ABSTRACT
This study investigates cortical reorganisation and hemodynamic responses in individuals with lower extremity amputation and replantation using functional near‐infrared spectroscopy (fNIRS). A total of 15 healthy controls, four left lower limb amputees and one replantation patient were included. Oxyhemoglobin (oxy‐Hb) activations were measured during 10 unilateral lower limb motor tasks (toe, ankle, knee and hip movements). Non‐parametric analyses revealed significant differences in cortical activation between amputees and controls, particularly during knee flexion and extension. Three‐dimensional contrast maps demonstrated that oxy‐Hb activity in amputees extended from the M1‐leg area into somatosensory regions, reflecting neuroplastic remapping. In contrast, the replantation patient exhibited activation patterns closer to the control group, especially in knee and hip tasks. These findings indicate that fNIRS can sensitively capture hemispheric dynamics during unilateral lower limb movements and highlight neuroplastic adaptations following amputation and replantation. Such insights may guide future neuroprosthetic design and rehabilitation strategies.
Keywords: fNIRS, lower extremity amputation, neuroplasticity, oxy‐Hb, replantation
Summary.
fNIRS was used to assess cortical hemodynamic responses during lower limb motor tasks in amputees and a replantation patient.
Significant hemispheric activation differences were observed between amputees and healthy controls, especially during knee movements.
Amputees showed extended oxy‐Hb activation from the M1‐leg region to somatosensory areas, indicating neuroplastic reorganization.
Findings highlight fNIRS sensitivity in detecting cortical adaptations that may inform neuroprosthetic design and rehabilitation strategies.
1. Introduction
The number of patients with traumatic limb injuries is increasing due to the rise in motor vehicle accidents, natural disasters and malignancies [1]. Amputation is defined as the complete or partial loss of a limb resulting from injuries that may be traumatic, vascular, oncological, or infectious in origin [2]. Replantation surgery involves the reattachment of severed body parts following trauma. The replantation of a traumatically amputated lower extremity is a known surgical procedure [3].
The human motor system performs a series of complex movement sequences by dynamically controlling multiple limbs, timings, external stimuli and following various trajectories [4]. After lower extremity amputation, structural asymmetry can lead to difficulties in motor control and coordination [5]. Consequently, movement dynamics may change at various levels following amputations [6]. To a certain extent, the brain has an extraordinary capacity for self‐repair [7, 8, 9]. Neuroplasticity is defined as the reorganisation of the brain's structure, function and connectivity within the central nervous system in response to incidents [8]. Partial severing of the peripheral nervous system due to amputation results in changes in brain dynamics and muscle activation as the nervous system undergoes restructuring [10]. Additionally, following lower extremity amputation, the rehabilitation period and prosthesis application aim to gradually restore the functionality of amputees [11, 12], which may lead to functional reorganisation in the sensory and motor areas of the brain.
Detailed analysis of brain functions is essential when discussing the effects of amputation. Amputation affects limb representation in the primary motor and somatosensory cortex. The primary motor cortex, primary somatosensory cortex, supplementary motor area, basal ganglia, thalamus and cerebellum connections are reduced. These changes in connectivity lead to a decrease in motor control and balance, making daily activities more challenging for individuals with lower extremity amputation [13].
These structural and functional changes can be interpreted as maladaptive reorganisation of the brain due to sensory deprivation and loss or reduced ability to autonomously control movements [14, 15]. Moreover, while studies have examined functional reorganisation of the brain following upper extremity amputation, studies focused on understanding brain changes in individuals with lower extremity amputation remain limited [13]. Understanding these neural mechanisms is critically important for providing theoretical knowledge that can aid in the development of new prosthetics and rehabilitation techniques.
Therefore, understanding brain plasticity in amputees and achieving more precise outcomes requires advanced technologies that can identify the mechanisms involved in post‐amputation sensorimotor cortex neural activity. In our study, we used functional near‐infrared spectroscopy (fNIRS), which has higher temporal resolution than functional magnetic resonance imaging (fMRI) and better spatial resolution than electroencephalography (EEG) [16]. fNIRS is an optical imaging method that takes advantage of the low absorption of light by biological tissues in the near‐infrared spectrum to non‐invasively measure changes in cerebral haemoglobin concentrations through intact human scalp and skull [17]. Compared to other techniques, fNIRS has a significant advantage in lower extremity studies due to its portability and resistance to motion artefacts [18, 19, 20, 21]. The ability of fNIRS to provide a low‐cost alternative to fMRI by measuring changes in oxyhemoglobin (HbO), deoxyhemoglobin (HbR) and total haemoglobin (HbT) concentrations and to evaluate whole‐brain hemodynamics is of great importance. Additionally, fNIRS has shown the highest correlation with fMRI BOLD measurements and has provided reliable results [22]. The real‐world applicability of fNIRS allows for more realistic results. In recent years, fNIRS has been used to study activation patterns in the human cerebral cortex during exercise [23, 24].
When fNIRS optodes are placed on the scalp, monitoring hemodynamic responses allows for an indirect estimation of cortical activity, so the placement of channels consisting of source and detector pairs is particularly important. It is known that the lower body area of the Penfield motor cortex is located on the medial surface of each hemisphere, extending downward into the longitudinal fissure [25]. However, it is unclear whether fNIRS has sufficient spatial resolution to distinguish adjacent regions of the cortex in understanding lower extremity movements. In our study, we determined an optode configuration that includes the foot area, which was not covered in most of the studies in the literature due to doubts about the sensitivity of fNIRS. A recent study by MacLennan et al. on healthy subjects emphasizes that fNIRS has the necessary sensitivity to measure hemispheric activity during unilateral lower body movement [26]. This finding, indicates that fNIRS has sufficient spatial resolution over the distinguishable longitudinal fissure during knee extension, knee flexion, plantarflexion and dorsiflexion movements (included in our experimental procedure). In addition to MacLennan et al., our study also includes hip and toe movements. Therefore, the inclusion of 10 motor movements in the study is critical for understanding the competence of fNIRS in lower extremity analysis.
Our study is designed to evaluate the cortical activation caused by primary movements of the right and left lower extremities in a wide configuration that includes the middle surface of each hemisphere and the somatosensory cortex in control, amputee and replant patients. Especially in this study, conducted with special patient groups such as amputees and replant patients, we have demonstrated the effectiveness of fNIRS in different motor movements in the lower extremities. Additionally, in this study, we have discussed in detail the HbO‐based reorganisation of cortical organisation in lower extremity amputee patients and replant patients. Our findings are likely to contribute to fNIRS‐brain‐computer interface research and lower extremity neuroprosthetics studies.
