Topology-aware deep learning model enhances EEG-based motor imagery decoding
Peer-Reviewed Publication
Updates every hour. Last Updated: 23-Nov-2025 21:11 ET (24-Nov-2025 02:11 GMT/UTC)
Motor imagery electroencephalography (MI-EEG) is crucial for brain-computer interfaces, serving as a valuable tool for motor function rehabilitation and fundamental neuroscience research. However, decoding MI-EEG signals is extremely challenging, and traditional methods overlook dependencies between spatiotemporal features and spectral-topological features. Now, researchers have developed a new topology-aware method that effectively captures the deep dependencies across different feature domains of EEG signals, ensuring accurate and robust decoding, paving the way for more brain-responsive technology.
Stroke is a major cause of non-traumatic disability worldwide. In a new IEEE/CAA Journal of Automatica Sinica breakthrough aimed at aiding patients with motor impairments, researchers demonstrated the development and control of CASIA-EXO upper-limb exoskeleton for motor learning in post-stroke rehabilitation. The proposed technology employs a novel control strategy to steadily individualize the training trajectory and intervention level as per the changing requirements and motor abilities of the patient.
Virtual reality (VR) use is often linked to reduced blinking, raising concerns about dry eye. This study examined the ocular surface during a 30-minute VR gaming session using a compact imaging system to observe the tear film in real time. Over the session, the tear film’s lipid-layer interference grade increased, indicating thickening of the lipid layer. Under these experimental conditions, this change may be associated with a lower likelihood of dry eye during VR use.
In recent years, significant advancements in educational technology, particularly the rise of online learning platforms, have transformed the way students engage with educational content. This study developed an intelligent digital human-based teaching system that leverages the capabilities of large language models to provide personalized, interactive, and real-time support.
A collaborative research team from institutions including China Agricultural University and Solomon Islands National University has addressed this question through in-depth analysis using a multi-country general equilibrium model. By leveraging trade data from 2001 to 2022 and combining it with a structural multi-country trade model, the team simulated the impact of China–Africa agricultural trade liberalization on the welfare and food security of both parties. The related paper has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025617).
Recently, a study led by Professor Qiran Zhao from the College of Economics and Management, China Agricultural University, for the first time incorporated physical fitness tests into the evaluation system. By analyzing the implementation effect of China’s Nutrition Improvement Program (NIP) for rural compulsory education students, the study provides a new perspective for optimizing SFPs worldwide, especially in African countries. The related article has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025611).
Recently, Professor Xu Tian from the College of Economics and Management at China Agricultural University, in collaboration with Mosses Lufuke from the Department of Economics at the University of Dodoma in Tanzania, uncovered the underlying reasons through data analysis. The related article has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025645).