Application of Transcranial Magnetic Stimulation with Electroencephalography in the Evaluation of Brain Function Changes after Stroke

Objective Based on transcranial magnetic stimulation (TMS) with electroencephalography technology, this study analyzed the rehabilitation mechanism of patients' motor function reconstruction and nerve remodeling after stroke. It revealed the function of the cerebral cortex network at a deeper level and established a set of prognostic marker evaluation indicators for the reconstruction of motor function after stroke. Methods Twenty-one patients treated at the Beijing Rehabilitation Hospital of Capital Medical University because of ischemic stroke in the territory supplied by the middle cerebral artery were selected as the experimental group. Neurophysiological evaluation, motor function evaluation, and clinical evaluation were performed 30 and 180 d after the onset of ischemic stroke. In the control group, neurophysiological evaluation was also performed as a reference index to evaluate the changes in cortical patterns after stroke. Results The brain topographic map showed the changes in energy or power spectral density (PSD) at 1,000 ms after stimulation as compared with before stimulation, but no difference was detected in these patients. The time-frequency analysis showed that when the left primary motor cortex (M1) area was stimulated using TMS, the PSD values of the left and right M1 and posterior occipital cortex areas produced an 8–40 Hz wave band in patients S1–S11. There was no significant energy change in patients S12–S16. Conclusions For patients with different injury types, degrees of injury, and different onset periods, individualized intervention methods should be adopted. The evaluation methods should be as diverse as possible, and the rehabilitation effects of patients should be assessed from multiple perspectives to avoid the limitations of single factors. Possible mechanism: After brain injury, the nervous system can change its structure and function through different ways and maintain it for a certain period of time. This plasticity change will change with the course of the disease.


