Genetic algorithm-based optimal repertoire selection for music therapy on therapeutic effects
Publicado en línea: 21 mar 2025
Recibido: 08 nov 2024
Aceptado: 10 feb 2025
DOI: https://doi.org/10.2478/amns-2025-0692
Palabras clave
© 2025 Ming Chen et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
As one of the components of Chinese medicine, traditional rehabilitation medicine has a long history, which not only has unique theories, but also has a variety of colorful and effective treatment methods, such as recreational therapy, affective therapy, as well as acupuncture, guided, dietary, and pharmacological therapies, which have made an important contribution to the development of China’s rehabilitation medical career [1-3]. According to the classification of modern rehabilitation medicine, music therapy is closely related to physical therapy, occupational therapy and psychotherapy, which embodies a comprehensive treatment modality and plays a positive role in disease prevention, clinical treatment and post-disease rehabilitation [4-6]. The history of treating diseases with “music” can be traced back to the form of witchcraft in primitive society. In ancient times, the ancestors knew little about the nature of life and health, as well as the mechanism of the formation of natural phenomena, such as thunder, lightning, rain and snow, so they often held some songs and dances in order to hope for the blessing of the gods [7-9]. In the process, some diseases of people were cured accidentally, so the song and dance treatment gradually became one of the treatment methods of ancient witch doctors. As people’s understanding of things in nature increased, especially after herbal medicine gradually became the main means of treating diseases, music therapy slowly faded out of the medical field [10-12]. Even so, music is still an important and indispensable part of people’s daily life. It is only in modern times that music therapy, an ancient and young therapeutic modality, has gradually returned to the field of medical treatment and has become an independent therapeutic discipline that has been widely accepted [13-15]. Traditional Chinese music therapy integrates various forms of expression such as song chanting, playing, listening and dancing, and can be combined with other therapies at the same time, with rich content and various forms, which is not only preserved in medical books of all times, but also recorded in history books and anthologies. Music therapy is a treasure trove to be developed, which can be applied to the rehabilitation of many diseases, and its value has been fully recognized by medical practitioners throughout the ages [16-19]. Different music tracks play different therapeutic effects, so it is necessary to explore the genetic algorithm-based music therapy optimization track selection on the therapeutic effect, for clinical rehabilitation reference, and hope to take this opportunity to increase the importance of music therapy, so that it develops into a relatively independent, universally applied rehabilitation therapy [20-23].
Starting from traditional music therapy, the article integrates genetic algorithms into music therapy to optimize repertoire selection for music therapy. The effective core repertoire in music therapy with patient relevance is mined, and a repertoire mining model based on an improved genetic algorithm is constructed. Taking the core effective repertoire as the benchmark, the music therapy repertoire decision-making recommendation algorithm is designed through similarity calculation to recommend personalized therapy repertoire for patients. Then the research design was carried out to develop a treatment effect measurement scale based on SDS, SAS, HAMD, HAMA, and PSQI scales. The pre- and post-test treatment effects of the experimental group and the control group are compared, and whether there is a significant difference between the treatment effects of the two groups before and after the experiment, so as to test the effectiveness of the music therapy based on the optimization of repertoire by genetic algorithm in this paper.
This study uses a quasi-experimental research design, which is a pre and post-test static group comparison design. Quasi-experimental research method is a method of social science research, which is different from the real experimental design in that the subjects of quasi-experimental design are not randomly assigned, and its advantage lies in the flexibility of the experimental conditions, which makes the quasi-experimental design widely applicable in the case where it is not possible to control the irrelevant variables that may affect the experimental results. In this study, the control group receives conventional medication and music therapy, and the experimental group receives medication and music therapy optimized by a genetic algorithm, and subjects in both groups complete pre and post-tests of the relevant scales, and the researcher will analyze the changes in the scores of the experimental group and the control group on the pre-test and the post-test.
