EurekAlert! News by Subject: Tech & Engineering - Home

Tech & Engineering

Updates every hour. Last Updated: 30-Oct-2026 20:15 ET (31-Oct-2026 00:15 GMT/UTC)

New framework renders AI trustworthy for cancer subtyping

Vanderbilt University Medical Center
Peer-Reviewed Publication

Reported June 23 in Nature Biomedical Engineering, researchers at Vanderbilt Health and centers in Hong Kong have created a versatile uncertainty-aware AI framework broadly adaptable as a wrapper for digital pathology AI systems. (An AI wrapper acts as an interface layer that customizes, formats and automates how users interact with the underlying intelligence.) They demonstrate their wrapper, called TRUECAM, primarily with reference to non-small cell lung cancer (NSCLC) subtyping using whole-slide images.

Journal: Nature Biomedical Engineering

Can scientists learn cells' language? UCLA-led team aims to decode cellular conversations

University of California - Los Angeles Health Sciences
Peer-Reviewed Publication

UCLA scientists are calling for a large-scale initiative to understand how human cells influence one another — a missing layer of biology that could reveal how cellular interactions drive disease and inform new therapies.

Journal: Nature Biotechnology

China deploys first 24-hour rapid intensification forecast model for typhoons

Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences
Business Announcement

The model, developed at the Shenzhen Institutes of Advanced Technology (SIAT) of the Chinese Academy of Sciences, recently completed operational deployment and real-world application testing at the country's National Meteorological Center.

Algorithms for species conservation

Universitaet Stuttgart
Peer-Reviewed Publication

Patterns reveal the individual: A novel AI algorithm equips researchers with a powerful new tool to accurately identify individual jaguars, zebras and giraffes in real time based on their unique coat patterns. Biologists and ecologists can now observe individual wild animals as they move from place to place, as well as their behavior and development over long periods of time, accurately, with much less effort and significantly faster than before – a major step forward for nature conservation.The scientists explain how this works in the journal "Methods in Ecology and Evolution".

Journal: Methods in Ecology and Evolution

Efficient power generation and cooling in cubic tin selenide

Maximum Academic Press
Peer-Reviewed Publication

Thermoelectrics are regarded as potential materials in energy technology because they can enable direct and reversible conversion between heat and electricity.

NTU Singapore scientists create optical skyrmions using a two-century-old light phenomenon

Nanyang Technological University
Peer-Reviewed Publication

Nanyang Technological University, Singapore (NTU Singapore) scientists have used a classic optical phenomenon known as the “Poisson spot” to create stable patterns of light called “optical skyrmions”. Optical skyrmions are tiny, swirling configurations in the properties of light – akin to the spikes of a hedgehog. The team used a laser directed at a small circular disc, instead of the complex and costly engineered materials commonly used to generate these skyrmions. This new method gives scientists a much simpler way to generate, study and adjust optical skyrmions.

Skyrmions are currently a hot scientific subject as they hold potential to store information, paving the way for future data storage, communications and computing systems. Published in a leading scientific journal, Optica, the discovery was led by Nanyang Assistant Professor Shen Yijie from NTU’s School of Physical and Mathematical Sciences and School of Electrical and Electronic Engineering.

Journal: Optica

Efficient algorithm enables translation of black-box AI models into interpretable medical decision knowledge

Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences
Peer-Reviewed Publication

In a study published in Nature Biomedical Engineering, Prof. CAI Yunpeng's team from the Shenzhen Institutes of Advanced Technology (SIAT) of the Chinese Academy of Sciences developed a mathematical framework named Class-Association Manifold Learning (CAML), which efficiently maps the decision rules of black-box AI models into a low-dimensional manifold that friendly visualizes knowledge to human experts.

Palatable food engages reward-stress neural circuit to relieve anxiety

Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences
Peer-Reviewed Publication

In a study published in Advanced Science, Dr. TU Jie's team from the Shenzhen Institute of Advanced Technology (SIAT) of the Chinese Academy of Sciences identified a neural circuit from dopamine 1 receptors (D1R) neurons in prefrontal cortex (PFC) to corticotropin-releasing factor (CRF) neurons in paraventricular nucleus (PVN) of hypothalamus—PFC^D1R→peri-PVN^CRFR1→PVN^CRF—that functionally connects the brain's reward system with the stress-response system and mediates reward-driven top-down regulation of stress.

AI and polygenic scores improve breast cancer risk assessment

Kaiser Permanente
Peer-Reviewed Publication

Kaiser Permanente study finds a mammography AI risk score used along with polygenic and clinical risk scores identifies more high-risk women.

Journal: JNCI Journal of the National Cancer Institute
Funder: National Cancer Institute, The Permanente Medical Group Delivery Science and Applied Research Program