Introduction
World Health Organization (WHO) survey results demonstrate that the incidence of stroke in China ranks frst in the world, twice higher than the United States. It is worth noting that stroke is no longer the "patient" of the elderly; more and more young people have joined them. In the two major categories of stroke, the incidence of ischemic stroke in China is 1.36 times the world average, and the incidence and mortality of hemorrhagic stroke are 2 times the world average. A population-based study showed that the prevalence of stroke in China continued to rise from 2013 to 2019 [1].
Tere are 2.5-3 million new stroke cases in China every year. With the improvement of diagnosis and treatment level, and rescue success rate in recent years, the survival rate of patients with stroke has increased. Although early treatment and intensive rehabilitation can signifcantly improve neurological impairment, more than 80% of stroke survivors and 40% of chronic patients [2] sufer from neurological dysfunction. Motor dysfunction is the most common symptom after stroke, afecting the independent living ability of patients [3] and increasing treatment costs and mortality.
Terefore, active and efective rehabilitation treatment to reduce disability is particularly important. Te rehabilitation treatment of motor function after stroke has gradually developed from neural promotion technology to physical factor therapy. In recent years, motor imagery therapy [4], constraint-induced motion therapy [5], functional electrical stimulation [6], enhanced occupational autonomic activity [7], and repetitive transcranial magnetic stimulation (rTMS) [8] have become new methods of rehabilitation treatment. Based on the nonrenewable nature of nerve cells, the specifc mechanism of motor function recovery after stroke is still unclear [9]. In general, the rehabilitation efect cannot currently meet expectations, and there is no consensus on the best rehabilitation method. What is the patient's potential rehabilitation capacity? Based on her/his clinical characteristics, what is the best rehabilitation strategy for this person? Which patients will beneft most from specifc interventions? In view of these problems, it is urgent to identify and verify accurate neurophysiological markers related to persistent disability and successful stroke rehabilitation to improve the ability to predict long-term outcomes after stroke.
Neurophysiological changes of nerve cells occur spontaneously within days to months after brain injury. Functional magnetic resonance imaging (MRI), positron emission tomography (PET), electroencephalogram (EEG), and TMS are widely used to study these changes, which can analyze the correlation between local blood fow, metabolism, and neuronal discharge to detect cortical excitability [10]. Using diferent neuroimaging methods to study the special role of brain regions in functional recovery, understand the potential mechanism of spontaneous recovery after stroke [11,12], and detect the target and key period of treatment intervention [13] has become the next research trend.
One of the four major technologies in brain science research is TMS. It uses electromagnetic induction to generate eddy current painlessly and generate action potentials in cortical neuron tissue [14,15]. It is often used in the rehabilitation treatment of stroke, including limb, speech, and cognitive function rehabilitation [16][17][18]. In general, TMS acts on the primary motor cortex (M1) and motor evoked potential (MEP) and is recorded by surface electromyography of the target muscle. Monopulse TMS (TEP) has been used to study the diference of corticospinal tract integrity and cortical excitability in patients with stroke [19]. Steiner and Ward found that the presence or absence of MEP in paralyzed limbs can predict the response to motor skill training in patients with chronic stroke [20] and the functional outcome of acute stroke [21]. TMS-associated measures of cortical function and plasticity (such as the short latency aferent inhibition, the short-interval intracortical inhibition, and the cortical silent period) might add useful information in most cases of secondary dementia, especially in combination with suggestive clinical features and other diagnostic tests [22]. In the TMS assessment of stroke neurophysiology, dependence on M1 and the integrity of the peripheral corticospinal tract pathway limits the characterization of the underlying neural mechanisms of motor dysfunction, as many patients may not be able to induce measurable MEP, especially in the early stage of stroke [23]. Although TMS can be used to identify patients with signifcant recovery potential, especially when used in combination with MRI and standard clinical examinations [21], the lack of MEP records may limit the predictive role of TMS in functional recovery.
With the deepening of research, TMS can be combined with neuroimaging and neuroelectrophysiological techniques to monitor its stimulating efects on brain regions, such as functional MRI, PET, EEG, and TMS [24,25]. Tese methods have their own advantages and disadvantages. Te combination of TMS and EEG provides the possibility for noninvasive detection of brain excitability, connectivity, and transient state. In general, EEG is a low-cost, noninvasive functional neuroimaging technology that uses electrodes located on the scalp to quantify nerve potentials in the brain. Action potentials come from the mixed efect of inhibitory and excitatory postsynaptic potentials in neurons, which can be captured at the scalp and analyzed to help diagnose, prognose, and monitor treatment responses to various neurological disorders. Since functional changes are earlier than morphological and structural changes, EEG signals can be more intuitively and sensitively analyzed after digital processing to help diagnose, evaluate prognosis, and monitor the treatment response of neurological dysfunction [26]. Te excellent time resolution (∼1 ms) combined with clinically feasible acquisition procedures makes EEG an attractive technique for assessing changes in brain activity after stroke.
Conversely, TMS induces the initial activation of the target region, followed by the triggering activity of signals transmitted by axons and synapses, resulting in late efects [27]. Studying the EEG signals of TMS (TMS-EEG) can provide a direct relationship between brain function and behavior and simultaneously evaluate a variety of neurophysiological processes, including cortical reactivity, local cortical excitation or inhibition, oscillatory activity, global brain connectivity [28], and neural plasticity. Unlike MEP, TMS-EEG can be used to evaluate the cortical response of almost any target brain to stimulation. It is only limited by the accessibility of the cortical target structure to TMS, is independent of spinal excitability, and is highly repeatable. Terefore, TMS-EEG provides the possibility of studying cortical interaction. Te frst EEG signal induced by TMS refects the excitability of stimulating the cortex, i.e., the functional state, while the response of other cortical regions induced by stimulation can be characterized by the EEG connectivity index [29]. Additionally, TMS-EEG can bypass the subcortical structure and directly evaluate the cortical excitability and connectivity. Compared with the traditional intervention, TMS-EEG has advantages for patients with severe subcortical damage. In evaluating the neurophysiological mechanisms related to brain network reorganization after stroke, TMS-EEG has good research and application prospects [21,30] and can be used as a neurophysiological marker for stroke recovery [31,32]. It provides a unique insight into efective connectivity and the description of causal interactions between regions, including how well the activation of one region explains the activation of another [33]. Notably, this approach provides a causal model on the origins of activation in neural activity patterns and might defne the functional strengths between regions. However, in clinical practice, there are relatively few systematic personalized TMS-EEG studies for patients with diferent types and degrees of injury, and there is no standard treatment system.
Te main purpose of this study was to explore the rehabilitation mechanism of motor function reconstruction and neural remodeling in patients with stroke based on TMS-EEG technology to further reveal the function of the cerebral cortex network. Additional objectives were to explore the correlation between EEG characteristic parameters (event-related potential, brain connectivity, etc.) and clinical evaluation indexes of motor function, to explore the correlation between TMS-induced changes in cortical excitability and connectivity and motor function changes, and to establish a set of prognostic markers for motor function reconstruction after stroke based on TMS-EEG technology. To understand the changes in cortical patterns after stroke events, the TEP and frequency oscillations of spontaneous cortical activity were compared between patients and controls.

