University researchers have developed a new computer program that gauges human feelings through speech, with substantially greater accuracy than existing approaches.
The program analyzes 12 features of speech, such as pitch and volume, to identify one of six emotions from a sound recording. And it achieves 81 percent accuracy—a significant improvement on earlier studies that achieved only about 55 percent accuracy.
The research has already been used to develop a prototype of an app that displays either a happy or sad face after it records and analyzes the user’s voice. It was built by graduate student Na Yang during a summer internship at Microsoft Research.
“The research is still in its early days,” says Wendi Heinzelman, professor of electrical and computer engineering and dean of graduate studies for Arts, Sciences & Engineering. “But it is easy to envision a more complex app that could use this technology for everything from adjusting the colors displayed on your mobile device to playing music fitting to how you’re feeling after recording your voice.”
Heinzelman and her team are collaborating with University psychologists Melissa Sturge-Apple and Patrick Davies, who are currently studying the interactions between teenagers and their parents. “A reliable way of categorizing emotions could be very useful in our research,” Sturge-Apple says. “It would mean that a researcher doesn’t have to listen to the conversations and manually input the emotion of different people at different stages.”
Learn more at www.ece.rochester.edu/projects/wcng/project_bridge.html.
A Medical Center analysis of more than 5,300 patients followed for eight months during treatment of spinal disorders showed that cigarette smokers reported far more pain than never-smokers or those who had quit.
Smoking cessation either prior to treatment or during the course of care was related to significant improvements in pain—a result that underlines the need for structured stop-smoking programs among the legions of patients who experience back pain due to degenerative disease, deformity, or musculoskeletal problems, says Caleb Behrend, chief resident in the Department of Orthopaedics and Rehabilitation.
Glenn Rechtine, a nationally recognized spinal surgeon and an adjunct faculty member at the Medical Center, led the study, which was published in the Journal of Bone and Joint Surgery.
Nearly all people will experience back pain at some point in their lives, and many will seek medical care. Because the cost of care and lost productivity for patients of spinal disorders is so great, researchers wanted to find out if improvements in pain could be achieved with a cost-effective intervention such as smoking cessation. Read more at www.urmc.rochester.edu/news.
A computer model of the heart wall has demonstrated the ability to predict risk of irregular heart rhythms and sudden cardiac death in patients, paving the way for the use of more complex cardiac models to calculate the consequences of genetic, lifestyle, and other changes to the heart.
Authors of the new study, published in the Journal of the American College of Cardiology, say this is the first report of cardiac modeling being used as an arrhythmic risk predictor for patients.
“This is a strong proof-of-principle study showing that computer simulation can be used to predict risk of cardiac arrhythmias,” says Coeli Lopes, lead study author and assistant professor at the Aab Cardiovascular Research Institute at the Medical Center. “With this model we can determine the influence of a single mutation on the much bigger overall response of the heart.”
Researchers plan on using IBM’s supercomputer, the Blue Gene/Q, to develop a more sophisticated model of the human heart.
They hope to predict the effects of new drugs on the electrical activity of the heart—one of the most challenging hurdles in the development of new drugs and an extremely important part of keeping potentially dangerous drugs off the market.
Computer models of the human heart have been in development since the 1960s, but the authors believe they are the first to publish a study demonstrating that they can be used to generate clinically relevant information that predicts patient risk. Read more at www.urmc.rochester.edu/news.
“Women going through menopausal transition have long complained of cognitive difficulties such as keeping track of information and struggling with mental tasks that would have otherwise been routine,” says Miriam Weber, assistant professor of neurology and lead author of the study. “This study suggests that these problems not only exist but become most evident in women in the first year following their final menstrual period.”
Study participants were assessed with a comprehensive battery of tests to evaluate a variety of cognitive skills. These included tests of attention, verbal learning and memory, fine motor skills and dexterity, and “working memory”—or the ability to not only take in and store new information but also manipulate it.
These tests are similar to daily tasks such as staying focused on something for a period of time, learning a new telephone number, and making a mental list of groceries and then recalling specific items as while walking the aisles of a grocery store.
The researchers found that women in the early stage of post menopause performed worse on measures of verbal learning, verbal memory, and fine motor skills than women in the late reproductive and late transition stages.www.urmc.rochester.edu/news
A study published in the Journal of Nutrition adds to the growing scientific evidence that when expectant mothers eat fish often, they are giving their future children a boost in brain development even though they are exposing their children to a neurotoxin, methyl mercury, present in fish.
The Food and Drug Administration advises pregnant woman to eat only two meals of fish a week to reduce the exposure of their babies’ developing brains to mercury. However, a recent joint report from the World Health Organization and the Food and Agriculture Organization of the United Nations recommended nations actually emphasize the benefits of eating fish for pregnant women and nursing mothers and the potential risks of not consuming fish to brain development.
Women in this new study, which was conducted in the Republic of Seychelles in the Indian Ocean, eat an average of 10 times as much fish as mothers in the U.S. In the study, children’s standard language development scores rose as levels of omega-3 fatty acids rose in mothers. The nutrients are important building blocks in the brain and are present in large amounts in fish. Fish are the primary dietary source of many of the fatty acids that play a crucial role in brain development.
Read more at www.urmc.rochester.edu/news.
A new study in the journal Plos One indicates that cosmic radiation—which would bombard astronauts on deep space missions to places like Mars—could accelerate the onset of Alzheimer’s disease.
“Galactic cosmic radiation poses a significant threat to future astronauts,” says M. Kerry O’Banion, professor of neurobiology and anatomy and the senior author of the study. “The possibility that radiation exposure in space may give rise to health problems such as cancer has long been recognized. However, the Medical Center study shows for the first time that exposure to radiation levels equivalent to a mission to Mars could produce cognitive problems and speed up changes in the brain that are associated with Alzheimer’s disease.”
While space is full of radiation, Earth’s magnetic field generally protects the planet and people in low Earth orbit from these particles. However, once astronauts leave orbit, they are exposed to a constant shower of radioactive particles. With appropriate warning, astronauts can be shielded from dangerous radiation associated with solar flares. But there are also other forms of cosmic radiation that, for all intents and purposes, cannot be effectively blocked.
Because this radiation exists in low levels, the longer an astronaut is in deep space, the greater the exposure. This is a concern for NASA as the agency is planning manned missions to a distant asteroid in 2021 and to Mars in 2035.
Read more at www.urmc.rochester.edu/news.