In the realm of smart manufacturing and digital engineering, a new technology named Data-Model Fusion (DMF) is gaining traction. A review paper in Engineering details how DMF integrates model-based and data-driven methods, addresses their limitations, and finds applications across the product lifecycle. It also explores DMF’s future directions, showing its potential to reshape industrial processes.

In a recent study published in Engineering, researchers from Xi’an Jiaotong University have developed a self-adaptive core-shell dry adhesive with a “live core”. This new adhesive addresses the long-standing issue of weak adhesion under non-parallel contact in engineering operations, offering improved performance and potential applications in various fields such as robotic gripping and optical component assembly.

A recent study in Engineering reveals that bank filtration (BF) can be a highly effective pretreatment for gravity-driven membrane (GDM) filtration. This approach addresses common issues in GDM systems, such as poor permeate quality and low stable flux when treating polluted water, offering a sustainable and practical solution.

In a paper recently published in Chinese Physics Letters, a research team from Peking University report their latest discovery in the field of high-temperature superconductivity, revealing the existence of pair density modulation within a single unit cell of iron-based superconductors. This finding provides unprecedented microscopic insights into unconventional Cooper pairing mechanisms at the atomic scale.

On February 11, the team from the Data Darkness Lab (DDL) at the Medical Imaging Intelligence and Robotics Research Center of the University of Science and Technology of China (USTC) Suzhou Institut introduced a new out-of-core mechanism, Capsule, for large-scale GNN training, which can achieve up to a 12.02× improvement in runtime efficiency, while using only 22.24% of the main memory, compared to SOTA out-of-core GNN systems. This work published on ACM Journals.

Given the multitude of conditions that must be optimized in synthesis routes, chemical synthesis remains a complex and multidimensional challenge. The rapid development of computational guidelines and machine learning (ML) techniques has brought exciting hope to this dilemma. A new study published in the journal National Science Review highlights the advancement of computationally guided and ML-assisted approaches in inorganic material synthesis.

Against the backdrop of accelerating global climate change and urbanization processes, urban transportation systems are confronting increasingly complex multi-hazard risks. Spatiotemporal big data, characterized by its high precision and information density, has demonstrated growing significance in transportation system resilience studies. Nevertheless, the current comprehension of the evolutionary trajectory of spatiotemporal big data applications in this domain remains fragmented. In this context, our study conducts a systematic review of global research, elucidating the practical implementations of spatiotemporal big data in transportation system resilience studies. The investigation reveals that multi-source big data with high spatiotemporal resolution has not only catalyzed methodological innovations in resilience assessment but has also potential to facilitate a paradigm shift in the field - transitioning from macro-scale to micro-scale analyses, from static evaluations to dynamic monitoring approaches, and from post-disaster emphasis to comprehensive lifecycle investigations. Journal of Geo-Information Science has published the study's results.

Researchers from the University of Science and Technology of China (USTC) achieved the first direct laboratory observation of ion acceleration through reflection off laser-generated magnetized collisionless shocks. This observation demonstrates how ions gain energy by bouncing off supercritical shocks, central to the Fermi acceleration mechanism. The research was published in Science Advances.

In a paper published in National Science Review, a Chinese team of scientists presents the role of pyrope garnet in water transport from the upper mantle to the topmost lower mantle. Pure single crystals of pyrope garnet were synthesized at high-pressure and high-temperature conditions of the upper mantle to the top lower mantle using a large-volume press. Pyrope garnet can contain up to 2000 wt. ppm water with a strong dependence on pressure and temperature in the transition zone and topmost lower mantle. Hydrated pyrope garnet may serve as a vital water carrier and reservoir in the deep mantle, offering new insights into water cycling up to the topmost lower mantle.

Medical image segmentation plays a crucial role in facilitating clinical diagnosis and treatment, yet it poses numerous challenges due to variations in object appearances and sizes with indistinct boundaries. This paper introduces the MHSAttResDU-Net architecture, a novel approach to automatic medical image segmentation. Drawing inspiration from the double U-Net, multi-head self-attention (MHSA) model, and residual connections, the proposed model is trained on images pre-processed by the innovative ranking-based color constancy approach (RCC). The MHSAttResDU-Net includes the integration of RCC to control model complexity and enhance generalization across diverse lighting conditions. Additionally, the incorporation of the sparse salient region pooling (SSRP) unit in the encoder-decoder blocks reduces the dimension of feature maps, capturing essential local and global channel descriptors without introducing learnable parameters. MHSA gates are strategically employed in both down-sampling and up-sampling paths, allowing the recollection of additional relevant dimensional data. This effectively addresses dissimilar feature representations, minimizing unfocused noise and artifacts while reducing computational costs. Furthermore, Leaky ReLU-based residual connections between the encoder and decoder enhance the model’s capability to recognize complex shapes and structures, ensuring improved gradient flow and faster convergence. Experimental results demonstrate the superiority of the MHSAttResDU-Net architecture across diverse datasets, including COVID-19, ISIC 2018, CVC-ClinicDB, and the 2018 Data Science Bowl. The model achieves state-of-the-art performance metrics, including an accuracy of 99%, representing a promising advancement in automated medical image analysis with potential implications for improving patient care and diagnostic accuracy.