Study Design.
From June 2019 to February 2020, 21 patients with ischemic stroke in Beijing Rehabilitation Hospital, Capital Medical University were selected as the experimental group, and seven healthy people were selected as the control group. Among them, fve patients were unable to continue to cooperate due to lung infection and other reasons. Te experimental group ended up with 16 patients. None of the 21 patients died at the time of discharge. Seven healthy people were compared as the control group.
Te inclusion criteria were as follows: (1) all patients met the diagnostic key points of various cerebrovascular diseases in the 2019 Guidelines for Diagnosis and Treatment of Acute Ischemic Stroke of the Cerebrovascular Disease Group of the Neurological Branch of the Chinese Medical Association [34]; (2) all patients were confrmed by computed tomography or MRI imaging examination; (3) the patients had the frst onset of stroke with stable vital signs and no obvious progress in neurological signs; (4) the cerebral infarction sites were the middle cerebral artery blood supply territories, including the thalamus, corpus callosum, and internal capsule; and (5) all patients signed an informed consent form. Tis study was conducted with approval from the Ethics Committee of Beijing Rehabilitation Hospital, Capital Medical University. (No. 2020bkky-022).
Te exclusion criteria were as follows: (1) failure of vital organs or insufciency of vital organs; (2) diseases with bleeding tendency; (3) unstable blood pressure control; (4) pacemaker and implantable defbrillator with other implantable devices in the body; (5) history of epilepsy; (6) any contraindications to TMS; or (7) failure to give informed consent. Seven healthy volunteers were selected as the control group, comprising two females and fve males, aged from 42 to 54 years old (T1-T7).

Neurophysiological Assessment.
Te motor function assessment and clinical evaluation were performed at Days 30 and 180 after the onset of ischemic stroke.

Electroencephalogram and Transcranial Magnetic Stimulation with Electroencephalography
Recording. Te scalp EEG was recorded with TMS-compatible EEG equipment (Brain Amp 32 MR Plus, Brain Products GmbH, Munich, Germany). Using TMS-compatible Ag/AgCl electrodes mounted on elastic caps, the EEG was continuously obtained using 32 scalp positions located by the 10-20 international system. Additional electrodes were used for grounding and reference. Te grounding electrode was located at the AFZ site, while the efective reference was located at the tip of the nose. Te sampling rate of the EEG and electrooculogram (EOG) signals were 5 kHz and the band-pass fltering range was 0.1-1,000 Hz. Te skin/electrode impedance was kept below 5kΩ. Horizontal and vertical eye movements were detected by recording EOG for ofine signal processing to monitor the online behavior of participants and reject the ofine trial through eye artifact.

Transcranial Magnetic Stimulation.
Te TEP was carried out by Magstim Super Rapid Stimulator (Magstim Ltd., Withland, Wales, UK). Te stimulator was connected to a supercharger module and a standard digital 8 shape coil with an external winding diameter of 70 mm (Magstim Ltd.), which produced a maximum output of 2.2 T. To defne resting motor threshold (RMT) [35], the coil was tangentially placed to the scalp, M1 on the left and right sides, with the handle pointing back and side 45°with the mid-sagittal axis of the participant's head, so that the current direction of the second stage was anteromedial. Te stimulation started with the overthreshold intensity. Te best stimulation site between the right and left frst dorsal interosseous that elicits MEPs is called the motor hotspot, which is identifed by placing the coil on the central sulcus and moving it to the scalp in steps of 0.5 cm on the left and right M1. Te RMT was assessed as the lowest intensity required to evoke an MEP with a peak-to-peak amplitude ≥50 μV in at least 5 out of 10 consecutive trials in the relaxed position. If MEP was not induced in the afected hemisphere (AH) at the maximum stimulator output (1% resolution), the intensity was set to 90% of the RMT of the unafected hemisphere (UH). To target the left and right posterior parietal cortex (PPC), the coil was positioned on the tail of the inner parietal groove, roughly corresponding to P3 and P4, and oriented 15°from the midline, thus inducing current from the back to the front direction. For PPC, both hemispheres were symmetrical stimuli. High reliability was ensured during each recording. Te coil location and the location on the hot spot were monitored online by using the SofTaxic neuronavigation system (EMS, Bologna, Italy) and a Polaris Vicra infrared International Journal of Clinical Practice 3 thermal imager. To avoid the infuence of auditory evoked potential brought by the TMS pulse sound on EEG signals, the subjects wore earplugs that played noise.

