This week’s list of data news highlights covers December 11, 2021 to December 17, 2021 and includes articles on entering the sun’s atmosphere for the first time to collect solar data and using an AI system to train crisis volunteers.
Researchers at the University of Waterloo and Darwin AI, a Canadian visual inspection quality company, have created an AI system that can predict a COVID-19 patient’s chances of requiring intensive care unit (ICU) treatment. The system uses over 200 clinical data points such as medical history and vital signs to determine a treatment plan. In tests, the system predicted the need for ICU treatment with over 95 percent accuracy.
A spacecraft has entered the sun’s atmosphere for the first time. NASA sent the Parker Solar Probe to the sun to collect data on solar wind, space weather, and solar eruptions.
Researchers at the University of Washington have developed a machine learning model that can predict lightning strikes in the southeastern United States two days earlier than previously leading techniques. The team trained the model with data on lightning strikes from 2010 to 2016.
Scientists at Harvard have proven the existence of quantum spin liquid, a phase of matter in which electrons do not abide by typical magnetic properties. Researchers can use quantum spin liquid to create topological qubits, a new form of qubit that can improve quantum computers.
The Trevor Project, a U.S.-based crisis intervention organization focused on LGBTQ+ youth, has created an AI system to train volunteers. The system creates written inquiries and responses that are similar to past submissions to the organization. Since implementing the system, the Trevor Project has added 1,000 volunteers to its counseling program.
Honda has partnered with the Ohio Department of Transportation to monitor road conditions in the state with connected car technology. Honda will use GPS systems and cameras to collect data on poor road conditions and report their location, type, and severity to transportation officials.
Researchers at the University of Exeter have developed an AI system that can predict a patient’s chances of developing dementia within the next 2 years with 92 percent accuracy. The team trained the model with medical data from 15,300 patients.
Researchers at Durham University have used a supercomputer to model impact scenarios that can explain Uranus’s unusual planetary tilt. The team found that the tilt may have been caused by an asteroid impact around four billion years ago.
Researchers at Microsoft have created an AI system that can identify common bugs in Python code. The team trained the system by creating two networks: one meant to create bugs in existing code and the other meant to find bugs in code.
Researchers at Massachusetts General Hospital have created a predictive model that can detect signs of lung cancer in asymptomatic patients from a blood sample. Doctors can use the model to screen patients for lung cancer and start treatment plans before the disease progresses.
Image credit: Flickr user NASA Goddard Space Flight Center