2. Materials and Methods
2.1. Participants
The study included a control group consisting of 15 healthy individuals (10 males, 5 females) aged between 23 and 45 years (32.33 ± 6.81), with no history of neurological, cardiorespiratory, or lower extremity disorders and who were right‐leg dominant. Additionally, the study involved one patient with left lower extremity replantation (with only the small toe intact; the other fingers were amputated) and four patients with different levels of left lower extremity impairment (Tables 1a and 1b). All amputation and replantation cases occurred as a result of traumatic events leading to sudden limb loss. The amputee participants were actively using prostheses at the time of inclusion in the study (except A3) and had completed their standard rehabilitation programs. In addition, the patient who underwent ankle replantation does not use a prosthesis.
TABLE 1a.
Subject information.
| Patient code name | Amputee/replant | Sex | Duration of loss (years) | Lost limb |
|---|---|---|---|---|
| R1 | Replant and amputee | M | 11 | Left leg, toes (replanted below ankle) |
| A1 | Amputee | M | 2 | Left leg; below knee |
| A2 | Amputee | M | 6 | Left leg; below knee |
| A3 | Amputee | F | 3 | Left leg; toes |
| A4 | Amputee | F | 4 | Left leg; knee |
TABLE 1b.
Patient photos.
| R1 | A1 | A2 | A3 | A4 |
|---|---|---|---|---|
|
|
|
|
|
After being informed about the experimental procedures, the participants provided written consent. Ethical approval for the study was obtained from the Akdeniz University, Ethics Committee on 23/10/2019, with the reference number 2012‐KAEK‐20/996.
2.2. fNIRS Data Recording
In this study, fNIRS was used for cortical analysis in patients with left lower extremity amputation and replantation. Recordings were conducted at the Akdeniz University Neuroscience Laboratory using a portable continuous‐wave NIRSport system (NIRSport2 NIRx Medical Technologies, Berlin, Germany) with two wavelengths (760 and 850 nm). Data acquisition was performed with a sampling frequency of 7.81 Hz, a 4 Hz low‐pass filter and 3.0 μV < dark noise < 15.0 μV.
2.3. fNIRS Channel Locations
The placement of optodes was carried out using the NIRsite software (NIRx Medical Technologies LLC, LA, USA) with automatic anatomical labeling based on the Brodmann area (BA) atlas [27]. According to the fNIRS Optode's Location Decider (fOLD) toolbox [28], the specificity levels are 20% for the primary somatosensory cortex (1–2–3), 19% for the primary motor cortex (4), 10% for the somatosensory association cortex (5) and 62% for the premotor and supplementary motor areas (6). The optodes primarily covered the primary somatosensory cortex, primary motor cortex, somatosensory association cortex, premotor and supplementary motor areas, representing BAs 1–6 (Figure 1). Brain areas not related to the study, such as auditory, frontal and visual areas, were excluded from this configuration. The system consists of 16 dual‐type NIRS optodes, with 8 sources and 8 detectors, forming 21 channels. Table 2 provides the fNIRS brain regions of interest for the 21 connections. The optodes were placed on a cap (EasyCap, GmbH, Germany) based on the 10–20 international standardised EEG system [29]. The source‐detector separation is approximately 3 cm apart. To ensure the correct positioning of the measurement cap, the Cz location was used as a reference point, and the cap was centered between the nasion and inion, as well as between the left and right preauricular points.
FIGURE 1.
(a) Experimental procedure, (b) shows the positions of optodes sources and detectors on the scalp. Labels starting with S (S01, S02, …, S08) (in red) represent the sources. Labels starting with D (D01, D02, …, D08) (in blue) represent the detectors. The channels, consisting of 21 source‐detector pairs, are represented by lines.
TABLE 2.
fNIRS brain regions of interest.
| Channels | Source, detector | Optodes | Anatomic brain region |
|---|---|---|---|
| 1 | S01, D01 | Cz‐FCz | Premotor and supplementary motor area |
| 2 | S01, D02 | Cz‐CPz | Primary motor cortex |
| 3 | S01, D05 | Cz‐C1 | Primary motor cortex |
| 4 | S01, D06 | Cz‐C2 | Primary motor cortex |
| 5 | S02, D02 | Pz‐CPz | Somatosensory association cortex |
| 6 | S03, D02 | CP1‐ Pz | Somatosensory association cortex |
| 7 | S03, D03 | CP1‐CP3 | Somatosensory association cortex—primary motor cortex |
| 8 | S03, D05 | CP1‐C1 | Somatosensory association cortex—primary motor cortex |
| 9 | S04, D02 | CP2‐ Pz | Primary motor cortex—somatosensory association cortex |
| 10 | S04, D04 | CP2‐CP4 | Primary motor cortex—somatosensory association cortex |
| 11 | S04, D06 | CP2‐C2 | Primary motor cortex—somatosensory association cortex |
| 12 | S05, D01 | FC1‐FCz | M1‐leg |
| 13 | S05, D05 | FC1‐C1 | M1‐leg |
| 14 | S06, D01 | FC2‐Cz | M1‐leg |
| 15 | S06, D06 | FC2‐C2 | M1‐leg |
| 16 | S07, D03 | C3‐CP3 | Primary somatosensory cortex |
| 17 | S07, D05 | C3‐C1 | Primary somatosensory cortex |
| 18 | S07, D07 | C3‐C5 | Primary somatosensory cortex |
| 19 | S08, D04 | C4‐ CP4 | Primary somatosensory cortex |
| 20 | S08, D06 | C4‐C2 | Primary somatosensory cortex |
| 21 | S08, D08 | C4‐C6 | Primary somatosensory cortex |
2.4. Experimental Setup and Study Design
fNIRS recordings were conducted for the control group and patient groups with different levels of lower left extremity amputation and replantation during the movements of right toe extension (RTE) and flexion (RTF), left toe extension (LTE) and flexion (LTF), right ankle extension (RFE) and flexion (RFF), left ankle extension (LFE) and flexion (LFF), right knee extension (RKE) and flexion (RKF), left knee extension (LKE) and flexion (LKF), right hip abduction (RHB) and adduction (RHD) and left hip abduction (LHB) and adduction (LHD). (Abbreviations for the movements used in the study will be used throughout the remainder of the article.) Patients were asked to perform each movement at the maximum contraction level they could achieve.