Exercise and Clinical
Assessment. Te Fugl-Meyer Assessment Scale (FMA) [36] was used to measure the recovery of limb motor function. Te Barthel Index [37] comprehensively evaluated daily life, quality of life, and clinical rehabilitation ability.
2.6. Electroencephalogram Analysis. Two sets of EEG data obtained during the quiescent state were analyzed ofine by MATLAB (MathWorks Inc., USA). Excessive drift, eye movement, blinking, and muscle activity were treated or excluded by independent component analysis. Te energy or power spectral density (PSD) of diferent frequency bands was estimated by fast Fourier transform (10% Hanning window; frequency resolution: 1 Hz), which was divided into four frequency bands: delta (2-4 Hz), theta (4-8 Hz), alpha (8)(9)(10)(11)(12), and beta (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30). Te average PSD of each group of EEG data were calculated as the superposition average within the frequency band, and the total power of each frequency band of all channels was calculated to evaluate the EEG diference between the two groups. Te analysis to evaluate the cortical response to TMS in the time domain showed that the TEP impulse response induced by TMS was recorded from the frst 100 ms to the last 500 ms at each evaluation time point and stimulation area. All periods were recorded for a period of 100 ms before baseline correction to TMS pulses.
Te analysis to assess the cortical response to TMS in the time-frequency domain showed that, from diferent cortical regions of AH and UH, the oscillatory response to TMS pulses were detected from 1 s to 1 s later.
In the brain connectivity calculation, the directed transfer function was used to construct the upper triangular and lower triangular matrices [35] to represent the bidirectional connectivity between the aferent and eferent directions, and the diagonal elements were not included. Te paired sample t-test was performed for the upper and lower triangular matrices of the state before and after the intervention, and the lead pairs with signifcant improvement were taken out to draw the brain connectivity map.
Te research process is shown in Figure 1.

Statistical Analysis.
Te paired t-test was performed on the measurement data. Te SPSS ™ Statistics v23.0 software was used for statistical analysis. A value of P < 0.05 indicated a statistical signifcance.

Results
Among the 21 patients, 5 could not continue to cooperate due to pulmonary infection or other reasons. Tese left 16 patients in the experimental group (S1-S16) ( Table 1). Tere are 13 males and 3 females in the experimental group, with an average age of 64. Tere were 3 males and 4 females in the healthy control group, with an average age of 51 years.

Te Brain Electrical Signals and Electrical Energy.
After TMS, the brain electrical signals of healthy patients and patients with stroke changed, including brain electrical energy. After stimulation, the connectivity of the brain was signifcantly increased compared with before stimulation, which further proves that TMS can be used as a poststroke treatment (Figure 2). Rehabilitation treatment for stroke patients can improve limb dysfunction, improve the quality of life, and enhance the connectivity of the EEG.

Time-Frequency Analysis
Chart. Appendix Figure 2 shows that the PSD values of 16 patients were calculated when the TMS of the left M1 area and the PSD of the left and right M1 area and the left and right posterior occipital cortex area under stimulation were changed, respectively. When the left M1 area was stimulated, the above four areas of S1-S11 produced an 8-40 Hz wave band as compared with the left and right M1 in patients S1-S7 and S9-S11. When the left M1 area was stimulated, the left M1 area energy increased higher than the right M1 area. Both sides were approximately equal for S8. Tere was no signifcant energy change in S12-S16.

