Teaching machines to catch lies across voice, face, and words
Peer-Reviewed Publication
Updates every hour. Last Updated: 8-Jun-2026 22:16 ET (9-Jun-2026 02:16 GMT/UTC)
Researchers at Mass General Brigham behind FaceAge, an AI tool that estimates biological age from facial photographs, found that tracking changes in FaceAge over time provides added insight into cancer prognosis. In a study of more than 2,200 patients, a faster Face Aging Rate (FAR) was strongly associated with worse survival, supporting its potential as a non‑invasive biomarker to inform treatment planning and follow‑up.
Artificial light at night not only alters the landscape, but also profoundly disrupts natural ecosystems. A recent study by the RPTU University Kaiserslautern-Landau shows that light pollution can significantly disrupt the exchange of energy and nutrients between bodies of water and their surrounding habitats – sometimes even more than non-native species. Thus, lighting along riverbanks or streams can have far-reaching ecological consequences. The results were published in the journal Functional Ecology.
Cancer treatment and other delicate medical procedures could one day be carried out using tiny microrobots guided precisely inside the body after scientists developed a new magnetic tool to control them.
Researchers created a bismuth-coordinated melanin material that provides in vivo physical radiation shielding and ROS scavenging. It markedly mitigates acute radiation syndrome and boosts mouse survival from 20% to 60% after lethal total body irradiation.
A new academic study has found that newer forms of blockchain technology could significantly cut the vast energy consumption associated with digital currencies, addressing one of the sector’s most persistent criticisms.
Researchers have developed a family of carbonyl rich carbon spheres with surface wrinkle structure for the efficient electrosynthesis of hydrogen peroxide. This unique wrinkled design significantly expands the accessible active area and utilizes carbonyl moieties to deliver exceptional H2O2 selectivity (>97.5%). The study provides a valuable surface modulation strategy for designing advanced heteroatom-doped carbon electrocatalysts.