Tag: MIT

Robot outperforms highly-skilled human surgeons on pig GI surgery

On By A.P.In Medicine, Robot, surgery, Sutures, The Future, The SingularityLeave a comment

A robot surgeon has been taught to perform a delicate procedure—stitching soft tissue together with a needle and thread—more precisely and reliably than even the best human doctor.

The Smart Tissue Autonomous Robot (STAR), developed by researchers at Children’s National Health System in Washington, D.C., uses an advanced 3-D imaging system and very precise force sensing to apply stitches with submillimeter precision. The system was designed to copy state-of-the art surgical practice, but in tests involving living pigs, it proved capable of outperforming its teachers.

Currently, most surgical robots are controlled remotely, and no automated surgical system has been used to manipulate soft tissue. So the work, described today in the journal Science Translational Medicine, shows the potential for automated surgical tools to improve patient outcomes. More than 45 million soft-tissue surgeries are performed in the U.S. each year. Examples include hernia operations and repairs of torn muscles.

“Imagine that you need a surgery, or your loved one needs a surgery,” says Peter Kim, a pediatric surgeon at Children’s National, who led the work. “Wouldn’t it be critical to have the best surgeon and the best surgical techniques available?”

Kim does not see the technology replacing human surgeons. He explains that a surgeon still oversees the robot’s work and will take over in an emergency, such as unexpected bleeding.

“Even though we take pride in our craft of doing surgical procedures, to have a machine or tool that works with us in ensuring better outcome safety and reducing complications—[there] would be a tremendous benefit,” Kim says. The new system is an impressive example of a robot performing delicate manipulation. If robots can master human-level dexterity, they could conceivably take on many more tasks and jobs.

STAR consists of an industrial robot equipped with several custom-made components. The researchers developed a force-sensitive device for suturing and, most important, a near-infrared camera capable of imaging soft tissue in detail when fluorescent markers are injected.

“It’s an important result,” says Ken Goldberg, a professor at UC Berkeley who is also developing robotic surgical systems. “The innovation in 3-D sensing is particularly interesting.”

Goldberg’s team is developed surgical robots that could be more flexible than STAR because instead of being manually programmed, they can learn automatically by observing expert surgeons. “Copying the skill of experts is really the next step here,” he says.

https://www.technologyreview.com/s/601378/nimble-fingered-robot-outperforms-the-best-human-surgeons/

Thanks to Kebmodee for bringing this to the It’s Interesting community.

New typing test may help diagnose Parkinson’s disease

On By A.P.In Medicine, Neuropsychiatric Disease, Parkinson's diseaseLeave a comment

Whether it’s on a keyboard, a smartphone, or even a credit card reader, you spend a lot of your day typing. Well, researchers at MIT noticed the value of this daily habit, and are putting it to a secondary use; they’ve developed software that can gauge the speed at which a typist is tapping the keyboard to help diagnose Parkinson’s disease.

In order to type a word, your brain has to send signals down through your spinal cord to the nerves that operate your fingers. If your central nervous system is functioning perfectly, then you should be able to tap most of the keys at a fairly constant rate. But a number of conditions might slow the signal from the brain to the fingers, such as sleep deprivation (which slows all motor skills) and diseases that affect the central nervous system, including Parkinson’s.

For the first version of this study, the researchers were looking at typing patterns that indicated whether a person was sleep-deprived or well rested. They created a browser plug-in that detected the timing at which the volunteers hit they keys and found that the people who were sleepy had a much wider variation in their typing speed. They found similar results in their preliminary test with Parkinson’s patients; the 21 typists with Parkinson’s tapped the keys at much more variable rates than the 15 healthy volunteers. The researchers called it a “window into the brain.”

Right now, the algorithm they’ve developed is not refined enough to distinguish Parkinson’s patients from people who are sleep deprived, though the results might be clearer after a number of trials. The researchers plan to conduct a study with a larger group of subjects, but they hope that this type test could eventually lead to earlier diagnoses of Parkinson’s–today most people are diagnosed after they have had symptoms for 5-10 years–and to distinguish Parkinson’s from other conditions that might affect a person’s motor skills, like rheumatoid arthritis. They are currently developing a smartphone app that can test participants even more easily.

http://www.popsci.com/type-test-diagnose-parkinsons