The experimental procedure was as follows: 5 s of empty sequence, 8 s of movement, 8 s of rest, 8 s of the second repetition of the movement, 8 s of rest and 5 s of empty sequence. Additionally, a user interface was designed in MATLAB for participants to follow the experimental procedures on a computer screen. Within this interface, the experimental procedures common to all recordings were automatically initiated and terminated at the specified times. fNIRS data were collected with participants seated comfortably, without muscle tension and with their feet not touching the ground. Additionally, the automatic command screen was positioned 2.5 m away. The experimental procedure and the representation of the 21 channels consisting of optode source‐detector pairs on the brain are shown in Figure 1.
2.5. fNIRS Data Processing
The HOMER3 MATLAB toolbox [30] and MATLAB R2017b were used for data processing. First, the optical density (OD) signals were specified according to the time intervals defined in the experimental procedure. Then, the spline and splineSG methods were applied to the raw OD recordings to detect initial head motion artefacts, followed by a wavelet correction specifically designed for sudden artefacts on the OD data [31]. The processed data were then converted into HbO/HbR concentration changes using the Beer–Lambert Law. The reflection and visualisation of statistically significant oximetric Hb values on the brain surface were performed using the AtlasViewer MATLAB toolbox, which utilises the ‘Colin27’ digital brain atlas [32, 33].
2.6. Statistical Analysis
In the study, statistical analyses between the fNIRS oxy‐Hb values of the control group and the left lower limb amputee patient group were conducted using SPSS 23 (SPSS Inc. Released 2021. PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc.). To identify differences in cortical activations between the control group and the left lower limb amputee patient group (since the number of amputee patients was less than 10), the Mann–Whitney U test, a non‐parametric test, was used.
For the statistical analysis involving the single patient with a replanted left ankle who only retained the small toe and was included in the study, the Crawford t test [34] was applied. Crawford et al. [35] suggest that this test is particularly useful in single‐case studies where the control samples being compared typically have a small N value.
A significance level of 95% (or an α = 0.05 margin of error) was used to determine differences in the analyses, and p ≤ 0.05 was considered significant. Additionally, descriptive statistics (mean, standard deviation) were provided for the demographic characteristics of all groups included in the study.
3. Results
The alterations in oxy‐Hb, deoxy‐Hb and total‐Hb concentrations were calculated for all movements. Given that the oxy‐Hb signal data exhibited a more pronounced activation than the deoxy‐Hb signal data [36, 37], all subsequent analyses were conducted exclusively on the oxy‐Hb data.
3.1. Statistical Results
No differences were observed in finger movements (RTE and RTF) and abduction‐adduction movements (RHB, RHD, LHB and LHD) between the control group and the left lower limb amputee patient group. In the RFE movement, except for the M1‐leg (FC2‐C2) region, and in the RFF movement, except for the somatosensory association cortex (CP1‐CPz), the results were similar to the control group, as shown in Table 3. There was no statistically significant difference in the RKF movement. However, in the RKE movement, differences were observed in the premotor and supplementary motor areas (Cz‐FCz), the primary motor cortex (Cz‐CPz), the somatosensory association cortex (CP1‐CPz) and the M1‐leg (FC1‐FCz, FC2‐C2) regions. In the RHF movement, a statistically significant difference was found in the primary motor cortex (Cz‐C2) between the control and amputee groups. The most statistically significant differences were found in the LKF movement in the M1‐leg (FC1‐C1) and the primary somatosensory cortex (C3‐C5, C4‐C2, C4‐CP4, C4‐C2, C4‐C6) regions. In the LHE movement, a statistically significant difference was found in the primary somatosensory cortex (C3‐C1) region. The left‐side movements (LKF‐LHE) in left amputee patients, which showed differences, exhibited maximum activations in the somatosensory areas (Table 3).
TABLE 3.
Comparison of left lower limb amputee patients with the control group (p < 0.05).