In this experiment, subjects were recruited in the outpatient clinic of the Department of Psychology at First People’s Hospital of W. The subjects were all adolescents (13-18 years old) with depression, and a total of 48 subjects were recruited and divided into the experimental group and the control group, each with 24 subjects. The cultural backgrounds of the subjects were basically similar.
Subjects’ inclusion criteria: physician-diagnosed behavioral and mood disorders, depressive disorders, and depressive states that usually develop in childhood and adolescence. They were seen in the outpatient clinic of the First People’s Hospital of W city and received only conventional medication and music therapy. All subjects’ results on the SDS and HAMD scales were significant depressive symptoms (i.e., SDS scores of 53 or more and HAMD scores of 7 or more). The subjects were informed and took part voluntarily, and their families were also informed and consented.
Subject exclusion criteria: patients with a diagnosis of mental illness or with psychotic symptoms such as hallucinations and delusions. Not receiving antidepressant medication. Concurrent treatment other than medication, such as psychotherapy, physiotherapy, etc. Accompanied by verbal communication disorder or hearing impairment.
The subjects were recommended by the outpatient psychiatrist, and were prescribed a psychological evaluation of the Self-Depression Scale (SDS), the Self-Anxiety Scale (SAS), the Hamilton Depression Scale (HAMD), the Hamilton Anxiety Scale (HAMA), the Pittsburgh Sleep Quality Index (PSQI), and the HAMD and HAMA scores by a regular psychometrician, HAMA) scores. After the recruitment of the experimental group, the subjects will undergo group music therapy, while the recruitment of the control group subjects will be carried out, and both groups will need to be interviewed and assessed before enrollment.
The control group of 24 subjects will receive conventional medication and music therapy. The experimental group of 24 subjects received medication and music therapy with repertoire optimization by a genetic algorithm. Measurement scores of SDS, SAS, HAMD, HAMA, PSQI scales of the experimental and control groups were collected four weeks before and after the experiment. The study was approved by the Ethics Committee of the First People’s Hospital of W. The subjects were introduced to the study process and signed an informed consent form.
The methods of music therapy are generally categorized into three types: receptive music therapy, improvisational music therapy, and recreational music therapy [24]. Receptive music therapy is also known as listening or passive music therapy, the specific method is to regulate the physical and mental health of the therapist by listening to music, under the correct guidance of the music therapist, by letting the therapist listen to the selected music tracks and then discussing the content of the music, and then guiding the therapist to carry out musical imagery or recollection of the music content within the music scene, and then passively accepting the music intervention techniques to induce the experience and feelings of the therapist towards music, and make him or her achieve the short-term music therapy. The passive acceptance of the music intervention technique causes the client to experience and feel the music, so that they can achieve the short-term music therapy goal. Re-creation music therapy mainly emphasizes the ability of the client to join in the music therapy on his/her own, to actively participate in music activities on the basis of listening to the music, or to sing or play the music, so as to achieve the short-term goals of music therapy. Improvisational music therapy mainly focuses on improvisation, which requires the music therapist to have the ability to play a variety of musical instruments or improvisation, generally using relatively simple and convenient instruments to guide the therapist to join the music performance process, so as to stimulate the therapist’s improvisational creativity and musical aesthetic ability, so as to achieve therapeutic effects.
Through the above methods, clients can be stimulated in music therapy to develop their listening ability, feeling ability, discriminating ability, imagination, memory, music aesthetics, and inner awareness. Listening, singing, playing, performing, and creating music through technical means can present different musical and artistic expressions through different objects. This study mainly uses receptive music therapy and recreational music therapy, and the specific music therapy methods include music synchronization, music warm-up, music meditation, music discussion, and song singing. Music therapy can stimulate the brain neurotransmitters to improve the cerebral cortex, play the mechanism of analgesic and emotional two-way regulation to relieve hemiplegic patients in the rehabilitation process of the pain in the affected limbs, relieve the negative emotions of the patients, so that the patients can release their hearts, regain confidence in music therapy, help patients to improve the ability of the affected limbs to move and take care of their own lives, improve the quality of life, and return to the society as soon as possible.
