Exploring the role of moxibustion robots in teaching: a cross-sectional study (2025)

Background

Artificial intelligence has gradually been used into various fields of medical education at present. Under the background of moxibustion robot teaching assistance, the study aims to explore the relationship and the internal mechanism between learning engagement and evaluation in three stages, preparation before class, participation in class, and consolidation after class.

Methods

Based on the data investigated in 250 youths in university via multistage cluster sampling following the self-administered questionnaire, structural equation model was built to discussing factors of study process about moxibustion robots.

Results

It was found after moxibustion robot teaching assistance that preparation before class, participation in class and consolidation after class positively predicted learning engagement. Learning engagement, preparation before class, participation in class, consolidation after class positively predicted effect evaluation. Learning engagement played a mediating role in the effect of preparation before class and consolidation after class on evaluation.

Conclusion

Employing artificial intelligence in three stages of class can improve the quality and efficiency of medicine education and promote its innovation and development. Serviceable and valuable reference and inspiration for future teaching improvement and industrial development can be provided via the systematic research and analysis of the practical application of moxibustion robot in teaching.

Background

Artificial Intelligence (AI) has gradually been used into various fields of medical education with the rapid development of technology, which brings revolutionary changes to medical education [1, 2] with its powerful data processing capabilities, accurate analysis algorithms and intelligent application scenarios. Combining medical education with artificial intelligence can not only improve the quality and efficiency of medical education, but also promote the innovation and development of it [3]. The application of artificial intelligence in medical education is gradually changing the traditional teaching ways and methods, providing medical students with a more rich, efficient and personalized learning experience [4].

On the one hand, medical education adopts intelligent assisted diagnosis system, which utilizes machine learning algorithms and big data analysis to help medical students carry out diagnosis and determine treatment plan [5]. By automatically identifying indicators such as the focal size, type and location and analyzing image features, the system can provide accurate diagnostic recommendations, which enhances students’ practical experience [6]. On the other hand, via using virtual simulation technology to re-render real medical scenarios, training opportunities to operate clinical surgery can be better offered. This technique reduces risk and stress for patients, creates a safe learning environment for students, and helps improve their judgment and decision-making skills [7,8,9].

Additionally, artificial intelligence plays a role in individualized teaching in medical education as well. By emphasizing an AI-integrated framework which enhance radiology education and take practices offered by our own institution for instance, Duong suggests that the era of AI-augmented Radiology could enable not only “precision medicine” but also “precision education”, where teaching is tailored to the students’ learning style and needs [10]. Pantic built a novel learning framework that fits the goals of the course and the expected level of computing skills of the students for programming instruction to novices through a synthesis of traditional objectivist and real-world constructivist approaches [11]. Thomaz designed an experimental platform with a simulated RL robot, analyzed real-time human teaching behavior found in a study, reported on how people manage feedback by using AI after learning from the study reinforcement classes [12]. Goh discussed a new paradigm [13] of biotechnology education involving AI coevolution to help students learn adaptively and build connections among new conception in the vast knowledge network of biotechnology. Cheng used AI to analyze the effectiveness of the realization of inverted classroom education in universities, and expanded the dimension and depth of the analysis of inverted sports behavior in the physical education teaching of inverted university classrooms [14]. Yilmaz compared the effects of learning through real-time intelligent teaching systems with that through human instructor-mediated training [15]. By using natural language processing, machine learning and other technologies, the system can comprehensively and accurately assess students’ abilities, provide real-time feedback and personalized learning suggestions by analyzing students’ learning process and level, and offer individualized learning content and plans, and recommend learning resources which are suitable for them [16, 17].

As a significant component of Traditional Chinese Medicine (TCM), acupuncture and moxibustion has profound theoretical heritage and practical requirements. In recent years, with the rapid development of artificial intelligence technology, its application in the teaching of acupuncture and moxibustion is increasingly extensive. Artificial intelligence records and evaluates students’ operations in real time. In addition, AI can not only provide accurate operational feedback and improvement suggestions through data analysis, which can help students correct errors in time and improve the accuracy and standardization of operations, but also simulate the real clinical environment [18] and allow students to practice acupuncture and moxibustion treatment in a virtual scene, which provides assistance for students in getting familiar with the clinical process in a safe environment and improving their clinical coping ability.

