Ebook: Artificial Intelligence, Medical Engineering and Education

The 8th International Conference on Artificial Intelligence, Medical Engineering, and Education (AIMEE2024) was held in Huangshi, China on 26 and 27 October 2024. This conference was jointly organized by Wuhan University of Technology, Hubei University of Technology, Wuhan Textile University, Wuhan Technology and Business University, the Polish Operational and Systems Society, the National Technical University of Ukraine, the International Center of Informatics and Computer Science, and the International Research Association of Modern Education and Computer Science.

This collection of papers is divided into three sections: Artificial Intelligence and Scientific Methodology; Systems Engineering and Analysis: Concepts, Methods, and Applications; and Education Reform and Innovation. A total of 227 submissions were received by AIMEE2024, with 80 papers selected for presentation and publication after a rigorous international peer-review process. These papers propose new concepts and methodologies that may pave the way for researchers across different disciplines to conduct further research.

We hope this volume will contribute to the rapid evolution of artificial intelligence and its applications in various fields by facilitating collaboration and knowledge sharing among the growing interdisciplinary workforce in this exciting field. We extend our gratitude to all those who helped to make this conference a success. We thank our keynote speakers for sharing their invaluable insights and setting the stage for thought-provoking discussions. We are grateful to the program committee members for their diligent efforts in reviewing submissions and assembling an outstanding technical program, and the same gratitude goes to all authors for contributing their innovative research results, and to the attendees for joining us and making the conference a vibrant forum for learning and communication.

Finally, our sincere appreciation goes to the publisher, IOS Press, for preparing and publishing this volume.

Conference Chairs

Prof. Z.B. Hu, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Ukraine

Prof. Qingying Zhang, Wuhan University of Technology, China

Prof. Matthew He, Nova Southeastern University, Fort Lauderdale, USA

Prof. Felix Yanovsky, Delft University of Technology, Netherlands

October 2024

The paper presents the components of developing secure and efficient information systems focusing on data protection technologies and architectural design using medical information systems. By evaluating architectural approaches, the analysis highlights the advantages of microservices architecture for modern medical systems due to its flexibility, modularity, and ability to integrate various technologies while minimizing security risks. Research explores the integration of cloud solutions with on-premises storage, offering practical insights for optimizing security and scalability in medical systems.

Crawler technology is a kind of technology that can automatically collect data from a large amount of web pages and data resources (such as videos, pictures, texts, etc.) on the Internet without manual intervention, and according to the automatic program or script technology, the relevant information scattered in different websites is summarized together for real-time monitoring. It can get information such as policy changes, demand changes, price fluctuations and other content changes in a short time. At present, in response to the changes in the domestic and foreign economic situation, China is promoting the construction of free trade pilot zones, the construction of the Belt and Road as the fulcrum, reshaping the pattern of opening up. This paper takes Jiangsu Province as a case to discuss the dynamic correlation between external opening up and regional economic growth, which can effectively help the government to have a real-time, accurate and comprehensive understanding of the dynamic correlation between external opening up and regional economic growth, hoping to provide a strong basis for the government to make relevant enterprise strategies and government decisions, and promote the prosperity and development of regional economy.

Classification experience, as an important way of product information recommendation in AI recommendation system, will have an important impact on consumers’ perceived behavior. Although there is a lot of research on intelligent recommendation, fewer existing studies have explored the impact of classification experience under AI recommendation on their willingness to continue to use it from the perspective of user experience. In this paper, based on the social exchange theory, we use the experimental method to collect data in three different AI recommendation scenarios (songs, novels, and shoes) to validate the mechanism of the influence of categorization experience on users’ willingness to continue to use the AI recommendation system, and reveal the mediating role of the users’ sense of being understood and the moderating role of the uniqueness demand. This study contributes to a better understanding of the different behavioral responses of users to different perceptions of AI recommendation results, and expands the use of AI in marketing.