Optimized design for repertoire selection Genetic Algorithm (GA) simulates the phenomena of reproduction, mating and mutation occurring in the process of natural selection and natural heredity, and according to the natural law of survival of the fittest and survival of the fittest, the population is made to evolve to better and better regions in the problem space through random selection, mutation and crossover from any initial population [25-26]. Genetic algorithm is an adaptive search technology, its selection, crossover, mutation and other operations are carried out in a probabilistic way, thus increasing the flexibility of the search process, at the same time, it can converge to the optimal solution with a large probability, and it has a better global optimization solving ability. In this paper, a genetic algorithm is introduced in music therapy to optimize the selection of therapeutic repertoire, which mainly involves the following steps: Determine the evaluation standard: first, according to the evaluation standard of effective diagnosis and treatment situation, the music therapy repertoire of all patients is initially screened to obtain an initial effective repertoire set. Genetic Algorithm Optimization Search: next, the initial effective prescription set is optimally searched using the parallel global search optimization search capability of the genetic algorithm. The goal of this step is to mine the core effective repertoire set that best characterizes the patient’s effective repertoire set. Recommendation degree analysis: finally, based on the core effective repertoire set obtained from the optimized search, the repertoire recommendation degree analysis is carried out according to the differences in the patients’ evidence types. The goal of this step is to customize the personalized recommended repertoire according to the patient’s specific condition and evidence type in order to improve the therapeutic effect. Core effective track mining model construction The construction of an effective track mining model based on genetic algorithms is mainly divided into the following six steps: Randomly generating the initial repertoire set: in this study, the set of repertoire sets corresponds to the population in the algorithm, each repertoire set corresponds to a chromosome, and repertoire corresponds to the genes in the chromosome. At the time the genetic algorithm (GA) was initiated, we were unable to determine which track sets were more effective because the original dataset had not yet been analyzed. Therefore, during the initial construction phase of the model, an initial track set is randomly generated by the algorithm to serve as the initial population for the algorithm. This repertoire set was represented by a series of binary strings of numbers, where each string of numbers represented a prescription. In these binary number strings, the number 1 indicates the presence of a repertoire in the corresponding position, while the number 0 indicates the absence of a repertoire in the corresponding position. Taking Moonlight Sonata as an example, if the first digit of the binary number string corresponding to track set 1 is 1, it means that track set 1 contains Moonlight Sonata and vice versa. This representation helps in the algorithm for randomly generating the initial population. Calculation of fitness for the repertoire set. The fitness calculation is performed for each track in the repertoire set and is the sole criterion for selecting which tracks will go to the next generation, reflecting the degree of merit of the tracks. In genetic algorithms, tracks with greater fitness are more likely to be retained during the iteration process, so the design and calculation of fitness have an important impact on the overall performance of the genetic algorithm. This study innovates on the basis of the traditional genetic algorithm and proposes an innovative method for calculating the track adaptation degree, which is given in the following equation (1):
where “
By iterating through the above fitness function, the core track set will be the brand new track set that can most comprehensively show the characteristics of the original dataset. Through this process, we get the core track set with high fitness value, which reflects its good matching degree with the original dataset.
Selection operation. The basic principle of the selection operation is that the track sets with better fitness have a higher probability of being selected, which helps to guide the GA in the direction of searching for more optimal solutions continuously. The selection method used in this study is the “roulette” selection method, which is simple and effective. It is a proportionality-based selection method, which is consistent with the basic principles of selection operators. In this method, the probability of a track set being selected is proportional to its own fitness value. Assuming that there are
It is clear from equation (2) that the higher the fitness value, the higher the probability that the set of tracks will be selected for the next generation.
