Volume 16 Issue 3
Published on April 2025Artificial intelligence (AI) is profoundly transforming various aspects of the medical field, from disease diagnosis and drug development to personalized treatment and health management. Currently, the research and applications of AI in the medical field mainly focus on medical image analysis, disease diagnosis, drug discovery, genomics, and personalized treatment. This paper reviews the main applications of AI in the medical field, including medical image analysis, genomics, drug discovery, clinical decision support, and patient monitoring. It also discusses the challenges faced by AI in medical applications, such as data privacy, algorithm bias, ethical issues, and regulatory difficulties. Finally, it looks forward to the future development trends of AI in the medical field, including the integration of AI with biotechnology, the rise of explainable AI (XAI), and AI-driven precision medicine. Research shows that AI has been widely applied in the medical field with significant value and broad prospects, but ethical and data privacy issues need to be addressed.
View pdf
As demand for fiber-optic communication capacity grows, traditional multiplexing technologies struggle to keep pace, prompting the rise of Optical Space Division Multiplexing (OSDM). By utilizing the spatial dimension of fibers like multi-core and few-mode fibers, OSDM enables parallel data transmission across independent channels. This paper explores its principles and applications in high-capacity networks, mobile backhaul, and microwave photonics. OSDM offers significant advantages, including enhanced transmission capacity and improved energy efficiency over conventional methods like wavelength and time division multiplexing. However, it faces challenges such as high manufacturing costs and complex crosstalk management. Despite these drawbacks, OSDM’s scalability and potential for integration with intelligent systems position it as a key technology for future optical communication networks.
View pdf
As intelligent driving technology matures and gains market acceptance, its application has become widespread across China. This paper delves into the developmental journey of intelligent driving within the country, while also examining its inherent advantages and disadvantages. Currently, the predominant approach in China is the use of multi-sensor fusion solutions, which enhance sensing capabilities by combining information from various sources. However, this method is not without its challenges, including high implementation costs, limitations in processing complex environments, and vulnerabilities within the chip supply chain. The Chinese government’s endorsement of smart driving car brands and technologies is escalating, reflecting a strong commitment to the sector. Public sentiment towards smart driving remains favorable, indicating a receptive market. Looking ahead, for China to maintain its momentum in the intelligent driving arena, it is crucial to pursue breakthroughs and innovations. Furthermore, fostering the integration of key industries with the driving sector will be essential to unlock new potentials and sustain competitive advantage in the global market.
View pdf
High-frequency noise, often caused by system jitter and environmental factors, can obscure the true motion of particles. This study presents a Fourier transform-based particle velocity optimization framework designed to improve the accuracy of velocity estimation in particle tracking experiments. High-frequency noise, often caused by system jitter and environmental factors, can obscure the true motion of particles. To address this, we propose an adaptive low-pass filtering approach where the cutoff frequency is optimized through a numerical search algorithm to minimize the error between the filtered velocity and the ground truth trajectory. Our results demonstrate that an optimal cutoff frequency of approximately 1 Hz offers the best balance between noise reduction and signal preservation. The framework is further enhanced by its adaptability to different experimental conditions, making it applicable to a wide range of particle tracking scenarios. This approach offers a more effective solution for overcoming noise-related challenges in particle tracking, providing a valuable tool for precise motion analysis in various scientific fields.
View pdf
In order to improve the degree of automation and efficiency of material sorting, virtual simulation technology is explored for verification in the design stage. This paper takes a robot for visual recognition and grasping a simple material model as an example to build a three-dimensional workstation model and integrate the robot's motion planning strategy. In the development environment of ABB's offline programming and simulation software Robotstudio, 3D modeling of related equipment was first carried out, and the spatial layout of workstation was completed according to the task requirements. Secondly, it is necessary to complete the selection of conveyor belts, the design of dynamic Smart components, the logical connection of workstations, and the loading and unloading path planning of mechanical arms. Finally, simulation verification is carried out. The simulation workstation realizes real-time collaborative optimization between the vision system and the robot controller, improving the flexibility and accuracy of the sorting process. The workstation helps to improve the sorting efficiency, and provides a certain reference for the follow-up research of industrial robot visual material sorting design.