Moxibustion robot assisted teaching

The team offered a modular and multi-functional experimental teaching platform through self-developed and homemade Chinese intelligent moxibustion robot (Fig.1), explained the moxibustion robot related technologies and knowledge in teaching, and gave students deep impression through on-site physical demonstration.

The project decomposed the research of moxibustion robot into multiple sub-modules such as robot ontology, multi-functional physiotherapy module, intelligent interaction and control system, acupoint recognition system and multi-dimensional perception system (Fig.2), and designed fundamental experiment content for each module to consolidate students’ basic knowledge and improve students’ hands-on ability. With the intelligent physiotherapy robot of traditional Chinese medicine as the carrier, the new information technology, artificial intelligence, big data and the internet of things, would guide students to independently design innovative experimental content, cultivate students’ practical ability, which lead an achievement of “basic practice-comprehensive training-competition innovation” step by step. From the perspective of theoretical training, modular grading design, sub-project assessment, the brand new, whole process, multi-level and three-dimensional training system of the ability to project research and development and innovation was constructed and the theoretical connection between medical science and artificial intelligence development was established, which cultivated the innovative thinking of students.

Theoretical frame

The relationship between artificial intelligence and assisted teaching

The application of artificial intelligence in the field of education has achieved remarkable progress. Through the application of intelligent teaching [19], individualized learning [20], interactive teaching [21], data analysis [22], content recommendation [23], intelligent assessment [19] and autonomous learning [24], artificial intelligence has brought revolutionary changes and innovative development to traditional teaching methods. With the continuous development and improvement of AI, its application in education will be more extensive and in-depth, injecting new vitality into the progress of education.

How assisted teaching makes an impression on the three stages of teaching

By using before class, during class and after class, moxibustion robot plays an important role in teaching.

In the stage of preparation before class, the introduction of moxibustion robots has brought a lot of convenience to teachers and students. The latest information and materials related to moxibustion teaching are collected and sorted out automatically through the network to provide rich teaching materials for teachers. The intelligent teaching system offers students preview materials and online tests to help them understand the content in advance. By analyzing students’ learning data and historical performance, the robot can provide teachers with personalized teaching program design suggestions.

In the stage of participation in class, the application of moxibustion robot brings a more vivid and efficient teaching process. Interacting with students in real time through speech recognition and artificial intelligence, it can simulate the real moxibustion operating environment, allow students to practice in the classroom, and reduce the risk and cost of real operation.

In the stage of consolidation after class, moxibustion robots still leaves a significant impression. The robot is able to continuously track student progress and performance, providing teachers with detailed learning reports and data analysis. By offering rich teaching resources, realizing interactive teaching, providing real-time feedback and individualized learning support and other functions, moxibustion robot can not only improve the teaching effect and efficiency, but also promote the independent learning and all-round development of students.

The importance of the three stages of teaching for learning engagement and evaluation

In the teaching process, learning engagement and evaluation are two crucial links. Learning engagement is the degree of students’ involvement in teaching activities. Evaluation is an essential means to test students’ learning results and teaching effects. Learning engagement is the starting point of teaching activities and the basis for students to acquire knowledge, skills and values. Using moxibustion robots to design interesting and challenging content and activities can enhance learning motivation and promote knowledge construction.

Preparation before class can improve students’ sense of involvement in learning. By preview in time, students can understand the course content in advance and form a preliminary cognitive structure. Preparation before class helps teachers to assess students’ learning level more accurately, understand students’ knowledge reserve and learning difficulties before class, and develop more targeted teaching programs.

Participation in class is the core link in the whole learning process. Combining moxibustion robot with vivid and interesting explanation and interactive discussion, students can be more actively involved in learning. It can not only promote the learning effect of students, but also cultivate their independent learning ability and innovative spirit, which is an important basis for teachers to evaluate the learning effect of students.