Crossover operation. The purpose of introducing the crossover operation is to enhance the GA’s ability to cover the global search, to better explore the combinations between different track sets, and to ensure the diversity of the track set collection. Through the crossover operation, completely new track sets can be generated, which ensures that the collection of track sets continues to evolve forward in the evolutionary process without falling into a stagnant state. Mutation operations. Mutation operation usually occurs with low probability, when the fitness values of all the tracks in the collection of tracks converge, it is difficult to generate new tracks with more obvious differentiation through selection and crossover operation, in this case, the mutation operation can ensure the generation of new tracks, thus avoiding the genetic algorithm to fall into the local optimal solution as much as possible. Termination condition judgment. The termination condition of the genetic algorithm can be that the fitness value reaches the expected value, or the number of iterations reaches the maximum number of iterations set by the algorithm. The study set the maximum number of iterations as 500 to ensure that relatively good optimization results are obtained within a certain range.
The Self-Depression Scale (SDS), which is commonly used to rate a person’s subjective state of depression and its change in treatment, contains 20 questions reflecting subjective feelings of depression, and the rater selects four different ratings based on their self-perceptions, 10 of which are positive and 10 negative. The cut-off value of the SDS International Standard Score is 53, and the rating scale is: 0-52 is normal, 53-62 is mild depression, 63-72 is moderate depression, and 73 or more is severe depression. The Self-Assessment Scale for Anxiety (SAS), commonly used to rate a person’s subjective state of anxiety and its changes in treatment, contains 20 questions reflecting subjective feelings of anxiety, and the rater selects four different scales to rate according to his or her self-perceptions, with 15 positive scales and 5 negative scales.The cut-off value of the international standardized score of the SAS is 50, and the rating scales are as follows: 0-49 normal, 50-59 mild anxiety, 60-69 moderate anxiety, and 69 or more severe anxiety. 50-59 is considered mild anxiety, 60-69 is considered moderate anxiety, and 69 or more is considered severe anxiety. The Hamilton Depression Scale (HAMD), which is administered by a trained rater who examines the patient for HAMD and generally scores the patient using conversation and observation. The rating scale is: less than 8 indicates no depressive symptoms, more than 8 indicates possible depression, more than 20 indicates significant depression, and more than 35 indicates severe depression. Hamilton Anxiety Scale (HAMA), which is mainly used to assess the severity of anxiety symptoms in neurological and other patients, is administered by a trained rater who examines the patient’s HAMA, and generally scores the patient using conversation and observation. The rating scale is: less than 7 is no anxiety symptoms, more than 7 is possible anxiety, more than 14 is possible significant anxiety, more than 21 is very significant anxiety, and more than 29 is possible severe anxiety. The Pittsburgh Sleep Quality Index (PSQI) is commonly used to assess the quality of sleep in patients with sleep disorders, psychiatric disorders, and also to assess the quality of sleep in the general population. The rating scale is: 0-5 points sleep quality is very good, 6-10 points sleep quality is OK, 11-15 points sleep quality is average, 16-21 points sleep quality is very poor.
Combining information from SDS, SAS, HAMD, HAMA, and PSQI scales, the Treatment Effectiveness Scale was compiled as a measure of the therapeutic effect of music therapy based on the optimization of repertoire by genetic algorithms. Self-impression, self-efficacy, emotional control, physical health, mental health, interpersonal adaptation, family atmosphere, balance of mind, and sleep quality were used as dimensions to measure the therapeutic effect.
Before the start of the formal experiment, the total scores of the treatment effect scale and the scores of each dimension of the subjects in the experimental group and the control group were statistically analyzed. The results of the pre-test are shown in Table 1, which shows that there is no significant difference between members of the experimental group and members of the control group in terms of the total scores of the scale and the scores of the dimensions, and it can be indicated that the members of the experimental group and the control group are homogeneous before the experiment.