Detailed Analysis for
Patients. S1 (damage to the left hemisphere): Te topographic map showed that after stimulating the left M1 area, the energy of the healthy right M1 dropped signifcantly. Te time-frequency analysis showed that, when stimulating the left M1 area using TMS, the energy increase of the left M1 area was higher than that of the right M1 area, and the efect was higher than that of contralateral stimulation. S2 (damage to the left hemisphere): Te topographic map showed that after stimulating the left M1 area, the left hemisphere part of the energy increased, which had an exciting efect on the afected side. According to the timefrequency analysis diagram, when stimulating the left M1 area using TMS, the energy increase and the excitation efect in the left M1 area were higher than in the right M1 area.
S3 (damage to the right hemisphere): Te topographic map showed that after stimulating the right M1 area, only the frontal parietal lobe had a small range of energy increase. Te time-frequency analysis diagram indicated that when stimulating the right M1 area using TMS, the energy increase and the excitation efect in the right M1 area were higher than in the healthy left M1 area.
S4 (damage to the left hemisphere): Te topographic map showed that after stimulating the left and right M1 areas, the energy of the occipital lobe area decreased, indicating an inhibitory efect but no obvious activation in the whole brain. Te time-frequency analysis indicated when stimulating the left M1 area using TMS, the energy increase and the excitation efect in the left M1 area were higher than in the right M1 area.
S5 (damage to the left hemisphere): Te topographic map showed that after stimulating the left M1 area, the frontal parietal lobe energy increased, indicating activation. Te time-frequency analysis indicated that when stimulating the left M1 area using TMS, the left M1 area energy increased more and the energy excitation efect was higher than in the right M1 area.
S6 (damage to the right hemisphere): Te topographic map showed that after stimulating the right M1 area, the right motor area showed a slight energy decrease in a small area, and the frontal and parietal area showed an increase in energy, but the scope of action was small. Te timefrequency analysis indicated that when stimulating the right M1 area using TMS, the energy increase in the right M1 area was higher than in the left M1 area and the stimulation efect on the right side was higher than on the contralateral side.
S7 (damage to the left hemisphere): Te topographic map showed that after stimulating the left M1 area, it showed an increase in energy, while the temporal lobe area showed a decrease in energy, indicating that TMS activated the patient's afected side's motor area. Te time-frequency analysis indicated that when stimulating the left M1 area using TMS, the energy increase of the left M1 area was higher than in the right M1 area, meaning that the left-side energy excitation efect was high on the contralateral side of the excitation efect.
S8 (damage to the right hemisphere): Te topographic map showed that after stimulating the right M1 area, the whole brain area showed a trend of energy reduction, indicating that there was an inhibition efect on the brain area. Te time-frequency analysis indicated that when stimulating the right M1 area using TMS, the energy of the right M1 area increased more than that of the left M1 area, and the excitation efect on the right side was higher than that on the healthy side.
S9 (damage to the left hemisphere): Te topographic map showed that after stimulating the left M1 area, the energy of the damaged left brain area increased signifcantly; that is, TMS activated the AH of the patient. Te time-Cerebral infarction in middle cerebral artery territory ( onset ≤ 1 month )

Meet all the inclusion and exclusion criteria
Baseline data collection Risk factors (gender, age, cerebrovascular disease, and size of infarction location, clinical movement function score and resting state of eeg data, etc.) sTMS was performed to record EEG data and clinical motor function score during stimulation ( 1 month after onset ) and after stimulation ( 3 months, 6 months after onset ) Collect data, statistical analysis International Journal of Clinical Practice  International Journal of Clinical Practice frequency analysis indicated that when stimulating the left M1 area using TMS, the energy increase and the excitation efect in the left M1 area were higher than in the right M1 area. S10 (damage to the left hemisphere): Te topographic map showed that after stimulating the left M1 area, the occipital lobe area decreased signifcantly. Te timefrequency analysis indicated that when stimulating the left M1 area using TMS, the energy increase of the left M1 area was higher than in the right M1 area. S11 (damage to the right hemisphere): Te topographic map showed that after stimulating the right M1 area, only the occipital lobe showed a small range of energy increases. Te time-frequency analysis indicated that when stimulating the right M1 area using TMS, the energy rise and the excitation efect in the right M1 area were higher than in the healthy left M1 area. S12 (damage to the left hemisphere): Te topographic map showed that after stimulating the left M1 area, the whole brain did not change signifcantly. Te timefrequency analysis indicated that when stimulating the right M1 area using TMS, the energy decreased briefy after stimulation and then rose back to the baseline level. S13 (damage to the left hemisphere): Te topographic map showed that after stimulating the left/right M1 area, the contralateral inhibition efect was signifcant, but no significant change from the time-frequency analysis was observed. S14 and S15 (damage to the right hemisphere): Tere was no signifcant change in the brain topographic map and time-frequency analysis map. International Journal of Clinical Practice S16, damage to the right hemisphere: Te topographic map showed that the energy of the occipital lobe was decreased slightly and it was inhibited, but no obvious change from the time-frequency analysis was observed.
In summary (Figure 3): Patients S1, S2, S7, and S9: Stimulation therapy using TMS should be applied to the afected motor area to enhance the activation of the afected side of the brain. At the same time, physical and occupational therapy should be provided to accelerate the nerve remodeling and motor function reconstruction of the afected side.
Patients S3, S5, S6, S10, and S11: After TMS treatment, the motor area of the injured side did not show obvious changes, and it was compensated by other brain regions. For these types of patients, the function of the compensatory brain area should be strengthened, combined with TMS, transcranial electrical stimulation, and other activation/inhibition treatment methods to activate the compensatory brain area. At the same time, the migration trend of the compensatory brain area should be observed, and the stimulation site should be adjusted accordingly.
Patients S4, S8, and S12-S16: TMS did not produce satisfactory energy activation phenomenon for such patients, so other treatments are suggested.