| Movements | Cz‐FCz | Cz‐CPz | Cz‐C1 | Cz‐C2 | Pz‐CPz | CP1‐CPz | CP1‐CP3 | CP1‐C1 | CP2‐CPz | CP2‐CP4 | CP2‐C2 | FC1‐FCz | FC1‐C1 | FC2‐FCz | FC2‐C2 | C3‐CP3 | C3‐C1 | C3‐C5 | C4‐CP4 | C4‐C2 | C4‐C6 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RTE | 0.194 | 0.317 | 0.162 | 0.134 | 0.194 | 0.484 | 0.484 | 0.134 | 0.764 | 0.230 | 0.072 | 0.162 | 0.057 | 0.230 | 0.110 | 0.194 | 0.194 | 0.194 | 0.134 | 0.162 | 0.368 |
| RTF | 0.194 | 0.841 | 0.089 | 0.230 | 0.271 | 0.549 | 0.549 | 0.638 | 0.617 | 0.230 | 0.368 | 0.134 | 0.072 | 0.424 | 0.549 | 0.162 | 0.110 | 0.072 | 0.424 | 0.368 | 0.368 |
| RFE | 0.424 | 0.368 | 0.424 | 0.089 | 0.484 | 1 | 0.368 | 0.194 | 0.484 | 0.368 | 0.271 | 0.764 | 0.134 | 0.134 | 0.036* | 0.230 | 0.089 | 0.057 | 0.230 | 0.089 | 0.617 |
| RFF | 0.072 | 0.230 | 0.764 | 0.549 | 0.057 | 0.036* | 0.134 | 0.271 | 0.057 | 0.194 | 0.484 | 0.134 | 0.459 | 0.134 | 0.230 | 0.689 | 0.920 | 1 | 0.230 | 0.134 | 0.317 |
| RKE | 0.021* | 0.007* | 0.194 | 0.271 | 0.271 | 0.046* | 0.317 | 0.920 | 0.134 | 0.162 | 1 | 0.021* | 0.617 | 0.072 | 0.021* | 0.134 | 0.230 | 0.484 | 0.271 | 0.549 | 1 |
| RKF | 0.920 | 0.920 | 0.317 | 0.368 | 0.424 | 1 | 1 | 0.764 | 0.368 | 0.110 | 0.271 | 0.484 | 0.617 | 0.920 | 0.271 | 0.764 | 0.110 | 0.841 | 0.689 | 0.134 | 0.162 |
| RHE | 0.549 | 0.217 | 0.368 | 0.484 | 0.110 | 0.484 | 0.549 | 0.689 | 0.484 | 0.617 | 0.484 | 0.764 | 0.134 | 0.617 | 0.841 | 0.484 | 0.317 | 0.230 | 0.110 | 0.368 | 0.424 |
| RHF | 0.544 | 0.396 | 0.467 | 0.039* | 0.146 | 0.716 | 0.182 | 0.090 | 0.544 | 0.275 | 0.090 | 0.544 | 0.808 | 0.467 | 0.182 | 0.090 | 0.182 | 0.225 | 0.716 | 0.544 | 0.903 |
| RHB | 0.368 | 0.549 | 0.484 | 0.368 | 0.841 | 0.689 | 0.689 | 0.841 | 0.999 | 0.484 | 0.368 | 0.162 | 0.999 | 0.194 | 0.617 | 0.764 | 0.689 | 0.617 | 0.920 | 0.424 | 0.764 |
| RHD | 0.314 | 0.441 | 0.374 | 0.859 | 0.260 | 0.594 | 0.678 | 0.767 | 0.594 | 0.515 | 0.859 | 0.859 | 0.953 | 0.441 | 0.515 | 0.767 | 0.859 | 0.594 | 0.374 | 0.767 | 0.594 |
| LKE | 0.999 | 0.484 | 0.424 | 0.549 | 0.617 | 0.230 | 0.368 | 0.194 | 0.920 | 0.689 | 0.271 | 0.841 | 0.689 | 0.841 | 0.999 | 0.920 | 0.271 | 0.841 | 0.841 | 0.230 | 0.764 |
| LKF | 0.194 | 0.841 | 0.841 | 0.689 | 0.549 | 0.920 | 0.764 | 0.317 | 0.134 | 0.230 | 0.230 | 0.999 | 0.028* | 0.841 | 0.424 | 0.549 | 0.134 | 0.028* | 0.046* | 0.046* | 0.016* |
| LHE | 0.764 | 0.689 | 0.368 | 0.089 | 0.920 | 0.920 | 0.920 | 0.089 | 0.134 | 0.841 | 0.617 | 0.317 | 0.110 | 0.230 | 0.317 | 0.317 | 0.036* | 0.162 | 0.841 | 0.162 | 0.689 |
| LHF | 0.057 | 0.424 | 0.689 | 0.549 | 0.368 | 0.424 | 0.194 | 0.689 | 0.764 | 0.920 | 0.317 | 0.424 | 0.689 | 0.764 | 0.689 | 0.999 | 0.271 | 0.920 | 0.920 | 0.317 | 0.689 |
| LHB | 0.457 | 0.671 | 0.524 | 0.111 | 0.832 | 0.915 | 0.750 | 0.457 | 0.832 | 0.595 | 0.396 | 0.457 | 0.750 | 0.750 | 0.750 | 0.915 | 0.750 | 0.595 | 0.457 | 0.339 | 0.288 |
| LHD | 0.167 | 0.671 | 0.203 | 0.595 | 0.457 | 0.243 | 0.595 | 0.999 | 0.524 | 0.750 | 0.288 | 0.915 | 0.396 | 0.832 | 0.524 | 0.832 | 0.339 | 0.288 | 0.671 | 0.243 | 0.915 |
p < 0.05.
In addition, the statistical results for the control group and the replanted patient (Table 4) revealed distinct cortical response characteristics across all movement conditions. While several movements such as RTE, RHE, RHF, RHD, LFE, LFF, LHE, LHF, LHB and LHD demonstrated localised activation differences limited to specific cortical areas, broader and more pronounced variations were detected during RKE, RHB, RTF and LKF movements. These results collectively illustrate that knee‐ and hip‐related motor tasks elicit stronger and more widespread cortical reorganisation patterns, consistent with the functional load and complexity of these movements in post‐traumatic lower‐limb adaptation.
TABLE 4.
Statistical comparison of the replant patient with the control group using the Crawford t test (p < 0.05).
| Movements | Cz‐FCz | Cz‐CPz | Cz‐C1 | Cz‐C2 | Pz‐CPz | CP1‐CPz | CP1‐CP3 | CP1‐C1 | CP2‐CPz | CP2‐CP4 | CP2‐C2 | FC1‐FCz | FC1‐C1 | FC2‐FCz | FC2‐C2 | C3‐CP3 | C3‐C1 | C3‐C5 | C4‐CP4 | C4‐C2 | C4‐C6 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RTE | 0.739 | 0.540 | 0.464 | 0.803 | 0.391 | 0.266 | 0.311 | 0.245 | 0.365 | 0.300 | 0.523 | 0.225 | 0.215 | 0.796 | 0.763 | 0.284 | 0.240 | 0.290 | 0.537 | 0.052* | 0.867 |
| RTF | 0.591 | 0.084 | 0.101 | 0.223 | 0.018* | 0.001* | 0.036* | 0.051 | 0.202 | 0.026* | 0.103 | 0.210 | 0.082 | 0.672 | 0.056* | 0.097 | 0.200 | 0.007* | 0.124 | 0.920 | 0.090 |
| RFE | 0.761 | 0.363 | 0.757 | 0.309 | 0.225 | 0.258 | 0.170 | 0.039* | 0.197 | 0.062 | 0.385 | 0.525 | 0.434 | 0.249 | 0.014* | 0.357 | 0.230 | 0.342 | 0.471 | 0.000* | 0.866 |
| RFF | 0.695 | 0.980 | 0.983 | 0.370 | 0.758 | 0.883 | 0.931 | 0.234 | 0.885 | 0.927 | 0.577 | 0.694 | 0.605 | 0.170 | 0.016* | 0.781 | 0.647 | 0.823 | 0.995 | 0.543 | 0.026* |
