Lateral walking gait recognition and hip angle prediction using a dual-task learning framework
Peer-Reviewed Publication
Updates every hour. Last Updated: 18-Jul-2025 14:11 ET (18-Jul-2025 18:11 GMT/UTC)
A research paper by scientists at Chinese Academy of Sciences proposed a dual-task learning framework, the “Twin Brother” model, which fuses convolutional neural network (CNN), long short-term memory (LSTM), neural networks (NNs), and the squeezing-elicited attention mechanism to classify the lateral gait stage and estimate the hip angle from electromyography (EMG) signals.
The new research paper, published on May. 1 in the journal Cyborg and Bionic Systems, provide a “Twin Brother” model. The model is a dual-task learning framework designed for simultaneous gait phases recognition for lateral walking and continuous hip angle prediction.
Addressing the urgent need for sustainable CO2 conversion, researchers at Tongji University developed a novel copper-based metal-organic framework (MOF) catalyst, TJE-ttfp, which achieves 99.2% Faradaic efficiency for C1 liquid fuels (formic acid and methanol) at a remarkably reduction potential of −0.1 V. By leveraging dynamic Cu(I)/Cu(II) interconversion and electron-rich ligands, the material suppresses competing hydrogen evolution while enhancing CO₂ activation, offering a breakthrough for energy-efficient carbon utilization.
Teaching STEM in Hong Kong’s preschools presents cultural, personal, and organizational challenges for female, ethnically diverse teachers. A recent study from The Education University of Hong Kong explores these barriers through the Concerns-Based Adoption Model, identifying five key concerns affecting STEM adoption. By examining confidence issues, traditional teaching philosophies, and limited resources, this study provides insights into the complex factors shaping early STEM education and calls for tailored support to enhance inclusivity.
Gliomas are among the deadliest brain tumors, with limited treatment options and poor survival rates. Scientists from China identified FAM111B, a DNA-repair-associated protein, as a key driver of glioma progression. The study shows that FAM111B overexpression enhances tumor growth and aggressiveness by activating the PI3K/AKT pathway. This is the first research to link FAM111B to gliomas, offering a promising new biomarker and therapeutic target for this intractable disease.
In this Science Bulletin review, the scientists from City University of Hong Kong summarized the recently developed metal-based Janus nanomaterials, including their synthesis methods and application in electrochemical carbon dioxide reduction.
By applying in situ ATR-SEIRAS, electrochemical measurements, DFT calculations and MD simulations, Wen-Bin Cai from Fudan University and Xiaohui Yang from Xiamen University discovered that the additive MBIS adsorbs nearly vertically on cobalt surfaces via thiolate, suppressing the faradaic efficiency of cobalt deposition. Their work also revealed that the difference in dynamic surface coverage of the additive is vital for achieving cobalt superfilling.
In a paper published in National Science Review, an international team of scientists presents the first global inventory of methane emissions from abandoned oil and gas wells. The study estimates that approximately 4.5 million abandoned wells across 127 countries collectively emitted around 400,000 tonnes of methane in 2022, with over 90% of these emissions originating from unplugged wells. By identifying high-priority wells and accelerating the implementation of targeted well-plugging strategies, global methane emissions from these wells could be reduced by up to 61% by 2050.
A comprehensive review highlights cutting-edge advancements in inspection, environmental monitoring, and safety assessment technologies for oil and gas pipelines. These innovations address growing challenges posed by aging infrastructure, geohazards, and complex operational demands, paving the way for smarter, safer energy networks.
A research team led by Prof. LIU Zhongmin and Prof. YE Mao from the Dalian Institute of Chemical Physics (DICP) of the Chinese Academy of Sciences (CAS) and their collaborators proposed a theoretical model describing the migration-aggregation behavior of confined metal clusters within individual zeolite.