Pre-test therapeutic effect comparison of experimental and control group
| Item | Group | N | M | SD | F | t | p |
|---|---|---|---|---|---|---|---|
| Self-impression | Experimental group | 24 | 8.45 | 1.84 | 0.324 | -0.429 | 0.826 |
| Control group | 24 | 8.65 | 1.64 | ||||
| Self-efficacy | Experimental group | 24 | 8.29 | 1.39 | 0.355 | 0.727 | 0.793 |
| Control group | 24 | 8.25 | 1.35 | ||||
| Emotional control | Experimental group | 24 | 8.84 | 1.62 | 0.688 | -1.069 | 0.691 |
| Control group | 24 | 9.06 | 1.86 | ||||
| Physical health | Experimental group | 24 | 8.27 | 1.46 | 0.223 | -0.684 | 0.467 |
| Control group | 24 | 8.43 | 1.79 | ||||
| Mental health | Experimental group | 24 | 8.55 | 1.50 | 1.021 | 0.545 | 0.489 |
| Control group | 24 | 8.38 | 1.68 | ||||
| Interpersonal adaptation | Experimental group | 24 | 8.16 | 1.68 | 0.122 | -0.809 | 0.414 |
| Control group | 24 | 8.37 | 1.78 | ||||
| Family atmosphere | Experimental group | 24 | 9.26 | 1.37 | 1.302 | 0.829 | 0.503 |
| Control group | 24 | 9.12 | 1.42 | ||||
| Mentality balance | Experimental group | 24 | 8.80 | 1.80 | 0.205 | -0.281 | 0.699 |
| Control group | 24 | 8.87 | 1.87 | ||||
| Sleep quality | Experimental group | 24 | 8.75 | 1.78 | 0.756 | -0.182 | 0.524 |
| Control group | 24 | 8.81 | 1.52 | ||||
| Total | Experimental group | 24 | 77.37 | 5.15 | 0.621 | -0.584 | 0.647 |
| Control group | 24 | 77.94 | 5.73 |
From Table 1, it can be seen that on the treatment effect scale and its nine dimensions, the between-group variance on the scores of the experimental group and the control group was chi-square in general, so the significance test of between-group means was conducted using the independent samples t-test, and it was found that the between-group differences of the ten indexes did not reach significance (p>0.05).
According to the above results, the experimental group and the control group were basically the same level in the total score of the Treatment Effectiveness Scale and the scores of the dimensions, and the starting point of all subjects was basically the same. If there is a difference in the results of the post-test, it can be attributed to a large extent to the fact that the music therapy based on genetic algorithm optimization of repertoire adopted in this study has played a role.
At the end of the eight-week treatment experiment, scale scores were measured using scales for both members of the experimental and control groups, and the measured data were statistically analyzed. In order to exclude the influence of maturity, history and other additional variables related to the course of time, and to better verify the effect of the treatment experiment, instead of directly testing the significance of the post-test scores of the two groups, group statistics and significance tests were carried out with the amount of growth obtained by the experimental group and the control group in the treatment effect and in each of the dimensions (growth=post-test-pre-test), and the results are shown in Table 2. Using independent samples t-test to test the significance of the difference between the means of the growth amount of each index of the treatment effect, the results show that the experimental group’s total score of the treatment effect scale and the scores of each dimension from the pre-test to the post-test showed a certain degree of progress trend, and the improvement of each dimension was around 5 points. The control group, on the other hand, showed very little improvement in the scores on the scale and the dimensions, none of which exceeded 0.5 points, without much change. The experimental results provide support for the hypothesis that music therapy based on genetic algorithm optimized repertoire can have a good effect on the experimental group, i.e., music therapy based on genetic algorithm optimized repertoire can enhance the therapeutic effect of adolescents.