| RKE | 0.000* | 0.120 | 0.001* | 0.000* | 0.176 | 0.065 | 0.109 | 0.058 | 0.087 | 0.107 | 0.068 | 0.023* | 0.048* | 0.000* | 0.002* | 0.301 | 0.051* | 0.325 | 0.289 | 0.992 | 0.564 |
| RKF | 0.863 | 0.361 | 0.289 | 0.852 | 0.096 | 0.065 | 0.143 | 0.106 | 0.114 | 0.149 | 0.386 | 0.663 | 0.294 | 0.973 | 0.421 | 0.420 | 0.279 | 0.750 | 0.020* | 0.019* | 0.983 |
| RHE | 0.894 | 0.971 | 0.907 | 0.623 | 0.855 | 0.956 | 0.744 | 0.808 | 0.822 | 0.569 | 0.816 | 0.900 | 0.971 | 0.739 | 0.510 | 0.172 | 0.544 | 0.549 | 0.591 | 0.018* | 0.166 |
| RHF | 0.782 | 0.355 | 0.616 | 0.333 | 0.230 | 0.306 | 0.417 | 0.220 | 0.522 | 0.252 | 0.270 | 0.291 | 0.144 | 0.042* | 0.287 | 0.496 | 0.272 | 0.329 | 0.477 | 0.102 | 0.860 |
| RHB | 0.006* | 0.155 | 0.083 | 0.002* | 0.182 | 0.133 | 0.060 | 0.600 | 0.018 | 0.009* | 0.016* | 0.003* | 0.262 | 0.000* | 0.065 | 0.210 | 0.345 | 0.395 | 0.103 | 0.001* | 0.435 |
| RHD | 0.312 | 0.218 | 0.145 | 0.489 | 0.304 | 0.371 | 0.143 | 0.113 | 0.229 | 0.420 | 0.436 | 0.100 | 0.130 | 0.323 | 0.001* | 0.277 | 0.162 | 0.162 | 0.993 | 0.332 | 0.215 |
| LFE | 0.547 | 0.743 | 0.946 | 0.959 | 0.632 | 0.945 | 0.579 | 0.967 | 0.367 | 0.234 | 0.357 | 0.551 | 0.827 | 0.188 | 0.026* | 0.474 | 0.482 | 0.811 | 0.565 | 0.073 | 0.759 |
| LFF | 0.876 | 0.048* | 0.154 | 0.235 | 0.138 | 0.673 | 0.122 | 0.586 | 0.121 | 0.017 | 0.088 | 0.159 | 0.039* | 0.747 | 0.253 | 0.143 | 0.134 | 0.148 | 0.457 | 0.002* | 0.904 |
| LKE | 0.044* | 0.891 | 0.758 | 0.944 | 0.746 | 0.635 | 0.695 | 0.472 | 0.412 | 0.429 | 0.612 | 0.878 | 0.955 | 0.006* | 0.853 | 0.430 | 0.716 | 0.471 | 0.723 | 0.072 | 0.692 |
| LKF | 0.677 | 0.007* | 0.148 | 0.077 | 0.000* | 0.018* | 0.040* | 0.131 | 0.001* | 0.385 | 0.075 | 0.645 | 0.956 | 0.105 | 0.090 | 0.160 | 0.368 | 0.872 | 0.029* | 0.885 | 0.755 |
| LHE | 0.001* | 0.488 | 0.846 | 0.878 | 0.355 | 0.664 | 0.404 | 0.874 | 0.623 | 0.836 | 0.926 | 0.747 | 0.652 | 0.840 | 0.351 | 0.545 | 0.678 | 0.336 | 0.904 | 0.662 | 0.772 |
| LHF | 0.806 | 0.222 | 0.475 | 0.958 | 0.937 | 0.325 | 0.492 | 0.436 | 0.313 | 0.709 | 0.959 | 0.254 | 0.474 | 0.012* | 0.929 | 0.417 | 0.665 | 0.929 | 0.679 | 0.905 | 0.871 |
| LHB | 0.054* | 0.180 | 0.290 | 0.577 | 0.315 | 0.195 | 0.204 | 0.671 | 0.276 | 0.218 | 0.856 | 0.386 | 0.591 | 0.069 | 0.758 | 0.701 | 0.645 | 0.465 | 0.234 | 0.891 | 0.687 |
| LHD | 0.027* | 0.538 | 0.523 | 0.912 | 0.611 | 0.657 | 0.677 | 0.794 | 0.686 | 0.652 | 0.722 | 0.443 | 0.604 | 0.398 | 0.156 | 0.995 | 0.737 | 0.960 | 0.749 | 0.998 | 0.295 |
p < 0.05.
To facilitate interpretation and improve readability, the overall oxy‐Hb activation differences between the left lower‐limb amputee patients and the control group were summarised in Table 5. This summary highlights the key channels that showed statistically significant oxy‐Hb differences, with median values, quartile ranges, effect sizes (r) and p values presented for each motor condition. By consolidating these data, Table 5 provides a clearer overview of hemispheric and regional activation trends that complement the detailed channel‐level comparisons shown in Table 3. Together, these summarised results highlight the predominant involvement of motor and somatosensory regions in lower‐limb control, reflecting cortical adaptations following amputation.
TABLE 5.
Comparison of oxy‐Hb activation between the left lower‐limb amputee patients and the control group. Only statistically significant channels are shown. Values are median (Q1–Q3), effect size (r) and p values (p < 0.05).
| Motion | Channels | Median (Q1–Q3), control amputee | r (Effect size) | p | |
|---|---|---|---|---|---|
| RFE | HbO 6.6 |
1.31 (−1.02 to 1.79) |
3.72 (0.87–4.95) |
−0.482 | 0.036* |
| RFF | HbO 3.2 |
0.87 (−0.35 to 2.24) |
−1.51 (−3.22 to −0.36) |
−0.482 | 0.036* |
| RKE | HbO 1.1 |
1.12 (−1.16 to 1.93) |
−3.1 (−3.78 to −1.52) |
−0.528 | 0.021* |
| HbO 1.2 |
0.33 (−0.44 to 4.03) |
−3.9 (−6.37 to −1.6) |
−0.62 | 0.007* | |
| HbO 3.2 |
0.70 (−0.91 to 2.91) |
−8.93 (−27.09 to −0.25) |
−0.459 | 0.046* | |
| HbO 5.1 |
0.06 (−2.56 to 0.71) |
−8.35 (−16.56 to 1.8) |
−0.528 | 0.021* | |
| HbO 6.6 |
0.32 (−1.29 to 1.58) |
−3.76 (13.59 to −1.07) |
−0.528 | 0.021* | |
| RHF | HbO 1.6 |
−0.33 (−2.02 to 1.51) |
5.12 (1.33–10.17) |
−0.474 | 0.039* |
| LKF | HbO 5.5 |
0.85 (−0.16 to 2.13) |
2.66 (2.13–4.86) |
−0.505 | 0.028* |
| HbO 7.7 |
0.87 (0.16–2.2) |
4.97 (1.94–11.67) |
−0.505 | 0.028* | |
| HbO 8.4 |
0.99 (−1.05 to 2.83) |
4.98 (1.75–10.2) |
−0.459 | 0.046* | |
| HbO 8.6 |
1.10 (−1.18 to 1.9) |
4.42 (0.29–7.55) |
−0.459 | 0.046* | |
| HbO 8.8 |
0.77 (−0.88 to 1.9) |
5.48 (2.19–14.41) |
−0.551 | 0.016* | |
| HbO 7.5 |
−0.43 (−2.13 to 1.51) |
5.12 (1.33–10.17) |
−0.482 | 0.036* | |
p < 0.05.
