The Emotional Computer
University of Cambridge (12/23/10)
A new University of Cambridge film explores professor Peter Robinson’s research on the role of emotions in human-computer interaction. Released on Cambridge’s YouTube channel, “The Emotional Computer” shows how emotions can be used to improve interaction between humans and computers. “We’re building emotionally intelligent computers, ones that can read my mind and know how I feel,” says Robinson, who is leading a team in the university’s computer laboratory. One system tracks features on a person’s face, calculates gestures being made and infers emotion, while a second analyzes speech intonation to infer emotions, and a third analyzes body posture and gestures. Research student Ian Davies is studying how to apply these technologies to command-and-control systems. “Even in something as simple as a car we need to know if the driver is concentrating and confused, so that we can avoid overloading him with distractions from a mobile phone, the radio, or a satellite navigation system,” Davies says. Robinson says computers also need to express emotion, and one team member is animating figures to mimic a person’s facial expression and another is experimenting with a robotic head modeled after Charles Babbage to explore empathy and rapport building.
Researchers Train Software to Help Monitor Climate Change
Penn State Live (12/22/10) Matthew Swayne
; Andrea Messer
Penn State researchers have developed software that automatically analyzes satellite data that could help monitor environmental conditions over time. The program uses probability to study satellite images and sensory data concerning ocean structures. “All of the data and information that is continually collected by satellites and sensors can cause tons of problems for scientists, who simply don’t have the time to analyze every pixel of every satellite image,” says Penn State professor James Wang. The researchers started the study by creating a database of ocean structures and the training the program to identify changes in them. They tested the technology using National Oceanographic and Atmospheric Administration data. “In almost all cases, the proposed methodology improves the accuracy rate and reduces the number of features necessary to get a good ocean structures classification,” says visiting professor Jose A. Piedra-Fernandez. The researchers say the program might uncover clues on small changes in ocean temperature that could have big effects on global climate conditions. The program uses the Bayesian system, which needs less data for learning than other probability-based decision systems, which reduces the computational cost of the system.
Smarter Systems Help Busy Doctors Remember
Northwestern University News Center (IL) (12/21/10) Marla Paul
Northwestern University researchers have developed a tool that analyzes electronic health records and alerts doctors during an exam when a patient’s care is not up to standards. The researchers say the software greatly improves physician’s performance in caring for patients with chronic conditions such as diabetes and cardiovascular disease. The program also enhances preventative methods in vaccinations and cancer and osteoporosis screenings. The new system involves a small yellow light on a computer that alerts the doctor when something is wrong with a patient’s care. The doctor clicks on the light and is shown what part of the patient’s treatment has been missed or overlooked. The system takes several standard aspects of patient care and compiles them in one location. “For this to work well, they have to view the alerts and reporting system as their personal quality improvement tools,” says Northwestern’s David Baker. During testing, the program increased the number of heart disease patients receiving cholesterol-reducing medication by 6 percent, the number of pneumonia vaccinations by 10 percent, and the number of colon cancer screenings by 5 percent. “The gains are modest, but if you are already at 90 percent and go to 94 percent, that’s important,” says Northwestern’s Stephen Persell.
Virtual Girl Can Read Nine Emotions
Business Day (New Zealand) (12/20/10) Claire Rogers
New Zealand researchers have developed Easy with Eve, an attractive avatar for teaching math to primary school students. Easy with Eve makes use of complex algorithms for detecting and responding to expressions and movements, motion-detection technology, and vision systems, and its developers believe the technology has other applications. Unitec professor Hossein Sarrafzadeh, who led the development of the avatar, is now working on Dr. Eve, a health advice application. “If you have a health question then Eve will come up on your monitor and hold a discussion with you and give you some medical advice,” Sarrafzadeh says. Unitec is assisting China Medical University in investigating the technology’s potential to detect cancer cells, and is working with Massey University on turning Eve into an intelligent sales agent that would distinguish between serious shoppers and browsers based on their facial expressions and gestures. Eve is still a few years away from being used in classrooms. “We’ve done some work on language translation so hopefully in the next couple of years Eve will be able to speak and understand different languages,” Sarrafzadeh says.
Software Improves Understanding of Mobility Problems
Economic & Social Research Council (12/16/10) Danielle Moore
; Jeanine Woolley
New software could help clinicians, health care practitioners, and design professionals better understand the mobility problems of the elderly. The software takes motion capture data and muscle strength measurements of older people performing everyday activities and generates a three-dimensional animated stick figure with visual representations at the joints showing the demands of the mechanics of movement. The visualization software is the result of a research collaboration between the Glasgow School of the Art and the University of Strathclyde, with support from the U.K. Research Councils’ New Dynamics of Aging program. Design professionals could use the visualization software to improve the design of products, landscapes, and buildings. The software also could help health care practitioners better understand mobility challenges, and help clinicians devise better approaches to assessment, diagnosis, and rehabilitation. “The visualization software is a simple yet highly effective tool to help older people and professionals explain, discuss, and address mobility problems,” says Glasgow professor Alastair Macdonald. “Better understanding of older people’s mobility can help health care professionals improve diagnosis or treatment of problems, and design professionals to adapt the way they design for older people.”
Filed under: ICT