Consolidation after class refers to the phase when students consolidate what they have learned and form long-term memory. By testing and feedback on students’ learning results, the learning situation can be acquired in time, providing basis for teachers to adjust teaching strategies and methods. The addition of moxibustion robot can provide timely feedback and guidance for teaching and promote teaching reflection and improvement.

Significance of study

First of all, via automatically determining the location of acupuncture points, smoke-free moxibustion environment and other functions, the moxibustion robot assisted teaching reduces the manual operation time and error in traditional teaching, improve teaching efficiency. Secondly, the moxibustion robot provides a more intelligent and convenient learning experience so that students can understand and learn moxibustion knowledge more intuitively, and enhance the interest and motivation of learning. Thirdly, the use of moxibustion robots brings a brand new stage where moxibustion teaching can no longer limited to specific teaching places and times, which provides new possibilities for distance education and autonomous learning.

As a new teaching tool, the application of moxibustion robot in teaching not only contribute to improving the teaching effect, but also promotes the cross-integration and development of medicine, education and other related disciplines. The systematic research and analysis of the practical application of moxibustion robot in teaching can give useful reference and inspiration to future teaching improvement and industrial development.

Aim of study

The study aimed to explore the factors influencing medical students’ engagement with moxibustion robot-assisted teaching and the mediating roles involved. Through multistage cluster sampling, this study collected and analyzed data from medical students who experienced the teaching of the moxibustion robot in an attempt to understand how the different stages of instruction, such as pre-class preparation, classroom participation, and post-class consolidation, affect learning engagement and ultimately the evaluation of the teaching process. The moxibustion robot was integrated into the teaching process to assist the instructor in delivering the content and facilitating learning. This study was conducted in this specific setting to evaluate the effectiveness and impact of the robot on student learning.

In the process of moxibustion robot assisted teaching (Fig.3), the research issues questionnaires to discuss the factors that affect the learning effect in the learning process, and further explore whether the combination of moxibustion robot with medical education and the students’ understanding level of medical theory and medical intelligent equipment technology can contribute to improving students’ practical operation ability and problem-solving capability.

Objects

The participants of the study were students in the TCM universities who had incorporated moxibustion robots into their teaching curriculum. They used the moxibustion robot as part of the teaching process in at least one class. To participant in the study, they had to be willing to complete questionnaires regarding their experience with the moxibustion robot, their level of commitment to learning, and their evaluation of the teaching process.

To minimize bias, multistage cluster sampling was employed to ensure that the sampling process was both random and systematic, enhancing the robustness and reliability of the study’s findings.

To obtain the sample size, we use the sample size calculation formula:

We set\(\:\:{\upalpha\:}=0.05\), p = 0.8, e = 0.05, the sample size n is 246.

Finally, a total of 250 questionnaires were collected from students who taught with moxibustion robots in a traditional Chinese medicine university, including 90 males (36.0%) and 160 females (64.0%).

Methods

The questionnaire focusing on the factors influencing the teaching effect of moxibustion robot in the teaching process set 5 latent variables including preparation before class (PBC), participation in class (PIC), consolidation after class (CAC), learning engagement, and evaluation. 5 observational variables (i.e. manifest variables) of each latent variable were determined, with a total of 25 items in the questionnaire. Every observable variable was measured on a 7-level Likert scale ranging from “1” (strongly disagree) to “7” (strongly agree). Specific model indicators and variable settings are shown in Table1.

Using SPSS 28.0 to test the reliability of the scale, the Cronbach’s α value of the total scale was 0.979, which is greater than 0.9, indicating that the reliability of the questionnaire results is high. The Cronbach’s α values for the five dimensions were all greater than 0.9, demonstrating that the reliability of the scale was also high. The Kaiser-Meyer-Olkin (KMO) value and Bartlett’s sphericity test were conducted on the questionnaire scale, with a KMO value of 0.957 and a significant result for Bartlett’s test of sphericity (x²= 8400.370, P < 0.001), indicating that the validity of the structural model was good.

