Teacher emotion recognition (TER) is crucial for classroom dynamics, yet limited by scarce multimodal data and feature modeling. We present a multimodal TER dataset with 102 lessons and 2,170 video segments, annotated with interaction-based emotional tags. We propose an Emotion Dual-Space Network (EDSN) with a Commonality Space (ECSC) using central moment differences to align modalities, and a Discrimination Space (EDSC) using gradient reversal and orthogonal projection to extract unique features. EDSN achieves 77.0% accuracy and 0.769 F1-score on the dataset, outperforming state-of-the-art models, demonstrating its effectiveness in educational emotion recognition. 

Human health is the Achilles heel of space travel. Researchers at ETH Zurich have now succeeded in printing complex muscle tissue in zero gravity, via parabolic flight experiments. Their work paves the way for beyond-Earth health solutions and testing for future space missions.

Recently, a research team led by Gengjie Jia at the Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, in collaboration with Yu Li at The Chinese University of Hong Kong, Xin Gao at King Abdullah University of Science and Technology, and Andrey Rzhetsky at The University of Chicago, published an article in Quantitative Biology titled “An effective encoding of human medical conditions in disease space provides a versatile framework for deciphering disease associations.” The study presents an embedding-based approach that encodes medical records into a high-dimensional disease space, providing a versatile framework for uncovering disease associations, facilitating genetic parameter estimation, and enabling data-driven disease classification. The authors also discuss the key challenges and future prospects of this emerging paradigm.

A practical, evidence-based checklist developed by scientists at the University of Surrey is helping everyone from keen gardeners to local councils plan their next greening project with confidence.  

A recent paper from the University of Chicago Pritzker School of Molecular Engineering (UChicago PME), published in Nature Communications, built an active learning model that was able to explore a virtual search space of one million potential battery electrolytes starting from just 58 data points. From this minimal data, the team from the lab of UChicago PME Asst. Prof. Chibueze Amanchukwu identified four distinct new electrolyte solvents that rival state-of-the-art electrolytes in performance.

Our galaxy’s most abundant type of planet could be rich in liquid water due to formative interactions between magma oceans and primitive atmospheres during their early years. New experimental work demonstrates that large quantities of water are created as a natural consequence of planet formation. It represents a major step forward in how we think about the search for distant worlds capable of hosting life.