Researchers at the University of Pennsylvania have released Observer, the first multimodal dataset of anonymized, real-world clinical encounters, combining video, audio, transcripts and health records. By capturing how care actually unfolds in the exam room, the dataset enables new research into communication, clinical practice and the responsible development of AI for health care.

An analysis by Shinshu University and the Distribution Economics Institute of Japan shows that wholesale prices for flounder—a key Fukushima catch—remained stable after the August 2023 ALPS-treated water discharge. Using a predictive market model based on pre-discharge data, the study found no significant price decline, providing reassurance for fisheries stakeholders and informing future policy debates on nuclear-water disposal.

Coordinating with public health officials, offering students laptops and showing them how to connect, creating Wi-Fi hubs for neighborhoods that lack them, developing new systems for distributing school lunches and offering in-person learning hubs for students whose parents were essential workers are among the many specific recommendations included in a new book about K-12 learning during a pandemic. 

An Osaka Metropolitan University-led research team conducted a review that examined the theoretical foundations of Artificial Superintelligence and explored how misaligned AI systems could optimize for wrong objectives, leading to patient harm and systemic failures.

Using satellite imagery and government data, researchers measured a 5 percent increase in voluntary, or non-subsidized, cover crop adoption by Arkansas farmers. Using data from 2013-2019, agricultural economists with the experiment station showed that cover crop adoption in Arkansas had the greatest association with a soybean-to-soybean cash crop rotation. The finding came out of research seeking to pinpoint how farmers were using cover crops and where, to help policymakers develop more targeted incentives for using cover crops.  

Environmental pollutants such as ozone can damage the internal structure of plants and impair their growth and productivity. Conventional assessments requiring sample processing may not accurately reflect internal damage. A new study highlights the application of a novel ‘optical coherence tomography’ scanner for the non-invasive temporal assessment of live plants exposed to ozone. The technique can aid the timely detection of deficiencies and stress-induced damage, thereby supporting better plant management and improving crop productivity.