Welcome to the September 2021 SIGCHI edition of ACM TechNews.

ACM TechNews - SIGCHI Edition is a sponsored special edition of the ACM TechNews news-briefing service focused on issues in Human Computer Interaction (HCI). This service serves as a resource for ACM-SIGCHI Members to keep abreast of the latest news in areas related to HCI and is distributed to all ACM SIGCHI members the first Tuesday of every month. ACM TechNews is a benefit of ACM membership and is distributed three times per week on Mondays, Wednesday, and Fridays to over 100,000 ACM members from over 100 countries around the world. ACM TechNews provides timely coverage of established and emerging areas of computer science, the latest trends in information technology, and related science, society, and technology news. For more information on ACM TechNews and joining the ACM, please click. The Interactions mobile app is available for free on iOS, Android, and Kindle platforms. Download it today and flip through the full-color magazine pages on your tablet or view it in a simplified low-bandwidth text mode on your phone. And be sure to check out the Interactions website, where you can access current and past articles and read the latest entries in our ever-expanding collection of blogs.

A Soft, Wearable Brain-Machine Interface
IEEE Spectrum
Payal Dhar
August 10, 2021

An international team of scientists has built a wearable brain-machine interface (BMI) that permits the wearer to control a robotic arm or wheelchair via thought. The BMI is part of a soft wireless scalp electronic system that reads and translates neural signals via electroencephalography. The Georgia Institute of Technology's Woon-Hong Yeo said the system boasts "miniaturized, imperceptible micro-needle electrodes, and flexible circuits with stretchable interconnectors" that streamline the form factor and enhance signal acquisition. Yeo also said the ultra-thin electrode needles minimize motion-related signal disruptions, while machine learning convolutional neural networks analyze and classify neural signals to identify the wearer's imagined motor activity.

Full Article

Toyota Halts All Self-Driving e-Palette Vehicles After Olympic Village Accident
Reuters
Sanjana Shivdas; Tim Kelly
August 28, 2021

Japanese automaker Toyota Motor has suspended the use of all electric-powered self-driving e-Palette transportation pods at the Tokyo Paralympic Games village, following a collision with a visually impaired athlete. In a YouTube video, Toyota CEO Akio Toyoda said the e-Palette had halted at a T junction and was about to turn under manual joystick control when it struck the athlete going at about 1 or 2 kilometers (0.2 to 1.2 miles) an hour. The e-Palette was adapted specifically for use during the Tokyo Olympic and Paralympic Games, with large doors and electric ramps so groups of athletes could board quickly. Toyoda said the accident highlighted the vehicle's problems operating under the village's special conditions, which "shows that autonomous vehicles are not yet realistic for normal roads."

Full Article

Expanding Human-Robot Collaboration in Manufacturing
Loughborough University (U.K.)
August 27, 2021

Researchers at Loughborough University in the U.K. have developed a method of training robots to detect a human's intended arm movements before they occur. The researchers trained an artificial intelligence (AI) system to identify pre-movement patterns in an electroencephalogram (EEG). The system was able to determine a human participant's impending arm movement an average of 300 milliseconds before the movement. A simulation of a human-robot collaborative scenario showed that this knowledge quickened completion time for a task by 8% to 11%, even when the EEG wrongly communicated the human's intention. Loughborough's Achim Buerkle said, "We hope this proposed technology could help towards a closer, symbiotic human-robot collaboration."

Full Article

China's Mental Health System Has Long Been Inadequate. Can AI Change That?
The Washington Post
Rebecca Tan; Alicia Chen
August 12, 2021

The Tree Hole Rescue Project uses an artificial intelligence (AI) algorithm developed by Chinese computer scientist Zhisheng Huang to scan posts on the social media site Weibo to identify users at risk of self-harm. When one is identified, the algorithm notifies Volunteers in a WeChat group, who message users, call their families or employers, or alert the police. University of Massachusetts Medical School's Xiaoduo Fan said there are concerns about the ethical use of AI as mental health tools, given the lack of regulation or transparency. Beth Israel Deaconess Medical Center's John Torous pointed out that algorithmic bias is another risk.

Full Article

Researchers Receive Best Paper Award at IoTDI for Work on Finger Motion Tracking
Pennsylvania State University News
Sarah Small
August 5, 2021

Researchers from Pennsylvania State University (Penn State) received the best paper award at the ACM/IEEE 2021 Conference on Internet of Things Design and Implementation for their paper on a new training data collection method for finger motion tracking. The paper details the ZeroNet system, which collects training data for performing inferences on wearable devices through transfer learning and domain adaptation techniques using publicly available videos. Penn State's Yilin Liu said that while finger motion tracking with cameras is very mature, “there is a dearth of training data for developing robust ML models for wearable sensors.” In response, the researchers created ZeroNet, which Liu described as “a system that shows the feasibility of developing ML models for wearable sensors with zero training overhead."

Full Article

Artificial Pancreas May Soon Help Patients with Type 2 Diabetes
News-Medical Life Sciences
Emily Henderson
August 4, 2021

A team from the University of Cambridge and Cambridge University Hospitals NHS Foundation Trust, working with researchers at Switzerland’s Bern University Hospital and University of Bern, have demonstrated an artificial pancreas’ ability to help patients with kidney failure manage their blood sugar levels and reduce the risk of low blood sugar. The system, which is worn externally, features a glucose sensor and an insulin pump, as well as an algorithm to calculate insulin dosages. Software in the user’s smartphone signals the pump when an adjustment to his/her insulin levels is needed; the glucose monitor calculates blood sugar levels and transmits them to the phone as well. Roman Hovorka at Cambridge’s Wellcome Trust-MRC Institute of Metabolic Science said in addition to managing their blood sugar levels, the artificial pancreas also “gave the users peace of mind.”