Discussion
After stroke, various neurophysiological characteristics of the brain tissue change. Te treatment strategy aims to increase neuronal plasticity, relearn processes and functional reorganization, form new neural networks and programs, improve functioning, and promote recovery. At present, the results of rTMS in the treatment of patients with stroke are diferent due to diferent parameters. Further research is needed to determine the best treatment method.
Lanza et al. found [2] TMS can be exploited as a noninvasive tool able to evaluate in vivo the cortical excitability, the propension to undergo neural plastic phenomena, and the underlying transmission pathways. Te combination of TMS and EEG technology is highly praised for its simplicity and high time resolution advantages. It can provide a variety of neurophysiological processes at the same time and gradually shows its potential in evaluating the neurophysiological mechanism related to brain network reorganization after stroke and becoming a neurophysiological marker for functional recovery of stroke.
In this study, the EEG parameters of TMS in patients with stroke were evaluated, such as EEG power spectrum density, TEP, and brain network connectivity and their correlation with injury type, injury degree, and motor function prognosis were analyzed.

Evaluation of Whole Brain Electroencephalogram Power
Spectral Density (Brain Topographic Map). Te lower frequency band of TMS and delta was selected to interfere with the motion of TMS and delta. Alpha (beta)-band energy can be used as an evaluation index. At present, the peak frequency of the alpha band is often used in current research, which varies among diferent populations. Quantitative EEG using spectrum content power spectrum analysis has been applied to stroke, and relevant studies have shown that there S9 S10 S11 S12 S13 S14 S16 S15 FMA MMSE ADL  International Journal of Clinical Practice is a correlation between spectrum power and the level of disability after stroke [38].
In the resting state, the motor area energy of patients with stroke was lower than that of the whole brain. According to the brain topographic map, the power spectrum changes of 1,000 ms after stimulation were not observed in patients with stroke compared with those before stimulation. Tere was no consistent diference in the brain topographic maps of healthy people, but the energy change of the brain topographic maps in task state had good consistency. Compared with the active task, TMS cannot cause regular changes in the whole brain energy distribution, suggesting that the latter may depend on spontaneous neurophysiological activities or may be due to the diferences in the location, degree, and duration of brain injury in patients with stroke. Te stimulation site and frequency of TMS are relatively fxed, which needs to be further studied.