In the realm of cross-modal Image-Text Retrieval (ITR), this paper focuses on the effectiveness in the Chinese language environment. Based on Chinese CLIP, the model pretrains on a large-scale Chinese image-text dataset and employs carefully selected vision and text encoders with a two-stage pretraining strategy, which enables the model to develop a nuanced semantic alignment between images and texts within a Chinese context. This approach facilitates a deep understanding of the matching between images and Chinese texts. To further enhance the model’s capabilities, the paper introduces domain-specific dataset to implement fine-tuning strategies. Well-designed experiments are conducted between the popular models and ours on public dataset Flickr30K-CN, COCO-CN, and domain-specific dataset AP-ID. Our model attains state-of-the-art (SOTA) results on Chinese text-image retrieval datasets, demonstrating its robustness and effectiveness in handling Chinese-language cross-modal ITR task.

With the rapid development of living standards and the leap up of national economy, the logistics industry has ushered in a vigorous development, which has played a great guiding role in promoting consumption and driving economic growth. To further expand their business and meet the different needs of customers, more and more enterprises have established many logistics distribution centers and applied a variety of car models to deliver different types of goods. Taking JD Logistics Company in Hubei Wuhan Province as an example, this paper has studied the optimization problem of multi-vehicle routing allocation, and gives corresponding solutions according to the actual situation of JD logistics in China based on scanning method and assignment model. The loading rate has been greatly improved and stabilized from 84.3% to about 95.5% based on the scanning method. The transportation time can reach to the minimum level based on the assignment model.

Digital-intelligence convergence promotes the construction of a new paradigm of intelligent education and is the core driving force for the construction of the New Liberal Arts lab. Driven by the digital-intelligence convergence technology, the New Liberal Arts lab presents new morphological characteristics, and its architectural design mainly includes five parts: a digital-intelligence convergence experiment platform based on 5G and cloud architecture; a multidisciplinary experimental big data center; an intelligent computing center providing arithmetic support, algorithmic models and intelligent control; a virtual simulation experiment center embodying panoramic rendering, holographic interaction, and virtual-reality coexistence; an experimental data sensing and monitoring system covering all nodes, objects, and processes; and an experimental intelligence system meeting the requirements of accurate evaluation and personalized learning. This architecture provides a reference for the construction of New Liberal Arts labs in colleges and universities at the technical level.

As a core element of Industry 4.0, intelligent manufacturing (IM) has influenced the direction of talent cultivation, and IM practice teaching has also become construction emphasis in various universities in recent years. The Center for Practice Innovations of Huazhong University of Science and Technology (HUST) relying on its strong mechanical discipline background and school enterprise advantages, has taken the lead in China in forming the IM practice education system guided by intelligence and driven by innovation. It has comprehensively summarized the measures and experiences in the construction of IM practical teaching platforms, practical teaching courses, and practical teaching modes. Through nearly five years of continuous improvement and practice, the Center for Practice Innovations has achieved the IM practice teaching covering various majors and four years of university, and has become a benchmark for the construction of IM practical platforms and practical courses in various universities.

The increase in computing power and the development of low-power computing, quantum computing, and the Internet of Things require lossless and reliable approaches to information processing. Based on previous results, this work aims to develop a new fault-tolerant reversible encryption device with an optimal number of components, delay time and auxiliary ports. Empirical analysis of known reversible reconfigurable encryptors and alternative reversible basis was used. The work results presented two new reversible encryptors based on the reconfigured extended basis of Fredkin gates. Comparisons with known reversible encryptors showed at least a 39% improvement in quantum cost and a reduction in circuit depth of more than 52%. Notably, the quantum cost was reduced from 79 to 19 (to 48 in case of variant with preservation of parity) due to the improvement of the circuit design. The circuit depth was reduced from 9 to 5 (to 6 in case of circuit with preservation of parity). One of the synthesised encryptors preserves parity, improving the fault tolerance of the device. The simulation of the encoder models was carried out in the Active-HDL environment, and the quantum parameters of the devices were determined using the IBM Quantum Lab. The outcomes of the study produced promising reconfigurable reversible circuits for data processing.

