In this feasibility study at a low-resource point-of-care setting, we used our wearable gait quantification shoe [38] to quantify gait performance changes due to a single session of ctDCS in chronic stroke survivors. The experimental setup comprised of (i) the gait quantification shoe [38], and (ii) the wireless ctDCS cap with STARSTIM 8 stimulator (Neuroelectrics, Spain), as shown in Figure 1. The gait quantification shoe characterized the gait parameters during overground walking in terms of Step Length [33], Walk Ratio [34], Gait Stability Ratio [35], Symmetry Index [36], and other relevant gait parameters [37], including Stride Time, Step Time, %Stance Time, %Swing Time, %Single Support Time, Cadence. Overground gait and balance evaluation were also performed based on the Ten-Meter walk test (TMWT) [39], Timed-Up-and-Go (TUG) [16], and Berg Balance Score (BBS) [40] before and after the ctDCS intervention, as shown in Figure 2. A single session of ctDCS intervention was investigated based on its acute effects on the gait and balance measures from chronic (> 6 months’ post-stroke) hemiplegic patients.
i. Study Participants
The hemiplegic stroke survivors, who (i) were aged between 18 and 90 years, (ii) could walk independently for at least 10 meters, (iii) could provide informed and written consent, and (iv) could understand instruction from the experimenter were contacted. Twelve post-stroke male subjects (P1-P12, Mean (SD) = 46(±13) years) were selected for this study (see Table 1) from fourteen volunteers recruited by convenience sampling from collaborating hospitals. We selected chronic (>6 months' post-stroke) stroke survivors with cerebral lesions but with an intact cerebellum (based on computerized tomography scan) so that the focal ctDCS electric field effects can be delivered to the cerebrum via the intact cerebellum [41]. Stroke survivors who underwent any recent surgery or were in the acute phase of stroke were excluded from the study. Written informed consent was obtained from each subject, and the multi-center research protocol for this study was approved by the All India Institute of Medical Sciences, New Delhi, India Institutional Review Board (IEC-129/07.04.2017), and Indian Institute of Technology Gandhinagar, India Institutional Review Board (IEC/2019-20/4/UL/046).
ii. Gait Quantification Shoe
In this study, we aimed to quantify gait-related indices by recording gait events using a pair of instrumented shoe [38]. Figure 1 shows the wearable device, namely the gait quantification shoes (GaitShoe henceforth) [38], that was used in this study to record the gait events. The GaitShoe consisted of insoles instrumented with force-sensitive resistors (FSRs) that were placed below the greater toe, lateral heel, and medial heel positions of each shoe to detect the gait events, e.g., heel-strike, toe-off, etc. These gait events were used to compute different gait-related indices, e.g., Step Length [33], Walk Ratio [34], Gait Stability Ratio [35], Symmetry Index [36], and other relevant gait parameters [37] including Stride Time, Step Time, %Stance Time, %Swing Time, %Single Support Time, Cadence. The GaitShoe transmitted the data wirelessly to a data logger computer for subsequent offline analysis.
a. Computation of Step Length
Step Length is the distance between two successive contralateral heel-strikes during gait. We wanted to study the effects of ctDCS on Step Length since this is an essential indicator of the functional gait ability of hemiplegic post-stroke patients [33]. Here, we computed the average Step Length using the average Step Time (recorded by the GaitShoe) and the average Walking Speed. The Step Time was measured by the GaitShoe from the time interval between two successive heel-strike events of the contralateral legs. Walking Speed (during the overground walk) was computed from the time taken to walk through a pre-defined distance. Subsequently, Step Length was calculated using Eq. (1). Finally, the Normalized Step Length was computed using the individualized height information [38] (Eq. (2)).
