Washington, September 14: The need for a better way to objectively measure the presence or absence of pain rather than relying on patient self-reporting has long been an elusive goal in medicine.
But now, with advances in imaging techniques, researchers, including one of Indian origin Stanford University School of Medicine trained a computer algorithm for magnetic resonance imaging (MRI) data from the brains interpret and if in pain to determine.
Scientists took eight subjects and put them in the brain scanning machine. A thermal probe was then applied to their forearms, causing moderate pain. The brains both with and without pain patterns were then recorded and interpreted by sophisticated computer algorithms to model what the pain seems to create. The process was repeated with another group of eight people.
The idea was to train a linear support vector machine - a computer algorithm invented in 1995 - a set of individuals, and use computer models to accurately classify pain into a whole new set of individuals.
The computer was then asked to brain scans of eight new items, and consider whether they could determine thermal pain.
"We have asked the computer to come up with what they think the pain looks," said Neil Chatterjee, currently an MD / PhD student at Northwestern University.
"Then we can measure how well the computer did it." And it did surprisingly well. The computer was successful 81 percent of the time.
The study was published recently in the online journal PLoS ONE. (ANI)
Posts
