Osaka Metropolitan University researchers have developed an AI model that can detect the presence of fatty liver disease from chest X-ray images. 

Tennessee Economic and Community Development Commissioner Stuart McWhorter visited UT Knoxville to explore research initiatives in nuclear energy, advanced materials, and AI. Collaborations with Oak Ridge National Lab and Y-12 aim to drive innovation, enhance Tennessee’s workforce, and address national priorities through multidisciplinary projects and record-breaking research expenditures.

A research team from ShanghaiTech University has created a new method for designing two-dimensional patterned hollow structures (2D-PHS) with improved mechanical properties for aerospace and automotive applications. By using Conditional Generative Adversarial Networks (cGAN) and Deep Q-Networks (DQN), they optimized the design of 2D-PHS much faster than traditional finite element analysis (FEA). Their optimization enhanced stress uniformity by 4.3% and reduced maximum stress concentrations by 23.1%. These improvements were validated through simulations and tensile tests on 3D-printed samples, which showed tensile strength increased from 5.9 to 6.6 MPa. This study highlights the effectiveness of AI in efficient material design.

USC researchers have identified a new brain imaging benchmark that may improve how researchers classify biologically meaningful changes associated with Alzheimer’s disease, especially in Hispanic and non-Hispanic white populations. Using an advanced brain imaging scan called tau PET, the research team studied over 675 older adults from the Health and Aging Brain Study–Health Disparities (HABS-HD) aiming to identify the optimal brain signal that distinguishes individuals with clinically-relevant biological markers of AD from those who are aging normally. They compared tau PET scans of study participants who were cognitively impaired with those who were not impaired based on cognitive tests to establish a tau cut-point that would indicate a higher risk for Alzheimer’s disease. The team used a new imaging tracer called 18F-PI-2620, to measure tau protein buildup in the brain. They found that when tau levels in the medial temporal lobe­—a region deep in the brain—exceeded a certain threshold, it strongly indicated cognitive impairment related to AD. But the tau cut-point was only effective when another abnormal protein, amyloid, was also present in those with cognitive impairment, and it only worked for Hispanic and non-Hispanic White participants. In non-Hispanic Black participants, the tau cut-point did not perform as expected. This suggests that other pathologies or conditions may be driving cognitive decline in this group. The findings reflect a growing focus in AD research on making sure diagnostic tools work for everyone—not just in narrow clinical trial populations.

Just the word “quantum” can make even seasoned science teachers break into a sweat. But a national pilot program led by The University of Texas at Arlington is helping take the mystery out of the subject for students and educators alike. This week, 50 high school students and science teachers gathered at Arlington Martin High School to dive into the topic through Quantum for All, a program launched by Karen Jo Matsler, a professor of practice and master teacher in UT Arlington’s UTeach program. 

In a commentary published today (June 26, 2025) in the journal Cell, Arc researchers introduce the independent nonprofit’s first “Virtual Cell Challenge,” a public competition with a grand prize worth $100,000 for the machine learning model that best predicts how cells will respond to genetic perturbations. Arc is introducing the competition to catalyze progress at the interface of artificial intelligence and biology, particularly by accelerating the creation of high quality datasets and sparking a conversation about rigorous standards for assessing how well AI models simulate cellular behavior.