3.2. Contrast Maps
As a result of the statistical analysis, the movements with channels showing differences between the left amputee patients and the healthy control group were visualised on a 3D brain model using the AtlasViewer software, corresponding to the standard channel positions.
According to Table 3, the contrast maps of movements with statistically significant differences detected in left amputee patients are presented in Table 6. Additionally, to observe the activation of the replanted patient in these movements, the contrast maps of the replanted patient have also been included in the table and compared. All movements' topographic maps were given in Supplementary Digital Content Figure 1.
TABLE 6.
Topographic maps of control, left amputee and left replant patients.
| Control | Left amputee | Left replant (R1) | |
|---|---|---|---|
| RFE |
|
|
|
| RFF |
|
|
|
| RKE |
|
|
|
| RHF |
|
|
|
| LKF |
|
|
|
| LHE |
|
|
|
When examining the topographic dominance of oxy‐Hb values in the control group, it is evident that the RKE, RFF, RFE, RHF, LKF and LHE movements show dominance on the medial surface of each hemisphere. In contrast, the topographic maps of left amputee patients reveal different dominance patterns compared to the control group. Even though the left lower limb is amputated, there are also differences in right motor movements. The widespread activation during the RFE and RFF movements in left amputee patients compared to the control group is clear. The greatest differences between the left lower limb amputee and control groups are observed in knee movements (RKE‐LKF).
It was found that the differences in left‐side movements (LKF‐LHE) in left lower limb amputee patients spread across somatosensory areas. Particularly, the LKF movement showed maximum activation in the somatosensory area of both hemispheres. Compared to the control group, the maximum and localised oxy‐Hb values observed in the M1‐leg area of the control group were found to be dispersed in the somatosensory areas of the left lower limb amputee group. In the RHF movement, although the maximum activity was in the M1‐leg area, the oxy‐Hb activity also spread to the somatosensory area.
In the left replanted patient (R1), dominant oxy‐Hb values similar to those in the control group were observed on the medial surface of each hemisphere during the LKF and LHE movements. Consequently, the left replanted patient's topographic map was closer to that of the control group.
The fact that the LHE movement involves a larger area of the brain compared to the LKF movement may explain the widespread activity in the control group, extending from the M1‐leg area to the primary motor cortex in each hemisphere. In the left amputee patients, although the LHE movement shows maximum activity in the M1‐leg and primary motor cortex areas of the right hemisphere, there is widespread oxy‐Hb activity across both hemispheres.
4. Discussion
While there have been studies examining brain functional reorganisation following upper limb amputation, the understanding of brain changes in individuals with lower limb amputation remains limited [13]. Although the present study focused on lower limb amputation and replantation, the oxy‐Hb activations in the cortex during lower limb motor movements in a control group of 15 healthy individuals were also examined. The observation that the Penfield motor cortex shows activation on the medial surface of each hemisphere, above the longitudinal fissure, indicates that fNIRS has the required sensitivity to measure hemispheric activity during unilateral lower body movement.
In the present study, the fNIRS oxy‐Hb values observed during right and left lower limb motor movements in four patients with a left lower limb amputation at different levels and one patient with a left ankle replant were statistically compared with those in a control group of 15 healthy volunteers. The results of the analysis demonstrated statistically significant differences in oxy‐Hb values between the left lower limb amputee patients and the healthy controls during the RFE, RFF, RKE, RHF, LKF and LHE movements.
A statistical comparison of the replant patient's oxy‐Hb values with those of the control group revealed that the differences in hip movements were limited to a single location. Furthermore, the patient exhibited oxy‐Hb values that were largely similar to those of the control group. It is noteworthy that there were differences among the control group channels in oxy‐Hb values during right and left knee movements, as illustrated in Tables 6 and 7. The changes in cortical activation triggered by the patient's recovery process in the replant region likely influenced the balance and control mechanisms, leading to a different cortical activation pattern in the healthy leg as the patient engaged in holistic motor movements focused on walking (Table 6).
TABLE 7.
Topographic maps of LFE and LFF motor movements for the left replant patient.
| LFE |
|
|
| LFF |
|
|
All amputation and replantation cases occurred as a result of traumatic events leading to sudden limb loss. The amputee participants were actively using prostheses at the time of inclusion in the study (except A3) and had completed their standard rehabilitation programs. In addition, the patient who underwent ankle replantation does not use a prosthesis. This clinical context indicates that the observed cortical activation patterns likely reflect the participants' functional adaptation stages following rehabilitation and prosthesis use.
The intact leg of the patient is of great importance for the maintenance of balance and control mechanisms. Structural asymmetry following lower limb amputation has been demonstrated to result in difficulties in motor control and coordination [5]. It can be therefore surmised that the loss of a limb may result in functional reorganisation within the sensory and motor regions of the brain. It has been observed that lower limb amputation results in the remapping of cortical topography, with the expansion of stump activation maps in the primary sensorimotor cortex of the deafferented hemisphere, spreading into adjacent areas representing the trunk and upper limbs [38]. The topographic results during motor movements in this study of amputee patients also demonstrate a clear expansion of the primary sensorimotor cortex area, which extends into the somatosensory cortex. Furthermore, in the topographic maps of the LFE and LFF movements in the patient who had undergone replantation of the ankle, a comparable distribution of the primary sensorimotor cortex was identified, as observed in amputee patients (Table 7). This suggests that lower limb amputation and replantation also lead to neuroplastic changes in the primary sensorimotor cortex. Therefore, dynamic changes may occur following amputation and replantation procedures at various levels of the lower limb.
