A Neurofunctional Study of Tactile Feedback in Stroke Rehabilitation Based on fNIRS

Academic Background

Stroke is a common neurological condition that profoundly affects patients’ daily lives and quality of life. Among the various functional impairments caused by stroke, hand dysfunction is particularly notable, manifesting in decreased muscle strength and severe limitations in finger movement control. These problems not only hinder a patient’s ability to perform basic life skills but also significantly reduce social participation and overall quality of life. Although traditional motor rehabilitation training has been shown to improve motor functions to some extent, over half of stroke patients continue to experience residual hand motor impairments even after rehabilitation.

In recent years, research has shown that motor rehabilitation incorporating tactile feedback (TF) holds promise as an effective intervention. By providing real-time tactile or visual information, TF helps patients perceive and adjust their movements, thereby enhancing their engagement and the efficacy of motor training. However, the neural mechanisms underlying TF in stroke rehabilitation remain insufficiently understood, particularly its effects on hand motor recovery and neuroplasticity. Investigating the impact of TF on brain-region activation and network connectivity could provide critical insights for developing personalized and more effective rehabilitation interventions.

Source of the Paper

This study was conducted by researchers Lingling Chen, Fanyao Meng, Congcong Huo, and others, from institutions including the School of Artificial Intelligence at Hebei University of Technology and the National Research Center for Rehabilitation Technical Aids. The paper was published in the February 1, 2025 issue of the journal Biomedical Optics Express, as part of the open-access articles.

Study Details

Research Process

The study primarily explored the relationship between tactile feedback and brain functional responses by using functional near-infrared spectroscopy (fNIRS) to evaluate brain-region activation and functional connectivity during hand-grasping tasks performed by stroke patients. The research process included the following steps:

1. Recruitment and Evaluation of Participants

The study recruited 15 stroke patients (in the subacute recovery phase), who met the following inclusion criteria: - Aged between 30 and 80 years; - History of unilateral subcortical stroke, at least 30 days post-onset; - Ability to comprehend and execute instructions; - First-ever stroke cases.

For comparative analysis, 15 age- and gender-matched healthy controls were also included. Hand motor function in patients was assessed using the Action Research Arm Test (ARAT), while basic demographic information and clinical characteristics (e.g., stroke type and lesion location) were recorded.

2. Experimental Design and Task Division

The experiment was designed with two types of grasping tasks: without tactile feedback (No-TF) and with tactile feedback (TF). The experiment followed a block design, where each block consisted of 30 seconds of grasping followed by 30 seconds of rest, repeated for 6 total blocks over a duration of 6 minutes. During the tasks, participants used a tactile feedback system to record grip force changes in real time, which were visualized on a screen. In the TF task, participants received both visual and auditory feedback when their grip force reached a pre-defined threshold, whereas no feedback was provided during the No-TF task.

3. Data Collection and Analysis

Brain activity data were collected using a multi-channel fNIRS device, covering brain regions such as the prefrontal cortex (PFC), premotor area (PMA), supplementary motor area (SMA), primary motor cortex (M1), and occipital lobe (OL). To minimize the impact of motion artifacts, the data underwent preprocessing steps like artifact correction and bandpass filtering. A general linear model (GLM) was used to analyze task-induced cortical activation responses.

Functional connectivity (FC) analysis was based on Pearson correlation matrices to construct brain functional networks. To enhance the robustness of statistical analysis, Fisher’s R-to-Z transformations were applied. Several network metrics were calculated, including clustering coefficient ©, global efficiency (GE), and transitivity (T).

---

Key Findings

1. Cortical Activation During Grasping Tasks

Stroke patients exhibited significantly lower brain activation levels compared to healthy controls during both types of grasping tasks: - In the TF grasping condition, stroke patients showed markedly reduced activation in the right prefrontal cortex (RPFC), left sensorimotor cortex (LSMC), and right premotor area (RPMA) compared to healthy controls. - Healthy participants demonstrated significantly broader brain activation, particularly in frontal and motor-related regions, during the TF task.

Furthermore, TF conditions led to higher activation levels in the right prefrontal cortex of stroke patients, suggesting that these patients may need to engage more cognitive resources for complex grasping tasks.

2. Functional Connectivity and Network Characteristics

• Functional Connectivity: In the TF grasping condition, healthy controls displayed significant increases in FC between frontal and motor areas, while such connectivity enhancements were limited in stroke patients, particularly in motor-related pathways.
• Brain Network Metrics: Stroke patients exhibited significantly decreased clustering coefficient and transitivity during TF grasping compared to healthy controls, indicating compromised network efficiency and local network structures post-stroke.

3. Behavioral Metrics and Network Characteristics Correlation

The study found that brain structural characteristics influencing grip strength correlated with recovery status in stroke patients: - The ratio of grip strength between the affected and unaffected sides was strongly correlated with ARAT scores (r=0.8447, p=0.0003). - Clustering coefficient (r=0.592, p=0.033) and transitivity (r=0.590, p=0.034) also positively correlated with grip strength ratios, highlighting the critical role of modularity and information integration efficiency in stroke recovery.

---

Significance and Value of the Study

1. Scientific Value

This study combines fNIRS and TF to open new perspectives on stroke hand rehabilitation research. Analyzing the relationship between network characteristics and grip strength advances the understanding of post-stroke neural reorganization mechanisms.

2. Practical Implications

The use of tactile feedback to enhance brain activation and cognitive engagement provides theoretical grounds for developing personalized rehabilitation strategies. The portable and cost-effective approach also makes it conveniently suitable for home-based recovery programs.

3. Research Highlights

• Innovative combination of TF and fNIRS analysis;
• Identification of the core role of FC and network features in stroke rehabilitation;
• Novel insights into the neural mechanisms underlying post-stroke hand dysfunction recovery.

---

Conclusion

This study demonstrated that tactile feedback modulates brain functional responses and network characteristics during grasping tasks in stroke patients, effectively improving behavioral outcomes. The research highlights the potential of tactile feedback as a promising tool for rehabilitating stroke patients. These findings offer crucial references for the development and optimization of future rehabilitation devices while providing new vigor to the field of stroke recovery research.