This week’s list of data news highlights covers April 24, 2021 – April 30, 2021 and includes articles about monitoring orca populations and preventing satellite collisions with space debris.
Businesses are turning to sensor data to gain insight into how workers utilize office spaces to help workers safely return to offices after social distancing guidelines are lifted. Skanska, a Swedish construction company, has installed sensors in a Texas office tower to collect anonymized data on the number of employees in any space at a given time. The sensors include cameras, parking scanners, and QR code readers on security turnstiles. Hines, a Texas real estate firm, has installed 150 sensors in office towers in Texas and Georgia to identify which rooms staff use most, data that tenants can use to inform their decision of how many spaces to rent.
The United Nations Department of Economic and Social Affairs (UN DESA), the UN Environment Program (UNEP), and the Basque Center for Climate Change have created an open-source tool that can help countries measure the contributions of nature to their economies. Officials can use the tool, called the Artificial Intelligence for Environment and Sustainability (ARIES), to track the extent, condition, and services that ecosystems, such as forests and wetlands, provide to society and the economy to shape policies that promote environmental health.
Researchers of the Sealife Response, Rehabilitation, and Researcher Center in Washington have collaborated with Oregon State University to use unmanned drones and machine learning to monitor orca whales in the Salish Sea, which is home to only 75 whales, the lowest number in the sea in 30 years. The researchers used the drones to capture 2,000 images of three orca communities and applied machine learning to the data to analyze, identify, and categorize the images based on the snout shape, blowhole, and eye patches. By using machine learning, scientists have been able to reduce the time it takes to categorize orcas from six months to six weeks.
Firefighters from the Guiyang Fire Department in China are using data collected from near field communication (NFC) chips, which facilitate wireless exchanges of data over short distances, to predict and detect electrical fire hazards. Public hospitals and conference centers have installed these chips to monitor the temperature, electricity current, and voltage in boxes that enclose electrical wires. The systems are connected to an Internet platform that sends alerts to firefighters about abnormally high temperatures.
Researchers from Delft University and Wageningen University in the Netherlands, together with researchers from Monash University in Australia, have discovered that phone signals can help measure smoke and air quality levels. They found that when there were large amounts of smoke in the atmosphere during the Melbourne bushfires that began in 2019, the smoke acted as a lid and caused radio links and mobile phone signals to broadcast irregularly, creating unique phone signal patterns. The researchers hope that by combining phone signal data with data from satellites, local governments can warn residents about potential fire dangers well in advance.
The European Space Agency (ESA) has developed AI algorithms that can help prevent low Earth orbit satellites from colliding with space debris. Currently, teams need to be on-call 24/7 across several days to implement measures that minimize the chance of a spacecraft colliding with another orbiting object. However, this is costly because each time operators move a satellite, it consumes a lot of fuel and requires lots of preparation. To address this, the scientists trained the algorithms to predict the likelihood of a satellite collision occurring in the next three days with a warning alert. Based on this prediction, the algorithms notify teams whether they need to move the satellite, which reduces the number of hours that teams are on-call and how often satellites consume excess fuel.
Community Health Network, a nonprofit care organization for the elderly in California, is implementing data automation tools to improve hospital processes. One tool reduces appointment no-shows by using machine learning to predict which patients are likely to miss their appointment. The tool uses data on their attendance history and sends them specialized text reminders. If they do not confirm via text, a provider contacts them directly to either reschedule or offer transportation services to help them attend. Another tool is a data platform that centralizes information so that nurses can find the information they need in one place, such as requests for an adjustable bed or applying for government benefits.
The New York State Office for the Aging has developed CV19 CheckUp, an AI-powered tool that evaluates a person’s risk of contracting COVID-19. The tool was designed to identify risks for elderly, low income, minority, or LGBTQ populations, groups that the U.S. Centers for Disease Control and Prevention (CDC) has found are less likely to have healthcare coverage and more likely to be immunosuppressed due to asthma, diabetes, heart conditions, and HIV. The tool uses people’s responses to an online questionnaire and data that the CDC and World Health Organization (WHO) have collected. Based on their identified risks, CV19 CheckUp provides recommendations and resources to individuals, such as increasing their social distancing practices or double-masking.
Montfort, an Israeli startup that focuses on developing technologies for psychiatric disorders, has created an AI-powered app that uses smartphone sensors to monitor a person’s motor and cognitive symptoms to help professionals diagnose and treat psychiatric disorders. The app prompts users with on-screen commands and uses the sensors behind phone screens to measure a user’s reaction times. Then, the app uses machine learning to analyze the sensor data to generate digital biomarkers, which psychiatrists can use to predict and determine whether there are disturbances in brain connectivity that reflect a specific disorder.
Researchers at the Institute of Photonic Sciences in Barcelona have developed a new wearable cap that doctors can use to see the brain activity of newborns, an alternative to using MRI scans, which emit harmful radiation. The cap uses near-infrared lasers, LED lights, and electrical conductors to send light signals into an infant’s brain and uses sensors to measure the concentration of oxygen in the blood surrounding the brain to build a 3D color image of a newborn’s brain in real time. With this image, physicians can monitor slight drops in critical oxygen levels to and from the brain that could lead to neurodevelopmental disabilities.
Image credit: Nitesh Jain