Step Length= Step Time * Walking Speed ……………………………………………………………………..(1)
Normalized Step Length = Step Length/Height………………………………………………………...……….(2)
Normalized Step Length was computed separately for the affected and the unaffected sides of the hemiplegics.
b. Computation of Gait Stability Ratio
The Gait Stability Ratio (GSR) depends on Cadence (steps/sec) and Walking Speed (m/sec) [38]. The GSR is a good indicator of balance deficits in older adults [35]. The GSR takes into account the changes in Walking Speed that can influence the Step Length. Here, a decrease in GSR indicates increased double support time during walking. The GSR was computed using Eq. (3).
Gait Stability Ratio= Cadence/Walking Speed …………………………………………………………….(3)
c. Computation of Walk Ratio
The Walk Ratio (WR) [34] can describe a relation between Step Length and Cadence during walking. Importantly, WR is invariant during different speeds, uneven surface conditions but is affected by dual task-condition [42]. The Walk Ratio was computed using Eq. (4).
Walk Ratio= Step Length/Cadence ……………………………………………………………………………(4)
Walk Ratio was computed separately for the affected and the unaffected sides of the hemiplegics.
d. Computation of Symmetry Index
The Symmetry Index (SI) is a measure of the extent to which one makes symmetrical use of both the legs during walking [36] – the smaller the value of SI, the better is the gait symmetry. One of the distinctive characteristics of post-stroke gait is the impaired gait symmetry, particularly in hemiplegic patients [43]. Here, we computed the SI using the %stance phase (of a gait cycle) measured using the GaitShoe while considering the %stance for each of the left (XL) and right legs (XR). The SI was calculated using Eq. (5).
SI= ((XL-XR) / 0.5 * (XL+XR) ) *100 …………………………………………………………………..………(5)
e. Computation of Stride Time and Step Time
The Stride Time, defined as the time interval between two successive heel-strike events of the ipsilateral legs [44], was computed using the GaitShoe. The Stride Time and Step Time were computed separately for the affected and the unaffected sides of the hemiplegics.
f. Computation of %Stance Time, %Swing Time
The gait cycle can be considered broadly comprising of two main phases, namely Swing Phase and Stance Phase. A healthy gait cycle (GC) can be characterized by ~60% GC in the Stance Phase and ~40% GC in Swing Phase [44]. The Stance Phase can be defined as the phase in which the foot stays in contact with the base of support (e.g., the floor) during the gait cycle, while the Swing Phase can be defined as the phase in which the foot is not in contact with the base of support [44]. In this study, the Stance Time was computed as the time interval between the successive heel-strike and toe-off events of the ipsilateral leg based on the foot contact with the floor. The %Stance time was computed by evaluating the Stance Time as a percentage of the gait cycle time. Similarly, the Swing Time was computed as the time interval between the successive toe-off and heel-strike events of the ipsilateral leg when the foot was not in contact with the floor. The %Swing Time was computed by evaluating the Swing Time as a percentage of the gait cycle time. The %Stance Time and %Swing Time were computed separately for the affected and the unaffected sides of the hemiplegics.
g. Computation of %Single Support Time
The Single Support Time (SST) is the duration of a gait cycle for which only one foot stays in contact with the base of support (such as the floor) while supporting the entire weight of the body on that leg under dynamic stability during a gait cycle which is important for fast walking [45]. The SST can be computed as the swing time of the contralateral leg. Gait cycle duration was measured using the time interval between the two consecutive heel-strike events of the same leg. Subsequently, the %SST was calculated for each leg using Eq. (6). The %SST was computed separately for the affected and the unaffected sides of the hemiplegics.
h. Computation of Cadence
Cadence can be defined as the number of steps walked per minute [37]. The Cadence was computed as the number of heel-strike events registered per minute, considering both the affected and the unaffected legs.
