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.

Researchers in Shanghai have developed a high-efficiency cryomodule with high quality factors, promising enhanced performance and accessibility for particle accelerator applications in healthcare, industry, and scientific research.

A research team comprising members from the National Institute for Materials Science (NIMS), Tohoku University, and the Photon Science Innovation Center (PhoSIC) has successfully identified the optimal composition of a Co-Mn-Si Heusler alloy exhibiting the highest spin polarization—a key property of half-metallic materials—in just one day of experiments. This achievement was made possible by fabricating a composition-spread thin film on a single substrate and conducting photoelectron measurements at a 3 GeV high-brilliance synchrotron radiation facility, NanoTerasu. This highly efficient optimization approach is expected to accelerate the development of high-performance spintronic materials.

Solid-state batteries have attracted significant attention due to their high safety and high energy density. However, the continuous lithium (Li) consumption will lead to decrease of the energy density and lifespan of the battery. Efforts have been devoted to exploring new strategies to increase the content of the active Li-ions, among which application of sacrificial additives is one of the most effective ways to compensate for the lost lithium. To spur more research into a stable and high-efficiency additive, researchers in China assessed the current advances of research and proposed Li_xC₆O₆ (x=2 and 4) as high-efficiency Li-compensation additives in cathode.

ADHD can exacerbate anxiety, and anxiety can exacerbate ADHD. Girls are particularly vulnerable to developing anxiety disorders, but researchers have recently discovered something that may help slow down – or prevent – the development of these types of disorders.

An investigation into Viking skeletons reveals a hidden story of violence, power, and the surprising differences between neighbouring Viking societies.

A recent study published in SCIENCE CHINA Earth Sciences offers a comprehensive examination of factors driving the Arctic amplification, while also comparing quantitative results from multiple studies. The review highlights the complex interactions of various drivers, including local feedbacks, atmospheric circulation, ocean currents, and aerosols. It also underscores significant uncertainties in quantifying their contributions and emphasizes the need for more reliable data and improved models to enhance understanding and predictions of Arctic climate change.

Ultrasound imaging is one of the most widely used diagnostic tools in modern medicine. Behind its non-invasive magic lies a class of materials known as piezoelectric single crystals, which can convert electrical signals into mechanical vibrations and vice versa. Now, in a world-first, a research team from Kumamoto University has successfully visualized how tiny structures inside one of these crystals respond to electric fields in real time—shedding light on the dynamics of nanostructure in materials used in ultrasound probes.