Statistics

Python 3.10 (Netherlands)、R4.3.3 (New Zealand) and SPSS 28.0 (IBM, the United States) were used for descriptive statistics and correlation analysis. By measuring the path coefficients in different dimensions, the model fit was modified according to Bollen-Stine bootstrapping method. The Bootstrapping method was used to sample 5000 times repeatedly. The influencing factors and intermediary effects of moxibustion robot’s participation in the teaching process were calculated as well.

Descriptive statistics and correlation analysis

Descriptive statistics and correlation analysis were conducted on the five dimensions of preparation before class, participation in class, consolidation after class, learning engagement and evaluation. According to the results, string graphs of different dimensions were drawn (Fig.4) to show the relationship and correlation strength among the five dimensions. The color blocks represent different variables, with the total length of each block determined by the sum of the corresponding correlation coefficients. The width of the bar indicates the strength of the correlation between two variables; the wider the bar, the stronger the correlation. It was found that learning engagement was strongly correlated with the other four dimensions. At the same time, preparation before class, participation in class, consolidation after class, learning engagement and evaluation are positively and significantly correlated with each other. The results are shown in Table2 below.

Scatter plot matrix displayed a series of regularly arranged scatter plots for each dimension of multidimensional data, which can be used to show the relationship between two variables through a set of points in a two-dimensional coordinate system. The scatter plot matrix of the five dimensions of preparation before class, participation in class, consolidation after class, learning engagement and evaluation is drawn according to different genders, as shown in Fig.5. The diagonal line shows the score distribution of the five dimensions, and the non-diagonal line shows the score distribution of the different value from different genders in the five dimensions. It is found that the data has an asymmetric tail feature. In order to further explore the relationship, the study used structural equation model to comprehensively analyze the influence factors of moxibustion robot participating in the teaching process on students.

Fit and modification of structural equation model

To explore the relationships and pathways of the moxibustion robots across different teaching stages, this paper employs maximum likelihood estimation to fit the initial Structural Equation Model(SEM), as illustrated in Fig.6-A.

After simulating the initial model, the modification indices among the five latent variables (PIC, PBC, CAC, Learning Engagement and Evaluation) were found to be large. To improve the model’s reliability, six residual paths were added: [e2-e3], [e8-e9], [e9-e10], [e11-e12], [e18 -e19] and [e23-e24]. After correlation, the P-values for each path were less than 0.05, indicating statistical significance. The corrected model is shown in Fig.6-B.

The fit index of the model includes Chi-square test (CMIN), the ratio of Chi-square to degree of freedom (CMIN/df), goodness fit index (GFI), adjusted goodness fit index (AGFI), comparative fit index (CFI), and root mean square of average approximate error (RMSEA) [25]. Bollen-Stine bootstrapping method was used to modify the fit fit of the model. After modification, each fit index meets the standard. The results are shown in Table3 below. The fit of the model in this paper is standard-compliant [26].

Standardized path coefficients of structural equation model

Maximum Likelihood Estimation (MLE) is performed on the data. The path coefficients are shown in Table4 below. P < 0.05 means significant. The standardized path coefficient from PBC to PIC was 0.839 (P < 0.05). The standardized path coefficient from PIC to CAC was 0.945 (P < 0.05). The standardized path coefficients from PBC, PIC and CAC to learning engagement were 0.155 (P < 0.05), 0.072 (P > 0.05) and 0.701 (P < 0.05). The standardized path coefficients from learning engagement, PBC, PIC and CAC to evaluation were 0.749 (P < 0.05), 0.090 (P > 0.05), 0.095(P > 0.05), -0.032 (P > 0.05), respectively. The model’s standardized path coefficients are all displayed in Fig.6-B.

