Machine learning (ML) is reshaping pipeline integrity management (PIM) from physics-based to data-driven paradigms. This systematic review comprehensively surveys ML applications across pipeline reliability design, integrity assessment, inspection, and maintenance; identifies key bottlenecks including data scarcity, poor interpretability, and limited field validation; and proposes a research roadmap toward tractable, trustworthy ML for full-lifecycle pipeline safety.

Researchers from the University of Oxford have demonstrated a new family of quantum superpositions – Schrödinger’s cat- like quantum states – from highly nonclassical building blocks. The work opens a new route towards quantum computing with non-binary systems, sensing and understanding quantum physics at a more fundamental level.

Monoterpenoid indole alkaloids (MIAs) are plant-derived molecules with unique pharmaceutical potential, but their highly complex and intricate structures make them difficult to produce in the laboratory. Now, researchers from Japan have developed a new strategy that enables the total synthesis of bisleuconothine A and bousigonine B, two biologically relevant oligomeric MIAs. Their findings lay the foundation for further studies of this family of molecules.  Monoterpenoid indole alkaloids (MIAs) are plant-derived molecules with unique pharmaceutical potential, but their highly complex and intricate structures make them difficult to produce in the laboratory. Now, researchers from Japan have developed a new strategy that enables the total synthesis of bisleuconothine A and bousigonine B, two biologically relevant oligomeric MIAs. Their findings lay the foundation for further studies of this family of molecules.  

Achieving high safety in energy storage systems is paramount but hindered by the catastrophic risks of thermal runaway propagation (TRP). This study develops a gradient-laminated ceramifiable silicone foam composite to resolve the inherent trade-off between thermal insulation and dynamic impact toughness. By integrating a polydimethylsiloxane foam matrix with a load-bearing glass fiber fabric skeleton, the material utilizes silane coupling agents for robust interfacial adhesion, while multiscale fillers promote synergistic ceramicization. Characterization reveals robust mechanical durability, maintaining stable elasticity across a wide temperature range (− 40 to 300 °C) and retaining 93% residual stress after 1,000 compression cycles. Under extreme thermal exposure, the foam transforms into a dense ceramic barrier, reducing total heat release by 54.4% and sustaining thermal protection for over 30 min. Crucially, during battery module testing, this architecture efficiently intercepts high-velocity gas jets and confines thermal runaway to a single cell. Fabricated via a scalable process, this composite paves a viable way for constructing intrinsically safe energy storage systems.

Inspired by firefly bioluminescence, scientists have created a flexible fiber sensor that converts electrical sensing signals into optical signals directly on the fiber itself. The ESOT FiSensor simultaneously detects vibration, pressure, temperature and strain through a single optical fiber, transmitting data over long distances with complete immunity to electromagnetic interference — far surpassing traditional sensors.

Freestanding oxide membranes are promising platforms for flexible electronics, sensors, and heterogeneous device integration, but cracks and wrinkles formed during release and transfer can compromise their performance and reliability. Researchers at the University of Science and Technology of China have developed a lock-in thermography-based method that converts these hidden structural imperfections into characteristic thermal signatures. The approach enables rapid, wide-area inspection of conductive oxide membranes and provides semi-quantitative information on crack size, crack orientation, and wrinkle morphology.