This week’s list of top data news highlights covers April 23, 2022 to April 29, 2022 and includes articles on using a supercomputer to model storm surges and eradicating invasive species with an AI system.
Researchers at the University of Texas at Austin have created an enzyme variant that can break down plastic materials in a few days. The team used an AI system to predict which mutations in a natural enzyme would enable bacteria to depolymerize plastics most quickly.
Researchers at the Lawrence Berkeley National Laboratory have used a supercomputer to simulate how climate change may impact flooding from storms in the San Francisco bay area. The team found that the region would likely experience 26 to 37 percent more flooding from storms by 2100.
The Nature Conservancy, a U.S.-based nonprofit organization dedicated to environmental conservation, has started using an AI system to detect signs of Himalayan ginger, an invasive weed, in Hawaiian rainforests. The system locates the weed in aerial images of forests and alerts the organization, which sends conservationists to remove it.
Pony.ai, a U.S.-based autonomous vehicle company, has received a license to operate an autonomous taxi service in Guangzhou, China. The company is the first to receive a taxi license in China and will begin taxi operations in May.
Researchers at the Vellore Institute of Technology in Chennai, India have created an AI system that can detect epileptic seizures from electroencephalography (EEG) signals better than current methods. The team trained the system with EEG data from healthy patients and patients experiencing a seizure.
The California Department of Public Health has launched a chatbot that can answer questions about COVID-19 in English and Spanish on WhatsApp. Users can text a designated number or scan a QR code to receive current information about vaccine safety, health recommendations, and digital vaccine records.
Meta has created a new formula for concrete for its data centers that releases less carbon dioxide. Currently, concrete accounts for 8 percent of global carbon dioxide emissions because of energy consumption and chemical reactions created during the manufacturing process. Meta used an AI system to create a formula that contains 40 percent less carbon but still meets appropriate industrial strength metrics.
Snapchat has launched an in-app shopping hub called Dress Up where users can virtually try on apparel and cosmetics with an augmented reality lens and send pictures of chosen outfits to their friends. The tool uses a machine learning model that can turn images of people or products into 3D models.
Researchers at the Open University of Catalonia have created an AI system that can distinguish between mild cognitive impairment and progressive cognitive impairment caused by Alzheimer’s disease in magnetic resonance images (MRIs). The team trained the system with MRIs from patients with Alzheimer’s disease and healthy patients. Healthcare researchers can use the system to better identify cases and personalize patient treatment plans.
Louisiana’s Coastal Protection and Restoration Authority has released new estimates of the financial impact of damage caused by storm surges. They used a supercomputer to simulate storms of varying sizes and found that coastal communities in Louisiana will likely face $5.5 billion in damages to structures, infrastructure loss, temporary relocation efforts, lost wages, sales, and rent by 2023 in the absence of additional mitigation efforts.