Full Article

Researchers Develop Bionic Arm That Restores Natural Behaviors in Patients with Upper Limb Amputations
Cleveland Clinic Newsroom
September 1, 2021

Researchers at the Cleveland Clinic, working with colleagues from Canada's universities of Alberta and New Brunswick, have built a bionic arm for patients with upper-limb amputations that lets wearers think, behave, and function naturally. The team integrated into the arm intuitive motor control, touch and grip kinesthesia, and the sensation of opening and closing the hand. The system is the first to test all three sensory and motor functions concurrently via a neural-machine interface in an artificial limb. The interface links to the wearer's limb nerves, where it sends impulses from the brain to the arm when they want to use or move it, and receives and relays physical environmental information back to the brain.

Full Article

'Likes,' 'Shares' Teach People to Express More Outrage Online
YaleNews
Bill Hathaway
August 13, 2021

A study by Yale University researchers found social media platforms magnify expressions of moral outrage because such language earns more "likes" and "shares." The researchers designed machine learning software to quantify such expressions on Twitter during actual inciting events; they found users who got more likes and retweets in response to tweets of outrage were more likely to express outrage in later posts. The team then analyzed behaviors of subjects in controlled experiments to demonstrate social media algorithms' ability to incentivize the articulation of outrage. The results indicated while people in politically extreme networks tend to express more outrage than those in politically moderate networks, social feedback tends to wield greater influence on moderates.

Full Article

Japanese Space Avatar Technology Will Allow Remote Space Tours
Interesting Engineering
Chris Young
August 31, 2021

The Space Avatar Project aims to advance avatar technology from Japanese space agency JAXA and avatar platform developer avatarin. JAXA said the project will further develop technology demonstrated at the International Space Station (ISS) last year, in which a publicly accessible space avatar was employed to communicate with astronauts aboard the Japanese experiment module. The partners said they will collaborate with University of Tokyo engineers to enable "remote space travel, remote work assistance, and remote space-themed experiences." Remote work assistance will allow astronauts to perform tasks on space stations like manipulating robotic arms and machines via an avatar; remote space travel will involve communication through an avatar screen controlled on Earth, which users can move freely throughout the ISS.

Full Article

Pianists Fitted with Robotic Thumb Can Learn to Play with 11 Digits
New Scientist
Jason Arunn Murugesu
August 7, 2021

A robotic "third thumb" developed by researchers at Imperial College London in the U.K. could help improve a user's fine motor skills in unfamiliar activities, like playing the piano. The robotic thumb is strapped to the user's hand next to the little finger and controlled by electrical signals generated when they wiggle their foot. The researchers tested the thumb on six people who play piano and six who do not, assessing their general timing and dexterity skills and then having them play simple piano pieces on an electric keyboard. They found that all the test subjects could play the keyboard using the thumb within an hour, and their effectiveness was predicted not by their prior piano-playing experience, but their timing and dexterity.

Full Article

App Helps Relieve Back Pain
Norwegian SciTech News
Steinar Brandslet
August 3, 2021

Researchers at the Norwegian University of Science and Technology (NTNU), working with colleagues at the U.K.’s University of Glasgow and Robert Gordon University and Denmark’s University of Southern Denmark, National Research Center for the Working Environment, and Trade Expansion, as well as the Netherlands’ Health Leads agency, developed the selfBACK app to facilitate self-management of back pain. The app advises users of actions they can take to deal with back pain, including physical activity and exercises. The researchers studied 461 patients with back pain from Norway and Denmark, half of whom received standard treatment from their health providers and the other half receiving standard treatment and also using selfBACK. More than half (52%) of participants using the app saw major improvement in their back pain after three months, compared to 39% in the control group. Similar differences were recorded after six and nine months.

Full Article

Schools Look for Help from AI Teacher's Assistants
The Wall Street Journal
Sarah Trent; Aydali Campa
August 6, 2021

Researchers are developing education products that leverage artificial intelligence (AI) to incorporate the performance of multiple functions into a single platform. Such tools could serve as peer learning partners, group facilitators, and teacher's assistants that can be personalized for individual students. Columbia University's Lalitha Vasudevan said AI tools could identify student behavior and performance patterns that could be missed by human teachers. Vasudevan explained that students could be given more opportunities to be seen, as AI tools could assess multiple communication cues, social dynamics, and academic performance. Computer scientists at Brandeis University and Colorado State University have developed Diana, a teaching assistant that responds to students’ nonverbal and visual cues and engages them in conversation, or prompts the teacher when students need help or are distracted. Other tools under development include a teacher's assistant chatbot at the Georgia Institute of Technology, and Columbia's ProJo, which aims to help students identify math and science errors.

Full Article

Robots Who Goof: Can We Trust Them Again?
University of Michigan News
August 10, 2021

University of Michigan (U-M) researchers assessed multiple strategies by robots to reestablish trust with human co-workers after committing errors. The researchers found some approaches are more effective than others, and often are influenced by a robot's appearance. They enlisted 164 participants to work with either a humanlike or a more mechanical-looking robot as they virtually loaded boxes onto a conveyor belt; the robots were programmed to intentionally make mistakes and either apologize, deny their error, explain it, or promise to avoid it in the future. Humanlike robots more easily repaired trust when providing explanations, and co-workers were more benevolent when offered apologies, denials, and explanations. U-M's Lionel Robert said, "Robots are definitely a technology but their interactions with humans are social, and we must account for these social interactions if we hope to have humans comfortably trust and rely on their robot co-workers."

Full Article