iii. Optimization of the Electrode Montage (age-specific computational modeling of ctDCS)
We used age-specific MRI templates that were obtained online at https://jerlab.sc.edu/projects/neurodevelopmental-mri-database/ with the permission of Dr. John Richards. The data comprised of average T1-weighted MRI with the segmentation priors for the gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). For this, we chose the age group that matched the age of our subjects for this study. A Realistic volumetric Approach to Simulate Transcranial Electric Stimulation (ROAST) [46] was used to create a tetrahedral volume mesh of the head. ROAST used SPM12 ("SPM - Statistical Parametric Mapping") to segment the head and brain. After segmentation, five tissues were identified for the tetrahedral volume mesh, namely, Scalp, Skull, Cerebrospinal Fluid (CSF), Gray Matter (GM), and White Matter (WM). These different brain tissues for the volume mesh were modeled as different volume conductors for Finite Element Analysis (FEA) in the ROAST. Here, isotropic conductivity used for the different brain tissues [46] were (in S/m): Scalp=0.465; Skull=0.01; CSF=1.654; GM=0.276; WM=0.126. For further details on the head modeling, please refer to our prior works [22], [23].
The Electric Field (EF) distribution was found for two different ctDCS montages based on the subject's age-specific head model that was created from MRI templates ( https://jerlab.sc.edu/projects/neurodevelopmental-mri-database/ ). The boundary condition was set as 2mA injection current (Neumann boundary condition) with the following electrode configurations from our prior work where we performed optimization of the electrode montage [23]:
a) Optimized configuration for dentate nuclei stimulation [23]: 3.14cm2 disc anode was PO10h (10/5 EEG system), and 3.14cm2disc cathode was placed at PO9h (10/5 EEG system) for ctDCS with 2mA direct current.
b) Optimized configuration for leg lobules VII-IX stimulation [23]: 3.14cm2 disc anode was Exx8 (electrodes defined by ROAST using "unambiguously illustrated (UI) 10/5 system" [47]), and 3.14cm2 disc cathode was placed at Exx7 (defined by ROAST) for ctDCS with 2mA direct current.
In all the simulations, the voxel size was considered as 1mm3. The contra-lesional anode and ipsilesional cathode injected the specified amount of current in the volume conductor, i.e., the head model. Finite Element Analysis (FEA) was conducted on each age-specific head model to compute the ctDCS induced EF in the brain tissues. The electric field was computed at all the voxels (voxel size 1mm3) of the cerebellar lobules that were normalized for flatmap using a spatially unbiased atlas for the cerebellum and brainstem (SUIT) [48]. Here, the cerebellar lobular electric field distribution was found as flatmap using SUIT [48] and T1-weighted images that were fitted to the SUIT template of the human cerebellum in SPM12 ("SPM - Statistical Parametric Mapping": https://www.fil.ion.ucl.ac.uk/spm/software/spm12/). The cerebellar mask was visually checked in MRIcron, and the non-linear deformation was then applied to each EF image. The volume of the cerebellar lobules, defined by the SUIT atlas [48], was used for the extraction of the lobular EF distribution. Also, we customized SUIT codes to extract the EF distribution at the left and the right dentate nucleus.
iv. Experimental Setup and Data Analysis
Figure 1 shows the experimental setup for the clinical study in a low-resource point-of-care setting with a subject walking on the 10-meter walkway for overground gait evaluation. The study required a commitment of about 30 minutes from each participant.
a. Cerebellar tDCS intervention
Based on our prior work [23], 15 minutes of 2mA bilateral ctDCS was delivered in a repeated measure single-blind crossover design using two bipolar montages with a circular (1cm radius) contra-lesional anode. The two bipolar montages were allocated in random order with 2-3 days’ washout period between the ctDCS sessions, and the subjects were blinded to the montage by keeping all the four stimulation electrodes (two anodes and two cathodes for two ctDCS montages) always embedded in their cap. The electrode locations in the cap were based on the ROAST toolbox [46], and "unambiguously illustrated (UI) 10/5 system" [47]; 1. PO9h – PO10h, and 2. Exx7 – Exx8. The experimental setup is shown in Figure 1 (see the right bottom inset with the neoprene cap), and the experimental protocol is shown in Figure 2, where overground quantitative gait, as well as clinical gait (TMWT [39]) and balance evaluations (TUG, BBS), were performed before and after the ctDCS intervention to compute a percent normalized change measure,
b. Experimental setup for overground gait analysis
The experimental setup for the overground gait analysis consisted of (i) 10 m long straight overground pathway (for TMWT [39]) marked with start and end lines, (ii) data-logger computer, and (iii) a pair of GaitShoes. We investigated the effects of ctDCS on gait characteristics during the 10 m overground walk – see Figure 1. Once the participant arrived at the study hall, they were asked to sit and relax for about 5 minutes. Then, the experimenter explained to the participant what he was expected to do in the study as well as the risks. After informed consent, the baseline clinical measures were recorded. Then, the experimenter helped the participant to wear the GaitShoe [38]. Subsequently, the experimenter prepared the participant for ctDCS by placing the neoprene cap combined with a battery-driven wireless stimulator, STARSTIM8 (Neuroelectrics, Spain), and the gel-based electrodes. The participants were informed that they could discontinue the study in case of any discomfort.