The findings of this study indicate that fNIRS imaging systems are capable of measuring hemispheric activity during unilateral lower body movements with sufficient sensitivity. Furthermore, our findings underscore the plastic changes induced by lower limb amputation and replantation.
These findings demonstrate that fNIRS can sensitively detect hemispheric changes, particularly during knee flexion and extension movements. This presents significant potential for the development of neuroprostheses targeting the lower extremities and for the optimization of rehabilitation programs. However, to more strongly support the study's claims, additional quantitative data from larger samples and studies that methodologically reveal levels of sensitivity and specificity are needed. While the small sample size in the present study is acknowledged, this limitation restricts the generalizability and functional interpretability of the findings.
Ethics Statement
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (2012‐KAEK‐20/996, 23.10.2019) and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Consent
Informed consent was obtained from all participants.
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgements
This study is funded by a grant from the Scientific and Technological Research Council of Turkey (Project No: 220S026).
Süzen E., Özkan Ö., Özkan Ö., et al., “Determination of Hemodynamic Response Using fNIRS in Lower Extremity Amputee and Replant Patients,” International Wound Journal 22, no. 11 (2025): e70792, 10.1111/iwj.70792.
Funding: This work was supported by a grant from the Scientific and Technological Research Council of Turkey (Project no: 220S026) and by Akdeniz University, Scientific Research Projects Supporting Unit.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Barbosa B. M. B., Monteiro R. A., Sparano L. F., Bareiro R. F. N., Passos A. D. C., and Engel E. E., “Incidence and Causes of Lower‐Limb Amputations in the City of Ribeirão Preto From 1985 to 2008: Evaluation of the Medical Records From 3,274 Cases,” Revista Brasileira de Epidemiologia 19, no. 2 (2016): 317–325. [DOI] [PubMed] [Google Scholar]
- 2. Zambudio Periago R., Prótesis, Órtesis y Ayudas Técnicas (Elsevier Masson, 2009). [Google Scholar]
- 3. Gayle L. B., Lineaweaver W. C., Buncke G. M., et al., “Lower Extremity Replantation,” Clinics in Plastic Surgery 18, no. 3 (1991): 437–447, 10.1016/S0094-1298(20)30842-7. [DOI] [PubMed] [Google Scholar]
- 4. Wolpert D. M. and Ghahramani Z., “Computational Principles of Movement Neuroscience,” Nature Neuroscience 3, no. 11 (2000): 1212–1217. [DOI] [PubMed] [Google Scholar]
- 5. Rueda F. M., Diego I. M. A., Sánchez A. M., Tejada M. C., Montero F. M. R., and Page J. C. M., “Knee and Hip Internal Moments and Upper‐Body Kinematics in the Frontal Plane in Unilateral Transtibial Amputees,” Gait & Posture 37, no. 3 (2013): 436–439. [DOI] [PubMed] [Google Scholar]
- 6. Nasri A., Abbasi A., Hadavi Z., Abbasi S., and Svoboda Z., “Lower‐Extremity Inter‐Joint Coordination Variability in Active Individuals With Transtibial Amputation and Healthy Males During Gait,” Scientific Reports 14, no. 1 (2024): 11668, 10.1038/s41598-024-62655-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Chen L. C., Sandmann P., Thorne J. D., Herrmann C. S., and Debener S., “Association of Concurrent fNIRS and EEG Signatures in Response to Auditory and Visual Stimuli,” Brain Topography 28, no. 5 (2015): 710–725, 10.1007/s10548-015-0424-8. [DOI] [PubMed] [Google Scholar]
- 8. Pascual‐Leone A., Amedi A., Fregni F., and Merabet L. B., “The Plastic Human Brain Cortex,” Annual Review of Neuroscience 28, no. 1 (2005): 377–401, 10.1146/annurev.neuro.27.070203.144216. [DOI] [PubMed] [Google Scholar]
- 9. Pascual‐Leone A., Freitas C., Oberman L., et al., “Characterizing Brain Cortical Plasticity and Network Dynamics Across the Age‐Span in Health and Disease With TMS‐EEG and TMS‐fMRI,” Brain Topography 24, no. 3–4 (2011): 302–315, 10.1007/s10548-011-0196-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Williams L., Pirouz N., Mizelle J., Cusack W., Kistenberg R., and Wheaton L. A., “Remodeling of Cortical Activity for Motor Control Following Upper Limb Loss,” Clinical Neurophysiology 127, no. 9 (2016): 3128–3134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Soares A. S. O. d. C., Yamaguti E. Y., Mochizuki L., Amadio A. C., and Serrão J. C., “Biomechanical Parameters of Gait Among Transtibial Amputees: A Review,” Sao Paulo Medical Journal 127 (2009): 302–309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Rusaw D. and Ramstrand N., “Motion‐Analysis Studies of Transtibial Prosthesis Users: A Systematic Review,” Prosthetics and Orthotics International 35, no. 1 (2011): 8–19. [DOI] [PubMed] [Google Scholar]
- 13. Molina‐Rueda F., Navarro‐Fernández C., Cuesta‐Gómez A., Alguacil‐Diego I. M., Molero‐Sánchez A., and Carratalá‐Tejada M., “Neuroplasticity Modifications Following a Lower‐Limb Amputation: A Systematic Review,” PM & R 11, no. 12 (2019): 1326–1334, 10.1002/pmrj.12167. [DOI] [PubMed] [Google Scholar]
- 14. MacIver K., Lloyd D., Kelly S., Roberts N., and Nurmikko T., “Phantom Limb Pain, Cortical Reorganization and the Therapeutic Effect of Mental Imagery,” Brain 131, no. 8 (2008): 2181–2191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Raffin E., Richard N., Giraux P., and Reilly K. T., “Primary Motor Cortex Changes After Amputation Correlate With Phantom Limb Pain and the Ability to Move the Phantom Limb,” NeuroImage 130 (2016): 134–144. [DOI] [PubMed] [Google Scholar]