Electroencephalogram Time-Frequency Analysis Results.
When stimulating the unilateral M1 area using TMS, the energy of the bilateral M1 area and the occipital 8-40 Hz band were signifcantly higher than that in the resting state. Most patients' stimulation side was higher than the contralateral side. Te energy of S1-S11 increased signifcantly for 100 ms and then returned to the initial state.
Tese results showed that TMS infuenced cortical electrophysiological activities. Some studies have found that the increase of alpha band energy is related to the improvement of clinical symptoms of stroke. Whether the energy increase of each band in this study is related to the recovery of motor function needs to be verifed. If the electrophysiological changes under the action of TMS are related to the prognosis, TMS can be used as a means of rehabilitation, and its corresponding electrophysiological indicators can also predict functional prognosis.
As there were diferences in bilateral energy increase after stimulation, if there are similar diferences in healthy controls, it indicates that there may be attenuation of cortical activity transmission after TMS, so the energy recorded on the opposite side of stimulation is lower.
If the healthy control is bilateral symmetrical, it may be that the cortical functions such as excitability, connectivity, and other characteristics change during stroke, which may cause reorganization, movement, subcortical circuit activation/inhibition, or neurotransmitter imbalance. Te recovery of motor function (Fugl-Meyer score diference) was compared between symmetrical and asymmetric patients with bilateral energy increase; if there was a statistical difference between the two, then this energy increase may refect the state of intercortical activity after stroke, which can be used as an indicator to assess prognosis.
Diferent people have diferent responses to TMS stimulation. In this study, bilateral EEG energy increased after TMS treatment, which is inconsistent with some previous studies. What needs to be studied in the future is the following: is the hemisphere with high responsiveness related to the location, size, and onset time (time window) of TMS stimulation? Is the prognosis of patients with a good prognosis related to high energy on the same side of the lesion? Will the interhemispheric inhibition (stimulating one side of the M1 area can inhibit the contralateral M1) in a physiological state change after a stroke? 4.3. Brain Connectivity. Te clinical outcome of stroke is related to local brain injury and to other parts of the connection. After a stroke, the coupling between hemispheres changes. Brain connectivity is considered as an index to study the mechanism of poststroke rehabilitation and evaluate the efect of some rehabilitation intervention measures. It can promote the rehabilitation process and produce adverse efects.
In this study, patients with stroke were found to have decreased brain connectivity compared with healthy controls in a resting state. After TMS stimulation, the brain connectivity of patients with stroke and healthy volunteers were signifcantly improved compared with forebrain stimulation, indicating that TMS can cause changes in the connectivity of the cerebral cortex and other parts but cannot quantitatively evaluate motor function and its correlation. Whether brain connectivity can be used as an effective guidance tool for TMS to treat stroke motor dysfunction remains to be studied.

Limitation of Tis Study.
Tere are some limitations to our study, most notably is the small number and the rather heterogeneity in clinical features within the control group. Te sample size is small, and it is impossible to judge the specifc connection between the exact brain injury site and the neurophysiological characteristics of brain tissue, such as the cortical connectivity and TEP. In the future, it is necessary to explore the abnormal patterns and quantitative indicators of cortical activity at diferent times and in different parts after stroke and the specifc relationship between abnormal patterns and neural function recovery, with an increased sample size. And the study of expanding the sample size is necessary to further determine the stability of the results. Due to the changes in cortical activity after stroke, the choice of TMS stimulation frequency and stimulation site needs to be personalized. Meanwhile, the changes in neurocortical plasticity at diferent stages of stroke should be considered and continuous observation is required. In addition, TMS itself generates noise, which may contaminate the activities induced by TMS. At present, there is no unifed standard for the postprocessing of EEG signals, and multicenter validation is still needed to determine the reliability of TMS-EEG-related neurophysiological markers. In addition, studies have shown that transcranial direct (tDCS) and alternating current (tACS) stimulation have low electrical stimulation intensity on the scalp and have been widely used in this feld [39], but there is a lack of comparative study of these three methods.

Conclusions
In healthy subjects and patients with stroke, TMS stimulation will enhance brain connectivity, and some brain regions will be activated, but diferent brains will react to it inconsistently, which seems to be related to the disease process and progress. In some patients, such changes in brain connectivity and activation of brain regions may be considered as adaptive responses to disease progression. Tis change of the brain changes with the time of stroke, its brain connectivity and activated brain areas will also change. Te diferentiation of TMS treatment efect seems to be consistent with it. Terefore, it is suggested that TMS treatment should be individualized rather than "one size fts all." Tere were no statistically signifcant diferences between the FMA, MMSE, and ADL indicators. For patients with different types of injury, degree of injury, and diferent onset period, individualized intervention measures should be adopted. Te evaluation methods should be as diverse as possible, and the rehabilitation efect of patients should be evaluated from multiple angles to avoid the limitation of any single evaluation factor.

Data Availability
Te data that support the fndings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.