Increment comparison of therapeutic effect of experimental and control group
| Item | Group | N | M | SD | F | t | p | Cohen’s d |
|---|---|---|---|---|---|---|---|---|
| Self-impression | Experimental group | 24 | 5.32 | 1.73 | 1.584 | 7.462 | 0.000 | 3.86 |
| Control group | 24 | 0.23 | 0.19 | |||||
| Self-efficacy | Experimental group | 24 | 5.79 | 3.26 | 3.624 | 7.952 | 0.000 | 4.12 |
| Control group | 24 | 0.17 | 0.08 | |||||
| Emotional control | Experimental group | 24 | 6.33 | 3.02 | 0.422 | 8.041 | 0.000 | 5.93 |
| Control group | 24 | 0.03 | -0.18 | |||||
| Physical health | Experimental group | 24 | 6.79 | 3.37 | 1.632 | 7.223 | 0.000 | 5.74 |
| Control group | 24 | 0.13 | 0.05 | |||||
| Mental health | Experimental group | 24 | 6.46 | 3.01 | 0.694 | 6.995 | 0.000 | 5.26 |
| Control group | 24 | 0.09 | -0.24 | |||||
| Interpersonal adaptation | Experimental group | 24 | 7.43 | 2.35 | 0.487 | 15.521 | 0.000 | 6.95 |
| Control group | 24 | 0.07 | 0.05 | |||||
| Family atmosphere | Experimental group | 24 | 5.20 | 3.42 | 0.589 | 7.346 | 0.000 | 3.92 |
| Control group | 24 | 0.17 | -0.09 | |||||
| Mentality balance | Experimental group | 24 | 4.01 | 1.65 | 1.702 | 7.558 | 0.000 | 2.84 |
| Control group | 24 | 0.06 | -0.25 | |||||
| Sleep quality | Experimental group | 24 | 4.34 | 2.10 | 1.632 | 8.027 | 0.000 | 2.76 |
| Control group | 24 | 0.06 | 0.00 | |||||
| Total | Experimental group | 24 | 51.67 | 5.15 | 0.412 | 42.565 | 0.000 | 18.54 |
| Control group | 24 | 1.01 | 0.28 |
After the total score of treatment effect and the scores of each dimension of the members of the experimental group before and after the experiment were tested for normality, the Wilcoxon signed ranks test was used for the indexes that did not conform to the normal distribution, and the paired samples t-test was used for other indexes, to compare whether there was any difference between before and after the experiment, and the results are shown in Table 3. The results showed that the total score of treatment effect of the members of the experimental group was 51.67 points higher than that of the pre-experiment, and the difference was statistically significant (P < 0.05). Analysis of the dimensions of the scale showed that in terms of self-impression, self-efficacy, emotional control, physical health, mental health, interpersonal adaptation, family atmosphere, balance of mind, and sleep quality, members of the experimental group scored higher than before the experiment, with a score difference ranging from 4.01 to 7.43, and the difference was statistically significant (P < 0.05).
Pre-test and post-test therapeutic effect comparison of experimental group
| Item | Pre-test | Post-test | t/Z | p | ||
|---|---|---|---|---|---|---|
| M | SD | M | SD | |||
| Self-impression | 8.45 | 1.84 | 13.77 | 3.57 | -6.588 | 0.000 |
| Self-efficacy | 8.29 | 1.39 | 14.08 | 4.65 | -7.420 | 0.000 |
| Emotional control | 8.84 | 1.62 | 15.17 | 4.64 | -6.614 | 0.000 |
| Physical health | 8.27 | 1.46 | 15.06 | 4.83 | -5.072 | 0.000 |
| Mental health | 8.55 | 1.5 | 15.01 | 4.51 | -6.423 | 0.000 |
| Interpersonal adaptation | 8.16 | 1.68 | 15.59 | 4.03 | -4.789 | 0.000 |
| Family atmosphere | 9.26 | 1.37 | 14.46 | 4.79 | -7.113 | 0.000 |
| Mentality balance | 8.80 | 1.80 | 12.81 | 3.45 | -6.348 | 0.000 |
| Sleep quality | 8.75 | 1.78 | 13.09 | 3.88 | -6.245 | 0.000 |
| Total | 77.37 | 5.15 | 129.04 | 10.30 | -10.985 | 0.000 |
Music therapy is a scientifically rigorous yet emotional form of treatment, where music plays a psychologically soothing role in the treatment of patients during the process of medication of the human organism. And traditional music therapy often generalizes a set of therapeutic programs. The music therapy proposed in this paper utilizes genetic algorithms to improve the selection of music repertoire and exclude the repertoire with higher adaptability. In addition, according to the patients’ different disease manifestations and musical interest orientations, the music therapy in this paper can also perform personalized music repertoire recommendation to explore the best treatment effect. From the experimental results, it can be seen that the effect of music therapy after the optimization of repertoire by genetic algorithm is indeed better than that of conventional music therapy, which confirms the research conjecture of this paper.
