This study presents a hybrid dynamic best model selection algorithm for real-time building fire prediction using IoT-enabled smoke, temperature, and humidity sensors. Validated on a custom laboratory dataset and live sensor streams, the system dynamically selected the best machine-learning classifier and achieved 99.57% accuracy, 99.37% precision, 99.79% recall, and low error, improving adaptability and early fire detection.

Propylene is vital for industry, but catalysts often fail due to heat. Researchers at Nankai University have engineered a robust Pt-Fe bimetallic catalyst by anchoring atomically dispersed Pt-O-Fe centers within MFI zeolite nanosheets for propane dehydrogenation. This framework-mediated architecture provides a thermodynamic barrier against metal sintering and coking, maintaining >95% propylene selectivity and full regenerability. By resolving the long-standing stability-activity trade-off, this structural design offers a transformative paradigm for industrial-scale, sustainable propylene manufacturing.

A new review in Journal of Management Science and Engineering maps nearly five decades of climate adaptation economics, covering 6,248 studies from 1978 to 2025. It shows how households, firms, markets and governments adapt to climate risk, why adaptation and mitigation must be coordinated, and where future research needs better data, causal evidence and models of inequality.

Patients with untreatable conditions such as sight loss or loss of motor-function could be closer to a viable technology for restoring their lost sense, within a faster time frame. This is due to the discovery that advanced brain interfacing technology used for both touch and vision prostheses, is in fact almost the same, despite being developed completely separately for more than 50 years. The comprehensive review in which this discovery has first been made, was published in Nature Reviews Bioengineering, and was led by Giacomo Valle, Assistant Professor at Chalmers University of Technology, in Sweden. 

Researchers developed a triple-synergy strategy that locks ultrafine ruthenium nanoparticles inside a nitrogen-doped porous carbon framework. Spatial confinement, nitrogenous defect anchoring, and electronic modulation work together to improve alkaline hydrogen evolution. The resulting Ru/3DSPNC catalyst delivers high activity and long-term durability in anion exchange membrane water electrolysis, operating stably at 1.9 V for 310 hours.

Researchers have developed a programmable genome evolution platform called RAMPAGE that rapidly generates beneficial mutations in the industrial yeast Pichia pastoris. The technology enabled the evolution of strains that tolerate extremely high methanol concentrations, grow at elevated temperatures, and overcome growth defects caused by glycoengineering. The platform could accelerate the development of more robust microbial cell factories for sustainable biomanufacturing.

Two-wheeled vehicles naturally lean during turns, making it challenging for rider-assistance systems to tell if the rider is making a planned turn or losing balance. In a recent study, scientists from Shibaura Institute of Technology in Japan created a machine-learning-based steer-by-wire bicycle that can differentiate between these situations. This innovation could make electric bicycles and motorcycles safer without disrupting the riders’ intended movements.

POSTECH and Incheon National University develop mussel adhesive protein to achieve “sustained immunity,” offering a technology that could ease the burden of repeat vaccination.