Smart learning bridges training gaps in cervical cancer prevention
Peer-Reviewed Publication
This month, we’re focusing on artificial intelligence (AI), a topic that continues to capture attention everywhere. Here, you’ll find the latest research news, insights, and discoveries shaping how AI is being developed and used across the world.
Updates every hour. Last Updated: 19-Nov-2025 16:11 ET (19-Nov-2025 21:11 GMT/UTC)
A research team has developed a deep learning-based framework that allows agricultural robots to identify new weed species using only a few training images.
Additive manufacturing (AM), with its high flexibility, cost-effectiveness, and customization, significantly accelerates the advancement of nanogenerators, contributing to sustainable energy solutions and the Internet of Things. In this review, an in-depth analysis of AM for piezoelectric and triboelectric nanogenerators is presented from the perspectives of fundamental mechanisms, recent advancements, and future prospects. It highlights AM-enabled advantages of versatility across materials, structural topology optimization, microstructure design, and integrated printing, which enhance critical performance indicators of nanogenerators, such as surface charge density and piezoelectric constant, thereby improving device performance compared to conventional fabrication. Common AM techniques for nanogenerators, including fused deposition modeling, direct ink writing, stereolithography, and digital light processing, are systematically examined in terms of their working principles, improved metrics (output voltage/current, power density), theoretical explanation, and application scopes. Hierarchical relationships connecting AM technologies with performance optimization and applications of nanogenerators are elucidated, providing a solid foundation for advancements in energy harvesting, self-powered sensors, wearable devices, and human–machine interaction. Furthermore, the challenges related to fabrication quality, cross-scale manufacturing, processing efficiency, and industrial deployment are critically discussed. Finally, the future prospects of AM for nanogenerators are explored, aiming to foster continuous progress and innovation in this field.
An international consortium of researchers has created the largest-ever database compiling records of brain activity during sleep and dream reports. One of the first analyses of the database confirmed that dreams do not occur only during REM sleep, but also during deeper and calmer NREM stages. In these cases, brain activity resembles wakefulness more than deep sleep, as if the brain were “partially awake”.
A research team from Wuhan University has developed an innovative machine learning framework that accelerates the discovery of materials with tailored thermal properties. By combining interpretable deep learning with multiscale computational techniques, the team achieved highly accurate predictions of lattice thermal conductivity (LTC) while also shedding light on the underlying physical mechanisms. Their approach maintains the predictive power of traditional "black-box" models but adds the physical interpretability. As result, several high-performance materials for thermal management were identified. This advancement not only speeds up the development of efficient thermoelectric devices and thermal control systems but also provides deeper insights into the fundamental process of heat transfer at the atomic scale.
The National Institute of Information and Communications Technology (NICT) and the Nagoya Institute of Technology (NITech), collaborated with the Japan Aerospace Exploration Agency (JAXA), have achieved the world’s first successful demonstration of next-generation error correction codes, mitigating the impact of atmospheric turbulence on ground-to-satellite laser communications.
Atmospheric turbulence in ground-to-satellite laser links is known to cause fading, resulting in burst data errors. Error correction codes are one of the key technologies to mitigate such effects. In this experiment, we transmitted next-generation error correction codes with high correction capability (5G NR LDPC and DVB-S2) and successfully corrected burst data errors caused by atmospheric turbulence in the laser link. This result confirmed that both codes can significantly improve communication quality compared to conventional schemes.
This achievement is expected to contribute to the practical implementation of ground-to-satellite laser communications by applying these codes.