View pdf
Sodium-ion batteries (SIBs) have gained increasing attention due to their low production cost, abundant raw materials, and relatively high energy density. In addition, SIBs exhibit a range of desirable characteristics, including high specific capacity, good high-temperature performance, safety, and environmental friendliness. Therefore, research into sodium-ion batteries is of paramount importance. This paper references a large number of studies on sodium-ion batteries, aiming to analyze and summarize the research issues related to SIBs and the impact of their development on societal progress. The paper primarily focuses on solid-state electrolytes, while also covering analysis of sodium-sulfur batteries, zebra batteries, sodium-air batteries, and aqueous sodium-ion batteries.
View pdf
In the era of rapid digital transformation, social networks generate huge amounts of textual data every day, making sentiment analysis an essential tool for understanding public opinion. This study focuses on the application of probabilistic and statistical methods to sentiment analysis in social networks, highlighting their effectiveness in dealing with uncertainty and modeling the distribution of emotions. The main objective is to evaluate the role of Naïve Bayesian (NB), Hidden Markov models (HMMs), and Bayesian networks in emotion classification, emotion propagation, and dynamic emotion tracking. Through literature review and comparative analysis, this study examines the existing research, computational efficiency, and real-world applications of probabilistic classification models. The results show that Naive Bayes is computationally efficient and effective for large-scale emotion classification, while HMM and Bayesian networks excel in sequential emotion prediction and user behavior modeling. The study highlights the advantages of probabilistic methods in sentiment analysis, while acknowledging their limitations, such as their reliance on probabilistic assumptions and the challenges of capturing deep contextual semantics. Future research should explore hybrid approaches that combine probabilistic models with deep learning techniques to improve the predictive performance and scalability of real-time sentiment analysis.
View pdf
In comparison to internationally renowned liquors, China’s fen-flavor liquor is characterized by its milder taste and healthier properties. However, with the advancement of intelligent brewing technologies, the quality of foreign liquors has significantly improved, increasingly challenging the market position of Chinese liquors. This study begins with an overview of the liquor brewing process, emphasizing the critical role of temperature control in fermentation. Subsequently, we propose an innovative real-time temperature monitoring system specifically designed for liquor production. The system incorporates a precisely engineered temperature sensor, complemented by a robust data transmission framework and an efficient storage module for real-time temperature visualization. To validate the system’s effectiveness, field tests were conducted at a distillery in Xinghua Village, focusing on real-time temperature measurement of brewing materials. The experimental results demonstrate that our system successfully achieves real-time monitoring of temperature variations during both the cylinder fermentation and material processing stages of liquor production. Compared with traditional methods relying on manual sensory evaluation and empirical judgment, this system enables more precise control over raw material proportions, thereby enhancing both the flavor profile and production yield of fen-flavor liquor. This technological advancement represents a significant step toward the intelligent transformation of traditional Chinese liquor-making techniques.
View pdf
Crowdfunding is a concept that emerged due to difficulties in raising funds for community business projects, social activities, small and micro enterprises, and startups. However, the success or failure of crowdfunding projects is often full of uncertainty. Therefore, predicting whether a crowdfunding project can succeed has become a question worthy of in-depth research. This article takes the crowdfunding website platform Kickstarter as an example and uses machine learning methods to construct an effective crowdfunding project status prediction model. This study aims to compare the performance of different machine learning algorithms in predicting crowdfunding results, while identifying the main factors that affect crowdfunding results and their relative importance. The dataset used in this study includes data from all projects published on Kickstarter between January 2017 and January 2018. This study used six classic classification algorithms to predict the status of crowdfunding activities, and calculated the accuracy of each classification model. The results showed that the accuracy of all six models was close to 1, indicating that they could effectively predict the success or failure of crowdfunding projects. However, the Gaussian Naive Bayes model had slightly lower accuracy than the other five models. Furthermore, the research results indicate that in successful crowdfunding projects, factors such as crowdfunding goals, funds raised, and the number of supporters are more crucial influencing indicators than others.
View pdf
Enterprise alliances and government agencies typically deploy distributed systems across subsidiaries and subordinate departments, where each system node collaborates via a network to present a unified interface to users. Addressing the challenges of log data dispersion, heterogeneity, and lack of credibility in distributed systems, this paper proposes a log anomaly detection framework based on a graph-chain architecture. The framework leverages the sequence analysis capabilities of a distilled Transformer model to detect anomalies in system logs at each node. Finally, by integrating blockchain smart contracts, it ensures tamper resistance and traceability. Experimental results demonstrate that the proposed framework achieves an anomaly detection accuracy of 99.6%, surpassing traditional methods.
View pdf