Mediating effect

In order to discuss the influence of moxibustion robot on the teaching process, Bootstrapping method was used to sample 5000 times repeatedly to measure the mediating role of learning engagement [25]. The results are shown in Table5. The mediating effect of learning engagement in the relationship between PBC and evaluation is significant(P < 0.001), as the 95% confidence interval ranged from 0.457 to 0.804, excluding 0. Learning engagement also played the mediating role in the effect of CAC on evaluation (P < 0.01), with a 95% confidence interval ranging from 0.226 to 1.133, excluding 0. However, the mediating effect that learning engagement in the relationship between PIC and evaluation is not significant(P > 0.05), as the 95% confidence interval ranged from − 0.065 to 0.895, including 0.

Learning engagement played a mediating role in the effect of preparation before class on evaluation, whose estimate was 0.631, which not only reflected the application value of modern science and technology in education and teaching, but also injected new vitality into the inheritance and development of traditional medicine. Moxibustion robot, as a combination of modern science and technology and traditional medicine, plays an increasingly important role in teaching [27]. In the pre-class preparation stage, the introduction of moxibustion robot significantly improved students’ learning engagement, and affected the overall evaluation of teaching effect [28]. With its vivid display and accurate operation demonstration, moxibustion robot can attract students’ attention in a short time, stimulate students’ curiosity and desire to explore. This novel teaching method not only enhances students’ learning motivation, but also promotes students’ understanding and mastery of moxibustion robot research and development [29, 30]. The moxibustion robot can provide personalized instruction tailored to each student’s specific needs. It achieves this by adjusting the intensity, frequency and acupuncture points of moxibustion to suit individual students. This personalized teaching method not only enhances the effectiveness of instruction but also makes students feel noticed and respected, increasing their willingness to engage in learning. The moxibustion robot is also capable of simulating professional moxibustion techniques, providing students with standardized teaching demonstrations.

The curriculum construction needs to change the teaching idea, reorganize the teaching content, and reform the teaching methods and means for the extensive use of artificial intelligence [31, 32]. The reform of teaching content needs to conform to the basic and frontier balance demand of the curriculum and the trend of interdisciplinary integration, and the innovation of teaching method needs to focus on solving the problem of limited class time and the “separation” of knowledge among various courses [18]. To cope with these challenges, this study introduces moxibustion robot into teaching as an assistant. Practice shows that with the assistance of moxibustion robots, students pay more attention to the content in the learning process, and actively participate in class discussion and practical operation, which deepening the in-depth understanding and skilled application of traditional Chinese moxibustion physiotherapy knowledge and traditional Chinese intelligent equipment technology. At the same time, the profound heritage and unique value of traditional Chinese medicine can be thoroughly appreciated. This kind of active learning state is not only conducive to students’ absorption and digestion of knowledge, but also helps to cultivate students’ innovative thinking and practical ability. Meanwhile, the classroom teaching is full of new vitality after the moxibustion robots take part. This teaching method not merely makes the classroom lively and interesting, but also greatly improves the teaching effect and education quality, and lays a solid foundation for students’ future academic exploration and practical operation [33].

Learning engagement played a mediating role in the effect of consolidation after class on evaluation, whose estimate is 0.525. The mediation effect indicates that moxibustion robot assisted teaching indirectly affects the overall evaluation of learning effect by affecting students’ learning engagement, which further proves the value and potential of moxibustion robot in education and provides new ideas and methods for teaching in university [34]. In the after-school consolidation stage, moxibustion robot, with its unique functions and advantages, offers an efficient and convenient way for students to review. The robot can also accurately demonstrate moxibustion professional techniques to help students independently review and consolidate the knowledge learned in class [35]. The moxibustion robot can also intelligently adjust the content and difficulty of teaching based on the students’ learning progress and feedback, ensuring that each student receives the personalized teaching that suits their needs. This intelligent teaching method helps optimize educational resources and improve teaching efficiency.