The measurement of the focal length of a bi-convex lens using auto-collimation imaging is a fundamental experiment in college. However, some teachers and students are confused by unwanted reflection images unaffected by the plane mirror when tracking the accurate image reflected by that mirror. By delving into the formulas of the reflection imaging of thin lenses through two different reflection imaging methods and introducing the notion of focal power, this paper obtained the inequalities determining the formation of the reflection images. These inequalities can explain the existence of the reflection images, analyze the general cases of reflection imaging using a thin lens, and do a quantitative calculation of a thin lens with specific parameters. This work can help teachers and students understand the reflection imaging of thin lenses, and it can be designed as an experimental procedure to foster greater scientific interest among students in this subject by combining different teaching methods.

This study explores the impact of six fundamental visual features—hue, lightness, saturation, contrast, clarity, and color complexity—on the emotional evaluation of images in terms of valence and arousal, using a digitalized affective picture system. Two sub-studies were conducted: Study 1 analyzed the differences in emotional evaluations among image sets with identical content but varying visual features, while Study 2 examined how changes in individual visual features influence emotional perception under controlled conditions. The results show that, although specific visual features significantly affect emotional evaluations with varying degrees of impact, the overall process of image-based emotion evaluation remains systemic, with interactions between visual features jointly shaping emotional responses. This research clarifies the underlying relationship between visual features and emotion perception, providing new empirical support and practical implications for advancing affective computing and image-based emotional assessment systems.

This study focuses on evaluating the effectiveness of AI-assisted AI-based instructional practices for teaching Python programming to business majors. The study analyzed the distribution of students’ final grades by comparing the class that used the traditional teaching model in the 1st semester of the 2022–2023 academic year with the class that introduced the AI-assisted teaching model in the 1st semester of the 2023–2024 academic year. The results showed that the classes with the AI-assisted teaching model showed a significant improvement in the students’ performance, with a significant increase in the percentage of high-performing bands and a significant decrease in the percentage of low-performing bands. Further, an independent samples t-test was conducted on the grades of the two classes using SPSS 26 software, and it was found that there was a significant difference between the two in terms of average grades, and this difference was highly statistically significant. This study demonstrates that the AI-assisted teaching model based on artificial intelligence has significant advantages in improving the teaching quality and student achievement in Python Programming course for business majors.

Since the COVID-19 virus epidemic, the world, the governments, and citizens of various countries and regions the importance of medical care has been enhanced to varying degrees, for the demand for medicines has also risen sharply. Traditional pharmaceutical logistics, in both the mode of operation and the level of technology needed to face the surge in demand, are quite stretched. So in this case, the traditional pharmaceutical logistics urgently need a big change, which also inspired many institutions and scholars in the “modern pharmaceutical logistics” this plate for further research. Given this phenomenon the author for nearly a decade of domestic and international pharmaceutical logistics research history summarized and refined, the use of citespace tools for its visualization. Through the analysis of the current research status to recognize the current problems and shortcomings of pharmaceutical logistics, propose targeted improvement measures, as well as to identify possible research trends in the next few years, for scholars in related fields to carry out further research.

Smart parking has become an important construction field of smart cities in China and an application field closely related to urban production and life. Establishing an urban smart parking evaluation index system is the top-level design of smart parking construction. Guided by the characteristics of new quality productivity, namely high technology, high efficiency, high quality, and sustainability, this paper constructs an urban smart parking evaluation framework of “one core, two goals, three levels, and six dimensions” on the basis of elaborating the concept and new connotations of smart parking, and extracts the key indicators for smart parking evaluation from six aspects: technological innovation, efficient governance, data-driven, high-quality service, sustainable industry, and people’s satisfaction. Finally, the calculation model and recommended weight value of the indicators are given, and the concept and calculation method of the “smart parking development index” are proposed. The research can provide evaluation methods and action guidelines for the construction of urban-level smart parking systems.