- 16. Quaresima V. and Ferrari M., “Functional Near‐Infrared Spectroscopy (fNIRS) for Assessing Cerebral Cortex Function During Human Behavior in Natural/Social Situations: A Concise Review,” Organizational Research Methods 22, no. 1 (2019): 46–68. [Google Scholar]
- 17. Boas D. A., Elwell C. E., Ferrari M., and Taga G., “Twenty Years of Functional Near‐Infrared Spectroscopy: Introduction for the Special Issue,” NeuroImage 85 (2014): 1–5. [DOI] [PubMed] [Google Scholar]
- 18. Hoshi Y., “Functional Near‐Infrared Spectroscopy: Potential and Limitations in Neuroimaging Studies,” International Review of Neurobiology 66 (2025): 237–266, 10.1016/S0074-7742(05)66008-4. [DOI] [PubMed] [Google Scholar]
- 19. Quaresima V. and Ferrari M., “Functional Near‐Infrared Spectroscopy (fNIRS) for Assessing Cerebral Cortex Function During Human Behavior in Natural/Social Situations: A Concise Review,” Organizational Research Methods 22, no. 1 (2016): 46–68, 10.1177/1094428116658959. [DOI] [Google Scholar]
- 20. Herold F., Orlowski K., Börmel S., and Müller N. G., “Cortical Activation During Balancing on a Balance Board,” Human Movement Science 51 (2017): 51–58. [DOI] [PubMed] [Google Scholar]
- 21. Pinti P., Aichelburg C., Gilbert S., et al., “A Review on the Use of Wearable Functional Near‐Infrared Spectroscopy in Naturalistic Environments,” Japanese Psychological Research 60, no. 4 (2018): 347–373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Strangman G., Culver J. P., Thompson J. H., and Boas D. A., “A Quantitative Comparison of Simultaneous BOLD fMRI and NIRS Recordings During Functional Brain Activation,” NeuroImage 17, no. 2 (2002): 719–731. [PubMed] [Google Scholar]
- 23. Pinti P., Tachtsidis I., Hamilton A., et al., “The Present and Future Use of Functional Near‐Infrared Spectroscopy (fNIRS) for Cognitive Neuroscience,” Annals of the New York Academy of Sciences 1464, no. 1 (2020): 5–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Menant J. C., Maidan I., Alcock L., et al., “A Consensus Guide to Using Functional Near‐Infrared Spectroscopy in Posture and Gait Research,” Gait & Posture 82 (2020): 254–265. [DOI] [PubMed] [Google Scholar]
- 25. Penfield W. and Boldrey E., “Somatic Motor and Sensory Representation in the Cerebral Cortex of Man as Studied by Electrical Stimulation,” Brain 60, no. 4 (1937): 389–443, 10.1093/brain/60.4.389. [DOI] [Google Scholar]
- 26. MacLennan R. J., Hernandez‐Sarabia J. A., Reese S. M., et al., “fNIRS Is Capable of Distinguishing Laterality of Lower Body Contractions,” Experimental Brain Research 242, no. 5 (2024): 1115–1126, 10.1007/s00221-024-06798-8. [DOI] [PubMed] [Google Scholar]
- 27. Rorden C. and Brett M., “Stereotaxic Display of Brain Lesions,” Behavioural Neurology 12, no. 4 (2000): 191–200. [DOI] [PubMed] [Google Scholar]
- 28. Zimeo Morais G. A., Balardin J. B., and Sato J. R., “fNIRS Optodes' Location Decider (fOLD): A Toolbox for Probe Arrangement Guided by Brain Regions‐of‐Interest,” Scientific Reports 8, no. 1 (2018): 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Herold F., Wiegel P., Scholkmann F., and Müller N. G., “Applications of Functional Near‐Infrared Spectroscopy (fNIRS) Neuroimaging in Exercise–Cognition Science: A Systematic, Methodology‐Focused Review,” Journal of Clinical Medicine 7, no. 12 (2018): 466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Huppert T. J., Diamond S. G., Franceschini M. A., and Boas D. A., “HomER: A Review of Time‐Series Analysis Methods for Near‐Infrared Spectroscopy of the Brain,” Applied Optics 48, no. 10 (2009): D280–D298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Molavi B. and Dumont G. A., “Wavelet‐Based Motion Artifact Removal for Functional Near‐Infrared Spectroscopy,” Physiological Measurement 33, no. 2 (2012): 259–270. [DOI] [PubMed] [Google Scholar]
- 32. Collins D. L., Zijdenbos A. P., Kollokian V., et al., “Design and Construction of a Realistic Digital Brain Phantom,” IEEE Transactions on Medical Imaging 17, no. 3 (1998): 463–468. [DOI] [PubMed] [Google Scholar]
- 33. Aasted C. M., Yücel M. A., Cooper R. J., et al., “Anatomical Guidance for Functional Near‐Infrared Spectroscopy: AtlasViewer Tutorial,” Neurophotonics 2, no. 2 (2015): 020801, 10.1117/1.NPh.2.2.020801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Crawford J. R. and Howell D. C., “Comparing an Individual's Test Score Against Norms Derived From Small Samples,” Clinical Neuropsychologist 12, no. 4 (1998): 482–486. [Google Scholar]
- 35. Crawford J. R. and Garthwaite P. H., “Investigation of the Single Case in Neuropsychology: Confidence Limits on the Abnormality of Test Scores and Test Score Differences,” Neuropsychologia 40, no. 8 (2002): 1196–1208, 10.1016/S0028-3932(01)00224-X. [DOI] [PubMed] [Google Scholar]
- 36. Xu J., Cai G., Li W., et al., “Robot‐Assisted Treatment Contributes to Regaining Upper Limb Motility in Stroke Patients: A Randomized‐Controlled Trial Based on Functional Near Infrared Spectroscopy,” Research Square (2024), 10.21203/rs.3.rs-4232249/v1. [DOI] [Google Scholar]
- 37. Bai Z., Fong K. N., Zhang J., and Hu Z., “Cortical Mapping of Mirror Visual Feedback Training for Unilateral Upper Extremity: A Functional Near‐Infrared Spectroscopy Study,” Brain and Behavior 10, no. 1 (2020): e01489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Simoes E. L., Bramati I., Rodrigues E., et al., “Functional Expansion of Sensorimotor Representation and Structural Reorganization of Callosal Connections in Lower Limb Amputees,” Journal of Neuroscience 32, no. 9 (2012): 3211–3220. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.