After the total score of treatment effect and its scores of each dimension of the members of the control group before and after the experiment were tested for normality, the paired-samples t-test was used to compare whether there was any difference before and after the experiment, and the results are shown in Table 4. The results show that the difference between the scores of the various dimensions of the treatment effect of the members of the control group before and after the experiment ranges from 0.03 to 0.23, and although the treatment effect after the experiment has been improved, the enhancement is too small, and the difference is not statistically significant (P > 0.05).
Pre-test and post-test therapeutic effect comparison of control group
| Item | Pre-test | Post-test | t/Z | p | ||
|---|---|---|---|---|---|---|
| M | SD | M | SD | |||
| Self-impression | 8.65 | 1.64 | 8.88 | 1.83 | -0.413 | 0.688 |
| Self-efficacy | 8.25 | 1.35 | 8.42 | 1.43 | -0.622 | 0.601 |
| Emotional control | 9.06 | 1.86 | 9.09 | 1.68 | -0.476 | 0.760 |
| Physical health | 8.43 | 1.79 | 8.56 | 1.84 | -0.373 | 0.508 |
| Mental health | 8.38 | 1.68 | 8.47 | 1.44 | -0.288 | 0.908 |
| Interpersonal adaptation | 8.37 | 1.78 | 8.44 | 1.83 | -0.458 | 0.446 |
| Family atmosphere | 9.12 | 1.42 | 9.29 | 1.33 | -0.268 | 0.735 |
| Mentality balance | 8.87 | 1.87 | 8.93 | 1.62 | -0.355 | 0.813 |
| Sleep quality | 8.81 | 1.52 | 8.87 | 1.52 | -0.523 | 0.681 |
| Total | 77.94 | 5.73 | 78.95 | 6.01 | -1.276 | 0.582 |
Although traditional music therapy has a regulating effect on the patient’s psychology, the “one-size-fits-all” type of “therapeutic formula” does not pay attention to the differences between individual patients, so the therapeutic effect is prone to uneven phenomena, the lack of targeted treatment leads to the overall treatment of The lack of targeted treatment has led to the overall treatment not seeing significant results.
The author introduces genetic algorithm into music therapy to optimize music repertoire selection in therapy through genetic algorithm for personalized repertoire decision recommendation. A related study is designed to compare music therapy based on genetic algorithm optimization of repertoire proposed in this paper with conventional music therapy to test the therapeutic effects of this paper’s method.
Before the experiment, the differences between the experimental and control groups in terms of the total score and the nine sub-dimensions of the treatment effect scale were not statistically significant (p>0.05). After the experiment, the treatment effect of the two groups showed a large difference, and the total score of the treatment effect of the experimental group was 50.09 points higher than that of the control group, and the experimental group was higher than the control group in the 9 dimensions of self-impression, self-efficacy, emotional control, physical health, mental health, interpersonal adaptation, family atmosphere, mental balance, and sleep quality, and the difference between the two groups in the total score and the dimensions were statistically significant (p< 0.05).
The experimental group achieved [4.01,7.43] improvement in all dimensions of the treatment effect, while the control group’s scores in all dimensions after the experiment were improved, but the improvement was not more than 0.5 points. The experimental group showed a significant difference (p<0.05) in the pre- and post-test treatment effect, while the control group did not show any significant difference (p>0.05).