This week’s list of data news highlights covers September 18, 2021 to September 24, 2021 and includes articles on repairing the International Space Station with augmented reality and predicting irreversible climate change.
1. Unearthing Wonder Materials
Researchers at the University of Liverpool have created a machine learning model that can identify wonder materials, physical substances such as graphene that exhibit powerful and surprising properties. So far, the model has found four new wonder materials, including one that can improve electric vehicle batteries.
2. Establishing Internet Connectivity
Alphabet has deployed sensor technology to establish a high-speed Internet connection between the Republic of Congo and the Democratic Republic of Congo. Instead of using fiber cables, the sensors are able to connect cities across the Congo River by beaming light to each other. So far, the sensors have enabled the delivery of over 700 terabytes of data.
3. Identifying Effective Drug Combinations
Researchers at The Institute of Cancer Research in London have partnered with BenevolentAI, a UK-based biotechnology company, to create effective drug combinations for children with a rare brain cancer. The team used an AI system that identified two drugs, everolimus and vandetanib, that could aid in treatment. In trials with mice, the drug combination extended survival by 14 percent.
4. Repairing the International Space Station
NASA has launched an augmented reality program to assist astronauts with low-stakes repairs on the International Space Station. Astronauts can wear a headset that will deliver real-time instructions and 3D directional cues.
Researchers at New York University Langone Health have created an AI system that can aid radiologists in detecting breast cancer from ultrasound images. In tests, radiologists using the system accurately diagnosed breast cancer in 96 percent of cases.
NASA has announced a new lunar program to explore the Moon’s south pole. The space agency will launch a rover with sensors to collect data on the terrain and build a map of ice in the south pole.
Researchers at the University of Waterloo have developed an AI system that can predict climate events capable of causing irreversible damage and accelerating climate change. The team trained the system on 500,000 models of extreme climate events like ocean acidification and extreme heat.
The city of Detroit has partnered with Kiwibot, a sidewalk delivery robot company, to launch an autonomous delivery program in Corktown, a historic neighborhood in downtown Detroit. The robots will deliver food and goods from local retailers and restaurants, as well as report data on operational issues to the city.
Unagi Scooters, an electric scooter company based in the United States, has developed a scooter with an AI system that warns of potential collisions. The system can distinguish between stoplights, stop signs, people, cars, and inanimate objects, and will provide audio and visual warnings for potential mishaps.
10. Predicting Long-Term Weather Forecasts
The Climate Prediction Center at the National Oceanic and Atmospheric Administration (NOAA) has determined 30-day and 90-day weather projections for the United States using a supercomputer. The Center used data on past weather events and climate patterns to predict that most of the northeast United States will experience warmer weather for the next 90 days.
Image credit: NASA