucial regions like the DLPFC, premotor and SMA, sensory motor cortex, and SAC.

Brain activation

When participants performed motor control tasks with their non-dominant leg, we found no statistically significant differences among the three groups in either the number of significant channels (H=2.77, P=0.25) or the total HbO₂ values of the active channels (H=2.71, P=0.25). The absence of discernible disparities underscores consistent brain activity responses to motor tasks across all the participant groups.

Discussion

In this study, we used fNIRS to investigate brain activity and functional connectivity in children with CP and TD during motor control tasks. Our findings provide valuable insights into the neural basis of motor control and its clinical implications. By focusing on brain regions affected by deafferentation, this research supports the development of targeted interventions that may enhance motor control in children with CP through neuroplasticity.

Cortical functional connectivity

Functional connectivity refers to the temporal correlation of neural activity between distinct brain regions, reflecting how these areas coordinate and communicate during cognitive or motor tasks. In the field of motor control, functional connectivity plays a critical role, as it supports the integration of sensory information, the formulation of motor plans, and the execution of coordinated movements (29).

Our study confirmed that the DLPFC, premotor and SMA, sensory motor cortex, and SAC form a functional network essential for coordinated motor function, consistent with existing literature (30). Moreover, we found that both intrahemispheric and interhemispheric functional connectivity in the motor cortex were weaker in children with CP than in TD children, which is consistent with the results of previous resting-state fMRI studies (31,32).

In children with BCP, brain injury typically involves the white matter of both cerebral hemispheres, causing widespread deficits in functional connectivity (32). This network disruption affects the entire motor system, including connectivity between the sensorimotor cortex and premotor area (33). Such extensive abnormalities are associated with more severe motor impairments, such as poor coordination of bilateral limbs or abnormal muscle tone. We observed a marked reduction in functional connectivity in the sensorimotor cortex, premotor and SMA, and SAC in these children. This reduction may stem from early developmental injuries common in CP, which impair structural and functional connectivity among these regions (6). Furthermore, motor deficits in CP may limit efficient movement execution, potentially triggering compensatory neural mechanisms that alter connectivity, as seen in the HCP group (17). These findings support the hypothesis proposed by Lee et al. (34), suggesting that reduced motor cortical functional connectivity in motor control regions, combined with focal lesions in the corticospinal tracts, may underlie the pathophysiological mechanisms contributing to motor dysfunction.

In children with HCP, brain injury is typically limited to one hemisphere, resulting in functional connectivity deficits primarily in the affected hemisphere. fMRI studies indicate that connectivity patterns in HCP show significant asymmetry, with reduced connectivity in the affected hemisphere and relatively preserved connectivity in the unaffected hemisphere (31). This asymmetry reflects the characteristics of unilateral injury and is consistent with the clinical presentation of unilateral motor impairments in HCP children. This asymmetry mirrors the unilateral injury and aligns with clinical motor impairments. The HCP group exhibited a trend of reduced connectivity in these regions compared to the TD group, though this difference was not statistically significant. We believe that this trend may relate to developmental abnormalities of the corpus callosum, where unilateral injury disrupts its formation, weakening interhemispheric connectivity (35). Besides, this may reflect milder motor impairments in hemiplegia, often more evident in hand function than lower limbs, with all HCP individuals classified as GMFCS level I, indicating mild motor impairment.

Cortical functional connectivity associated with motor control

This study found a positive correlation between the non-dominant SCALE score and functional connectivity of the DLPFC, premotor and SMA, sensory motor cortex, and SAC in children with CP. This correlation underscores the importance of synchronized neural communication for proficient motor control and suggests that enhanced functional connectivity may play a critical role in improving motor control in this population. These insights also highlight the potential of fNIRS as a complementary tool for assessing motor control in children with CP. Based on our findings, neuromodulation techniques, such as Repetitive Transcranial Magnetic Stimulation, which have been shown to enhance functional connectivity in Alzheimer’s disease, warrant investigation for their applicability in strengthening functional connectivity in crucial motor regions in children with CP (36).

Brain activation

In contrast to the findings of de Campos et al., our study did not observe significant differences in brain activation across the three groups during motor control tasks (10). This discrepancy might be due to the younger age of participants in our study. Emerging evidence suggests that younger individuals may exhibit lower levels of brain activation during task performance, compensated by broader functional connectivity networks (37). These findings highlight the need for network-based rehabilitation approaches that focus on restoring functional connectivity between brain regions, rather than just targeting isolated cortical activation.

There are several limitations in this study. First, fNIRS can only detect cortical activity and not deeper structures, such as the basal ganglia, which play a crucial role in motor control. Second, the relatively small sample size may limit the statistical power to detect anticipated differences in functional connectivity. Additionally, the focus solely on a lower limb motor task without including an upper limb task restricts the generalizability of our findings, as upper limb dysfunction is prevalent in children with CP and significantly affects daily activities such as self-care and fine motor skills. Future research should address these limitations by increasing sample sizes, exploring techniques to assess deeper brain structures, and incorporating upper limb tasks to better elucidate motor control deficits in CP.

Conclusions

To conclude, our study leverages fNIRS technology to uncover diminished functional connectivity in premotor and SMA, sensory motor cortex, and SAC of children with CP, suggesting a neural basis for their motor control difficulties. Targeted interventions to enhance functional connectivity of these key cortical regions may improve motor control and coordination in children with CP.

Acknowledgments

We sincerely appreciate the children and their families for participating in this study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-11/rc

Data Sharing Statement: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-11/dss

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-22025-11/prf

Funding: The study was funded by STI 2030—Major Projects (No. 2021ZD0200500), the General Guidance Project of Guangzhou Municipal Health Commission (No. 20241A011030), Featured Clinical Technique of Guangzhou (No. 2023C-TS59), and Guangzhou Municipal Science and Technology Project (Nos. 2024A03J01274, 202201020627). The funders played no role in the design, conduct, or reporting of this study.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-11/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Guangzhou Women and Children’s Medical Center (No. 2023017A01, date: 6 February 2023) and informed consent was taken from all the patients’ guardians.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Li N, Zhou P, Tang H, et al. In-depth analysis reveals complex molecular aetiology in a cohort of idiopathic cerebral palsy. Brain 2022;145:119-41. [Crossref] [PubMed]
2. Bekteshi S, Monbaliu E, McIntyre S, et al. Towards functional improvement of motor disorders associated with cerebral palsy. Lancet Neurol 2023;22:229-43. [Crossref] [PubMed]
3. Kim WH, Park EY. Causal relation between spasticity, strength, gross motor function, and functional outcome in children with cerebral palsy: a path analysis. Dev Med Child Neurol 2011;53:68-73. [Crossref] [PubMed]
4. Sanger TD, Chen D, Delgado MR, et al. Definition and classification of negative motor signs in childhood. Pediatrics 2006;118:2159-67. [Crossref] [PubMed]
5. Cahill-Rowley K, Rose J. Etiology of impaired selective motor control: emerging evidence and its implications for research and treatment in cerebral palsy. Dev Med Child Neurol 2014;56:522-8. [Crossref] [PubMed]
6. Grefkes C, Fink GR. Connectivity-based approaches in stroke and recovery of function. Lancet Neurol 2014;13:206-16. [Crossref] [PubMed]
7. Hutchison RM, Womelsdorf T, Allen EA, et al. Dynamic functional connectivity: promise, issues, and interpretations. Neuroimage 2013;80:360-78. [Crossref] [PubMed]
8. Kraus D, Horowitz-Kraus T. Functional MRI research involving healthy children: Ethics, safety and recommended procedures. Acta Paediatr 2022;111:741-9. [Crossref] [PubMed]
9. Nishiyori R, Bisconti S, Ulrich B. Motor Cortex Activity During Functional Motor Skills: An fNIRS Study. Brain Topogr 2016;29:42-55. [Crossref] [PubMed]
10. de Campos AC, Sukal-Moulton T, Huppert T, et al. Brain activation patterns underlying upper limb bilateral motor coordination in unilateral cerebral palsy: an fNIRS study. Dev Med Child Neurol 2020;62:625-32. [Crossref] [PubMed]
11. Zhong X, Wang C, Xu M, et al. Physical training improves inhibitory control in children aged 7-12 years: An fNIRS study. Behav Brain Res 2024;463:114902. [Crossref] [PubMed]
12. Sukal-Moulton T, de Campos AC, Alter KE, et al. Relationship between sensorimotor cortical activation as assessed by functional near infrared spectroscopy and lower extremity motor coordination in bilateral cerebral palsy. Neuroimage Clin 2018;20:275-85. [Crossref] [PubMed]
13. Nakayama Y, Sugawara SK, Fukunaga M, et al. The dorsal premotor cortex encodes the step-by-step planning processes for goal-directed motor behavior in humans. Neuroimage 2022;256:119221. [Crossref] [PubMed]
14. Haar S, Donchin O. A Revised Computational Neuroanatomy for Motor Control. J Cogn Neurosci 2020;32:1823-36. [Crossref] [PubMed]
15. Russo AA, Khajeh R, Bittner SR, et al. Neural Trajectories in the Supplementary Motor Area and Motor Cortex Exhibit Distinct Geometries, Compatible with Different Classes of Computation. Neuron 2020;107:745-758.e6. [Crossref] [PubMed]
16. Gale DJ, Flanagan JR, Gallivan JP. Human Somatosensory Cortex Is Modulated during Motor Planning. J Neurosci 2021;41:5909-22. [Crossref] [PubMed]
17. Staudt M. (Re-)organization of the developing human brain following periventricular white matter lesions. Neurosci Biobehav Rev 2007;31:1150-6. [Crossref] [PubMed]
18. Beltramini GC, Cendes F, Yasuda CL. The effects of antiepileptic drugs on cognitive functional magnetic resonance imaging. Quant Imaging Med Surg 2015;5:238-46. [Crossref] [PubMed]
19. Sukal-Moulton T, de Campos AC, Alter KE, et al. Functional near-infrared spectroscopy to assess sensorimotor cortical activity during hand squeezing and ankle dorsiflexion in individuals with and without bilateral and unilateral cerebral palsy. Neurophotonics 2020;7:045001. [Crossref] [PubMed]
20. Piscitelli D, Ferrarello F, Ugolini A, et al. Measurement properties of the Gross Motor Function Classification System, Gross Motor Function Classification System-Expanded & Revised, Manual Ability Classification System, and Communication Function Classification System in cerebral palsy: a systematic review with meta-analysis. Dev Med Child Neurol 2021;63:1251-61. [Crossref] [PubMed]
21. Fowler EG, Staudt LA, Greenberg MB, et al. Selective Control Assessment of the Lower Extremity (SCALE): development, validation, and interrater reliability of a clinical tool for patients with cerebral palsy. Dev Med Child Neurol 2009;51:607-14. [Crossref] [PubMed]
22. Chen S, Zhang X, Chen X, et al. The assessment of interhemispheric imbalance using functional near-infrared spectroscopic and transcranial magnetic stimulation for predicting motor outcome after stroke. Front Neurosci 2023;17:1231693. [Crossref] [PubMed]
23. Zhao Y, Xiao X, Jiang YH, et al. Transcranial brain atlas-based optimization for functional near-infrared spectroscopy optode arrangement: Theory, algorithm, and application. Hum Brain Mapp 2021;42:1657-69. [Crossref] [PubMed]
24. Scholkmann F, Spichtig S, Muehlemann T, et al. How to detect and reduce movement artifacts in near-infrared imaging using moving standard deviation and spline interpolation. Physiol Meas 2010;31:649-62. [Crossref] [PubMed]
25. Wang J, Dong Q, Niu H. The minimum resting-state fNIRS imaging duration for accurate and stable mapping of brain connectivity network in children. Sci Rep 2017;7:6461. [Crossref] [PubMed]
26. Pinti P, Merla A, Aichelburg C, et al. A novel GLM-based method for the Automatic IDentification of functional Events (AIDE) in fNIRS data recorded in naturalistic environments. Neuroimage 2017;155:291-304. [Crossref] [PubMed]
27. Chen X. False discovery rate control for multiple testing based on discrete p-values. Biom J 2020;62:1060-79. [Crossref] [PubMed]
28. Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. Connecticut Appleton & Lange 2013;53.
29. Roth RH, Ding JB. Cortico-basal ganglia plasticity in motor learning. Neuron 2024;112:2486-502. [Crossref] [PubMed]
30. Rehme AK, Eickhoff SB, Grefkes C. State-dependent differences between functional and effective connectivity of the human cortical motor system. Neuroimage 2013;67:237-46. [Crossref] [PubMed]
31. Woodward KE, Carlson HL, Kuczynski A, et al. Sensory-motor network functional connectivity in children with unilateral cerebral palsy secondary to perinatal stroke. Neuroimage Clin 2019;21:101670. [Crossref] [PubMed]
32. Qin Y, Li Y, Sun B, et al. Functional Connectivity Alterations in Children with Spastic and Dyskinetic Cerebral Palsy. Neural Plast 2018;2018:7058953. [Crossref] [PubMed]
33. Vallinoja J, Nurmi T, Jaatela J, et al. Functional connectivity of sensorimotor network is enhanced in spastic diplegic cerebral palsy: A multimodal study using fMRI and MEG. Clin Neurophysiol 2024;157:4-14. [Crossref] [PubMed]
34. Lee JD, Park HJ, Park ES, et al. Motor pathway injury in patients with periventricular leucomalacia and spastic diplegia. Brain 2011;134:1199-210. [Crossref] [PubMed]
35. Laporta-Hoyos O, Pannek K, Pagnozzi AM, et al. White matter microstructure and receptive vocabulary in children with cerebral palsy: The role of interhemispheric connectivity. PLoS One 2023;18:e0280055. [Crossref] [PubMed]
36. Yao Q, Tang F, Wang Y, et al. Effect of cerebellum stimulation on cognitive recovery in patients with Alzheimer disease: A randomized clinical trial. Brain Stimul 2022;15:910-20. [Crossref] [PubMed]
37. Huang W, Li X, Xie H, et al. Different Cortex Activation and Functional Connectivity in Executive Function Between Young and Elder People During Stroop Test: An fNIRS Study. Front Aging Neurosci 2022;14:864662. [Crossref] [PubMed]