The collaboration between moxibustion robots and teachers promotes educational objects to engage in experiential learning, deeply integrate into educational situations, and learn to live harmoniously with others in practice [36]. Moxibustion robot has a positive effect on learning effect through the improvement of learning engagement. Students invest more time and energy in the consolidation after class, and have a deeper understanding and mastery of moxibustion technology. This kind of investment not merely helps to consolidate what they have learned from courses, but also promotes students’ practical application of TCM moxibustion physiotherapy knowledge and TCM intelligent equipment technology, and develops their innovative ability. The role of moxibustion robot in teaching is not only reflected in the classroom teaching stage, but also extended to the after-school consolidation stage. By influencing students’ learning engagement, the effect of consolidation after class and the overall evaluation of learning effect are effectually improved. The intermediary effect provides new ideas and methods for the teaching in university, which contributes to promoting the innovation and development of education [37].

The moxibustion robot offers significant advantages in TCM education. By accurately simulating the traditional moxibustion process, it provides students with an intuitive and vivid teaching demonstration, helping them gain a deeper understanding of the principles and operation techniques of moxibustion. Compared to traditional teaching methods, the moxibustion robot ensures the safety of the practice, effectively avoiding the risk of burns and other hazards. This allows students to practice repeatedly in a risk-free setting until they achieve mastery. Additionally, the moxibustion robot is programmable and can simulate a variety of moxibustion scenarios, catering to different levels and stages of teaching and learning [38]. This adaptability enhances students’ comprehensive abilities and practical skills. With the integration of modern technology through moxibustion robots, TCM education can preserve the essence of traditional practices while cultivating a new generation of professionals who are well-versed in both classical Chinese medicine and modern technological applications.

Strengths and limitations

This study conducts a comprehensive analysis of the impact of the moxibustion robot on student engagement across three key learning stages—preparation before class, participation during class, and consolidation after class—providing a holistic understanding of how AI tools can influence the learning process within a medical education context. This multi-stage approach is valuable as it addresses engagement and effectiveness across the entire learning cycle, rather than focusing on a single phase. Additionally, the use of SEM allows for an in-depth exploration of the complex relationships among various factors affecting student learning, engagement, and outcome evaluation. This statistical approach enhances the rigor of the study and enables a nuanced understanding of both direct and indirect effects, including the mediating role of learning engagement. While the study effectively examines student engagement and evaluation within a classroom setting, its findings may have limited applicability to broader areas, such as objective assessment for professional exams or standardized evaluations. Future studies could explore how moxibustion robots and similar AI tools may be integrated into these settings to objectively assess skill proficiency.

Author notes

1. Wei Lin, Lin Xu and Tao Yin contributed equally to this work.

Authors and Affiliations

1. School of Acupuncture and Tuina, Chengdu University of TCM, Chengdu, China

   Wei Lin,Tao Yin,Yang Chen,Jiaqi Chen&Fang Zeng

2. Key Laboratory of Acupuncture for Senile Disease, (Chengdu University of TCM), Ministry of Education, Chengdu, China

   Wei Lin,Tao Yin&Fang Zeng

3. School of Intelligent Medicine, Chengdu University of TCM, Chengdu, China

   Wei Lin,Lin Xu,Yujie Zhang&Xiabin Zhang

4. Shanghai Key Laboratory of Psychotic Disorders, Shanghai Mental Health Center, School of Medicine, Shanghai Jiao Tong University, Shanghai, China

   Binxin Huang

5. School of Psychology, Shanghai Jiao Tong University, Shanghai, China

   Binxin Huang

Contributions

WL, TY and LX drafted the manuscript. WL, TY and LX contributed equally to this work. JQC collected the data. YJZ, BXH, XBZ and YC revised the manuscript. FZ conceived and designed the study. All authors reviewed and approved the final manuscript as submitted and agreed to be responsible for all aspects of the work.

Corresponding author

Correspondence to Fang Zeng.

Ethics approval and consent to participate

All procedures performed in the study involving human participants were reviewed and approved by the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study was approved by Sichuan Regional Ethics Review Committee on Traditional Chinese Medicine (ID: 2021KL-059). The participants provided their written informed consent to participate in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions