This week’s list of data news highlights covers March 20, 2021 – March 26, 2021 and includes articles about detecting freshwater pollution from sensors on mussels and using AI to identify viable embryos.
1. Helping People with Low Vision or Blindness Navigate Public Spaces
Researchers at the University of Georgia are using IBM’s AI software to develop backpacks that help people with blindness or low vision better navigate public spaces. The backpacks are embedded with AI cameras that use computer vision to read signs, detect crosswalks, and identify changes in elevation, and also include a match-box-sized GPS unit. Wearers can communicate with the system using bluetooth earphones to ask for location information and the system can warn wearers of nearby threats, such as an oncoming pedestrian.
2. Creating an Internet-of-Mussels to Detect Toxins in Freshwater
Researchers at North Carolina State University have developed a system that allows them to remotely monitor the behavior of freshwater mussels, which are one of the first creatures to react to toxic substances in aquatic ecosystems. When a group of mussels feed, they each open their shells to filter tiny organisms but if one of them identifies a harmful substance, they will all suddenly close their mouths in unison. The researchers have attached sensors to some of the mussels, allowing them to detect if any of the mussels identify toxins and more quickly begin investigating the source of pollution.
3. Making Sense of Car Accidents
Israeli computer-vision company Nexar has partnered with Japanese insurance company Mitsui Sumitomo Insurance to incorporate its automatic crash reconstruction system into cars in Japan. The system makes sense of car accidents by combining camera footage, information from the car’s sensors, and GPS data to map out the accident. Then it uses AI to draw conclusions from the information, such as the timeline of events and who was at fault. There are 200,000 cars in Japan using the system.
4. Determining the Relationship Between Coffee and Heart Health
Researchers from the University of Colorado School of Medicine have used machine-learning algorithms to look for links between thousands of patient characteristics collected in a well-known cardiovascular study and the odds of those patients developing heart failure. Their initial results suggested that two, three, or more cups of coffee per day may lower the risk of heart failure. To test their hypothesis, the researchers used the algorithms to predict the relationship between coffee intake and heart failure in two other respected data sets and found convincing results.
5. Analyzing Properties of Electron Beams
Researchers at the U.S. Department of Energy’s SLAC National Accelerator Laboratory have developed a tool that can analyze and adjust the quality of electron beams, which power devices such as electron microscopes, X-ray lasers, and medical accelerators. Conventional beam diagnostic tools physically interact with a beam to measure properties such as intensity and shape, but physical interaction often destroys the beam. Instead, the researchers trained an algorithm to accurately predict beam properties for experimental situations and found that their approach can provide more detailed information about each of the millions of electrons in a beam.
6. Helping Choose Viable and Healthy Embryos for IVF
Researchers at the Kaunas University of Technology in Lithuania and Esco Medical Technologies, a manufacturer of medical equipment, have developed a tool that tracks and analyzes the development of embryos to increase the chances of choosing the most viable for transfer to the uterus. The tool takes pictures of developing embryos every five minutes and uses an algorithm to process the collected visual data and identify events and anomalies that may indicate embryos that are less likely to have successful further development.
7. Converting Radio Waves into Electricity for Wearables
Scientists from the Penn State Department of Engineering Science and Mechanics have developed a system that can harness energy from radio waves to power wearable devices. Radio waves permeate the environment when people consume energy but are also sources of energy themselves. The researchers’ tool is a stretchable antenna system that can convert energy from electromagnetic waves into electricity, which can then be used to power wireless devices or to charge energy storage devices, such as batteries.
8. Detecting Errors in Patients’ Medication Self-administration Methods
Researchers at the Massachusetts Institute of Technology have developed a system that can determine when a patient is using an insulin pen or inhaler and reduce self-administration errors. The system sits in the background at home, much like a Wi-Fi router, and emits low-power radio waves that reflect off a patient’s body. The system then uses AI to analyze the radio waves to identify when a patient is using a medical device and flag anomalies in any particular step, such as using an insulin pen for five seconds instead of the recommended ten seconds. When tested, the system was able to detect 96 percent of insulin pen administration and 99 percent of inhaler uses.
9. Tracking the Impact of COVID-19 of Education
Researchers at John Hopkins University in Baltimore have developed a tool that tracks the impact the COVID-19 pandemic has had on education. The tool, called the Global Education Recovery Tracker, is a collaborative effort between the university, the World Bank, and UNICEF and captures information on the status of schooling, modalities of learning, availability of remedial education support, and status of vaccines for teachers in countries around the world. Data from the tool shows that 51 countries have fully returned to in-person education as of early March 2021.
10. Creating Tools to Increase Access to Financial Sector in Brazil
Brazil’s central bank has launched a digital platform called Pix to help more unbanked Brazilians enter the financial system and increase the number of fintechs. Pix, which was launched in November, already has more than 170 million registered individuals and made 275 million transactions in February, representing 71 percent of total transfers. Together with its open-banking mandate last month, Brazil hopes to increase data-driven innovation in the financial sector.
Image credits: Pixnio.