The concept of smart ports represents a novel trend within the port industry, characterised by the integration of advanced technologies such as 5G, IoT and AI, which collectively facilitate the realisation of intelligent facilities and efficient operations. Shanghai Port plays a pivotal role in China’s global engagement and integration into the international economic system. It is also among the first ports in China to have introduced and constructed smart ports. This paper takes the aforementioned seaport as a case study and employs the SBM-DEA model, combined with the Malmquist index, to evaluate the logistics efficiency of Shanghai Port and other major Chinese ports over the past ten years, from 2014 to 2023. The evaluation is conducted from three dimensions and is followed by an analysis and recommendations.

With the prevalence of AI, AI literacy has become a critical capability for citizens in the new era. It necessitates that practitioners not only master AI technologies but also possess skills in critical assessment and effective application. This article explores the fundamental theoretical framework of AI literacy, including its definition, essence, and connotation, emphasizing the importance of a deep understanding of AI technology, critical evaluation, effective application, and ethical considerations. In the context of intelligent manufacturing, a content framework for AI literacy education is proposed, covering four aspects: AI cognition, technical practice, ethics and social responsibility, and interdisciplinary thinking. In mechanical automation education, reform strategies for integrating AI literacy are suggested, including course content integration, innovative teaching methods, enhancement of practical components, optimization of assessment systems, development of faculty teams, and multidisciplinary collaboration among multiple stakeholders. The aim is to cultivate future professionals in the field of intelligent manufacturing who possess a solid theoretical foundation, practical skills, innovative thinking, and ethical awareness.

This article addresses the issues of insufficient automation, high reliance on manual labor, and data silos in the traditional stirrup production process. It employs various information technologies such as building information modeling (BIM), enterprise resource planning (ERP), manufacturing execution systems (MES), cyber physical systems (CPS), central control systems (CCS), and warehouse management systems (WMS) to establish a solution for intelligent stirrup production. Based on this foundation, a flexible robotic grasping fixture is designed, integrating automated equipment including material handling robots, stacking robots, automatic packaging machines, and AGV forklifts. This configuration creates a highly automated, digitized, and intelligent stirrup workstation, facilitating the automation and flexibility of stirrup production. This solution eliminates data silos and achieves comprehensive data integration from BIM design to production execution, allowing for seamless coordination between online scheduling and offline automated execution in the stirrup production process. Through these measures, stirrup production has attained high standards of intelligent manufacturing, driving the factory’s transformation towards digitization and intelligence.

To solve the problem that the traditional retail market is limited to single-dimensional data analysis of sales when predicting commodity sales, which ignores the long-term trend, seasonality and holiday within the sales data, making it difficult to fully capture the data characteristics and accurately predict sales. This paper constructs a Prophet-LightGBM combined machine learning model. Firstly, the Prophet model is used to automatically decompose data characteristics and flexibly adjust their impact on sales. Then, the LightGBM model is used to build an efficient prediction model for the multi-dimensional features identified by Prophet to improve the reliability of sales forecasts. Experiments have shown that the prediction R² of the combined model has reached 0.85, which is better than the prediction effect of a single model. It is proved that the proposed combined model has better performance in commodity sales forecasting, and provides strong theoretical data support for enterprises to accurately grasp market demand, optimize production planning and inventory management.

The article analyses the impact of digital technologies, particularly artificial intelligence (AI), on consumer behaviour in the financial services sector, with a focus on improving financial literacy and decision-making through personal financial trackers. The study explores how AI-based applications, such as adaptive learning and gamification, can enhance users’ financial literacy. Key outcomes include a significant improvement in users’ ability to manage personal finances, with AI tools offering personalized recommendations, automatic expense categorization, and forecasting. Additionally, the study identifies challenges related to data security, ethical concerns, and algorithmic transparency. The research concludes that while AI-powered financial tools can greatly enhance financial literacy, their effectiveness depends on user engagement and the resolution of ethical and regulatory challenges.

Optical flow estimation is a crucial task in computer vision, aiming to infer motion information in scenes by analyzing pixel movements in video sequences. Traditional methods rely on photometric consistency assumptions and motion smoothness constraints but often struggle with large motions, complex scenes, or occlusions. Deep learning-based approaches have significantly advanced the field, with notable contributions from FlowNet, FlowNet2.0, PWC-Net, and RAFT, enhancing accuracy and robustness. However, challenges remain, including robustness in complex environments and accurate estimation of large displacements. To address these, we propose MSI-Net, a deep optical flow estimation network that integrates multi-scale feature extraction, deformable convolution, and a layer-by-layer refinement strategy. MSI-Net enhances robustness and accuracy, especially when handling large motions and occlusions. Experimental results on synthetic (Sintel) and real-world (KITTI) datasets demonstrate the effectiveness of our approach, which achieves comparable or superior accuracy to state-of-the-art methods while maintaining high efficiency.

Since its establishment in 1993, SF Express has grown alongside the domestic express delivery industry and private economy. The rise of online and video shopping has further fueled its rapid growth. SF Express’s achievements can be attributed to effective cost control, yet improvements in frontline operations are needed. This article first analyzes SF Express’s current status in the industry and then explores alternative plans for its Hubei Province distribution center using the Analytic Hierarchy Process (AHP). Leveraging delivery volume data from Hubei outlets, a vehicle routing model is constructed and solved using the Saving Algorithm, yielding an optimized plan for SF Express’s logistics in Hubei. This study offers practical optimization suggestions for SF Express and provides valuable reference for third-party logistics enterprises.

Knowledge discovery and data mining is a complex, multifaceted task that requires the integration of various approaches, including machine learning, statistics, and data processing algorithms. The main challenge is the need to extract useful information from large, often unstructured and noisy data, and to interpret the results to make effective decisions. The proposed method is based on the morphological approach framework, which involves the use of morphological analysis to extract useful information from structured Big Data. The morphological approach focuses on analyzing and interpreting morphological structures of systems to extract the required information. The considered approach is based on systems theory, set theory and cluster analysis. A similarity measure is introduced to evaluate the correct partitioning of a morphological set. The use of big data identifies patterns, trends and relationships between attributes of systems. Through the use of Big Data, morphological analysis can be more accurate and efficient, advancing fields such as knowledge discovery and data mining. Solving these problems opens up great opportunities for the use of data in all areas of human endeavor. The proposed approach has been used in applied engineering fields such as Aerospace, IT, technological innovation.

Mathematical modeling can optimize dietary plans, but relevant research is scarce. This paper adopts the Analytic Hierarchy Process (AHP) and single-objective optimization model in mathematical modeling. Based on the “Chinese Food Composition Table”, an evaluation model is constructed, and This paper is presented from the perspective of food structure, energy, meal ratio, nutrients, and protein amino acids. Through Matlab programming, the diet is optimized to maximize the protein amino acid score and reduce costs. The results show that single optimization is prone to imbalance or high costs, while comprehensive optimization achieves both balanced nutrition and economy. This scheme provides a scientific basis for dietary optimization and promotes the development of healthy diets.

When passing through a toll station, vehicles need to slow down or stop to pay. Therefore, vehicles on the toll station section are in a state of “stopping and starting” and waiting in line, which can easily lead to traffic congestion and driving safety issues. Vehicles will lose kinetic energy due to deceleration, referred to as energy consumption. Based on the cellular automaton NaSch traffic flow model, a vehicle energy consumption model for toll station sections under periodic boundary conditions is established and simulation analysis is performed to study the impact of the random deceleration probability of vehicles on vehicle energy consumption. The results of research show that the energy consumption of vehicles on the road decreases with the increase of random deceleration probability, and the traffic flow and average speed of vehicles also become smaller, and the earlier the traffic jam occurs.