Once the participant was ready to start the study, they were asked to walk on a 10m long straight path (overground) marked with a start and stop lines at their self-selected comfortable speed, and the participant's overground Walking Speed (SpeedOG) was computed. After this, the participant was asked to sit and relax on a chair for about 5 mins. Subsequently, ctDCS was administered using one of the two ctDCS montages for 15 minutes at the rest condition with a dosage of 2 mA [23]. Following this, the participant repeated the 10m overground walk, followed by an assessment of the clinical gait and balance measures (TMWT, TUG, and BBS). The gait performance of the post-stroke participants was also quantified using GaitShoe in terms of the gait-related indices, as described earlier. Therefore, the post-stroke participants performed two trials of the overground walk, pre, and post ctDCS intervention, at their self-selected walking speed while wearing the GaitShoe, as illustrated in Figure 2. We also evaluated the acceptability of the ctDCS intervention in post-stroke subjects based on a questionnaire (see Supplementary Materials) where we collected subjective feedback from the post-stroke participants prior to (PretDCS), during (ActivetDCS), and post (PosttDCS) application of ctDCS.
c. Statistical analysis and the partial least squares regression
A two-sided Wilcoxon rank-sum test was performed at the 5% significance level on the percent normalized change measures
the null hypothesis that the two ctDCS montages led to the same percent normalized change in the quantitative gait parameters from the same continuous distributions with equal medians. Multivariate regression analysis was conducted to relate the changes in the balance and gait measures to the lobular electric field distribution due to ctDCS montages. Here, multicollinearity can occur when independent variables (predictors) are correlated. In our prior work [23], we have presented principal component regression analysis for multivariate linear regression of the lobular electric field distribution as the predictor with the behavioral outcomes as the response variables. The goal is to extract the relation between electric field distribution and the behavioral effects of ctDCS where Partial Least Squares (PLS) can be a promising multivariate statistical technique that can combine the information about the variances of both the predictors and the responses while also considering the correlations among them [49]. In this study, we applied a PLS regression (PLSR) approach to analyze the associations between the lobular electric field distribution as the predictor with the gait outcome measures as the response variables. Although statistical inference is the strength of the PLSR approach using computational cross-validation methods (e.g., jackknife, bootstrap) [49]; however, we will apply PLS as a correlation technique in this study. The matrix of correlations between the lobular electric field distribution as the predictor with the gait outcome measures as the response variables is subjected to the singular value decomposition that results in the singular vectors called saliences. The lobular electric field distribution as the predictor with the gait outcome measures as the response variables can be projected onto their respective saliences, which creates latent variables that are linear combinations of the original variables. Here, PLS searches for latent variables that express the largest amount of information common to both the lobular electric field distribution as the predictor and the gait outcome measures as the response variables. This is a fixed-effect model where the results can only be interpreted with respect to the current data sets from this study. In this study, PLS analysis was performed on the percent normalized change measures,
of gait indices from the Gaitshoe as the response variable, where the lobular electric field distribution for both the montages across all the subjects (found after centering the data and then singular value decomposition) was the predictor.