Falling is a common phenomenon in the life of the elderly, and it is also one of the 10 main causes of serious health injuries and death of the elderly. In order to prevent falling of the elderly, a real-time fall prediction system is installed on the wearable intelligent device, which can timely trigger the alarm and reduce the accidental injury caused by falls. At present, most algorithms based on single-sensor data cannot accurately describe the fall state, while the fall detection algorithm based on multisensor data integration can improve the sensitivity and specificity of prediction. In this study, we design a fall detection system based on multisensor data fusion and analyze the four stages of falls using the data of 100 volunteers simulating falls and daily activities. In this paper, data fusion method is used to extract three characteristic parameters representing human body acceleration and posture change, and the effectiveness of the multisensor data fusion algorithm is verified. The sensitivity is 96.67%, and the specificity is 97%. It is found that the recognition rate is the highest when the training set contains the largest number of samples in the training set. Therefore, after training the model based on a large amount of effective data, its recognition ability can be improved, and the prevention of fall possibility will gradually increase. In order to compare the applicability of random forest and support vector machine (SVM) in the development of wearable intelligent devices, two fall posture recognition models were established, respectively, and the training time and recognition time of the models are compared. The results show that SVM is more suitable for the development of wearable intelligent devices.

With the continuous improvement of human living standards, population aging has become a global development trend. At present, China has entered an aging society, the health and safety of the elderly have become the focus of social concern. Due to the aging of physiological structure and the decline of physical function, the probability and frequency of accidental falls in the elderly are very high. Under the above background, the purpose of this study is based on a heterogeneous sensor data fusion algorithm in an intelligent wearable sensor network. This article proposes a heterogeneous sensor data fusion algorithm based on wearable wireless body area network technology, and constructs a high-precision and stable wearable elderly activities of daily living (ADLs) and fall monitoring system. We first select the three-axis acceleration sensor, three-axis magnetic sensor, and three-axis angular velocity sensor to monitor the activities of the elderly. Then, we use Bluetooth to transmit the data collected by heterogeneous sensors to smartphones, and communicate with service centers and users through the mobile phone communication network, Family members interact to form a wireless city network based on wearable technology. Our proposed data fusion approach is based on the Kalman filter algorithm, which can reduce the system noise and improve the stability of the system. The experimental results demonstrate that the fall detection system proposed and implemented in this study can well detect accidental falls in the daily activities of the elderly, the sensitivity and specificity of the fall detection system are 98.7% and 98.5% respectively. The study in this article has a high research value and practical application significance in protecting the healthy life of the elderly.

The livelihood problem, especially the medical wisdom, has played an important role during the process of the building of smart cities. For the medical wisdom, the fall detection has attracted the considerable attention from the global researchers and medical institutions. It is very difficult for the traditional fall detection strategies to realize the intelligent detection with the following three reasons: (i) the data collection cannot reach the real-time level; (ii) the adopted detection methods cannot satisfy the enough stability; and (iii) the computation overhead of collection device is very high, which causes the barely satisfactory detection effect. Therefore, this paper proposes Convolutional Neural Network (CNN)-based fall detection strategy with edge computing consideration, where the global network view ability of Software-Defined Networking (SDN) is used to collect the generated data from smartphone. Meanwhile, on one hand, the edge computing is exploited to put some computation tasks at the edge server by the scheduling technique. On the other hand, CNN is equipped with both edge server and smartphone, and it is leveraged to train the related data and further give the guidance of fall detection. The experimental results show that the novel fall detection strategy has a more accurate rate, transmission delay, and stability than two cutting-edge strategies.

Micro-expressions, fleeting facial movements lasting 1/25 to 1/3 of a second, offer crucial insights into genuine emotions, particularly valuable in online education settings. The rapid growth of English learning livestreams has heightened the need for accurate, real-time micro-expression recognition to enhance learner engagement and instructional effectiveness. However, existing methods need help with the subtle nature of these expressions, especially in dynamic, low-resolution streaming environments. This paper presents TSG–MER–ELL, a novel end-to-end network for micro-expression recognition in English learning livestreams, integrating temporal–spatial correlation and graph attention mechanisms. The framework addresses the unique challenges of real-time emotion analysis in online language education, where subtle facial cues are crucial in understanding learner engagement and comprehension. The temporal–spatial correlation module employs action units with spatio-temporal graph convolution to aggregate features from diverse facial regions, while transformer encoders construct long-range correlations. The graph attention module builds upon local facial areas to guide self-attention computations, yielding precise local correlation features. These global and local features are fused for the final micro-expression classification. We introduce an adaptive loss function that balances accuracy, efficiency, and relevance to linguistic context. Extensive experiments on SMIC, CASME II, and SAMM datasets, adapted for English learning scenarios, demonstrate TSG–MER–ELL’s superior performance over ten state-of-the-art baselines. The TSG–MER–ELL framework achieves top UF1 and UAR scores across all datasets, significantly improving recognition speed and accuracy. Ablation studies and visualizations of temporal–spatial features and graph attention weights provide insights into the framework’s effectiveness in capturing subtle emotional cues. TSG–MER–ELL’s robust performance in varied online learning conditions highlights its potential to enhance engagement, personalize instruction, and improve overall outcomes in virtual English language education.

With the increasing awareness of consumer health and the arrival of the big data era, how to efficiently extract valuable knowledge from massive health text information has become an urgent problem. Large Language Models (LLMs) provide a new approach for intelligent processing of consumer health information with its powerful natural language processing. This article proposes a method for LLMs-driven health text information analysis in consumer electronics. It demonstrates how LLMs effectively analyze consumer health text information and improve the accuracy of health information generation. Based on LLMs, the proposed method first establishes a health text information component recognition model (HTICRM), which aims to identify the location and type of key information input into consumer health text. Then when there are diverse positions of text information components that cannot be accurately analyzed, this article uses LLMs reflection and P-Tuning v2 to generate health text information components to narrow the feature gap between the generated data and general consumer health domain data. Finally, the experimental results verify that the proposed method can help to promote the further application and development of LLMs in the field of consumer health, providing consumers with more scientific and accurate health guidance.