"Long before it's in the papers"
May 17, 2012

RETURN TO THE WORLD SCIENCE HOME PAGE

Robotic arm for the paralyzed reported to reach new level of sophistication

May 17, 2012
Courtesy of Nature
and World Science staff

A new study re­ports that two par­a­lyzed peo­ple were able to con­trol a robotic arm through brain sig­nals pick­ed up by a com­put­er sys­tem, and use it to make point-to-point reaches and grasps.

One pa­tient was able to use the de­vice, which reads elec­tri­cal nerve sig­nals in the brain through a ti­ny im­plant, to pick up a bot­tle fit­ted with a straw and drink from it.

The work, pub­lished this week in the re­search jour­nal Na­ture, il­lus­trates con­sid­erably more com­plex robotic con­trol than shown in pre­vi­ous stud­ies of this na­ture, ac­cord­ing to sci­en­tists. Re­search­ers are hop­ing to use such de­vices, called neu­ral in­ter­face sys­tems, to re­store lost arm func­tion in eve­ry­day tasks for peo­ple with pa­ral­y­sis. 

The de­vices work by trans­lat­ing brain ac­ti­vity di­rectly in­to con­trol sig­nals for as­sist­ive de­vices. Pre­vi­ous studies had shown that par­a­lyzed peo­ple could use such tech­nol­o­gy to con­trol com­put­er cur­sors. Mon­keys were al­so found to be able to con­trol a robotic arm us­ing the de­vices.

In the new work, John Donoghue of Brown Uni­vers­ity in Prov­i­dence, R.I. and col­leagues fit­ted neu­ral in­ter­face sys­tems to two pa­tients with long-term pa­ral­y­sis due to brain stem stroke. They used two types of robotic arms op­er­ated by user-driven, brain con­trol of hand move­ments. 

The sys­tem de­cod­ed reach-related brain sig­nals to con­tin­u­ously up­date an es­ti­mate of the par­ti­ci­pants’ in­tend­ed hand move­ment and turn them in­to hand ac­tions. The arms were not di­rectly at­tached to the users’ bod­ies but mount­ed on sep­a­rate sup­ports.

The team first tested abil­i­ties by ask­ing the two pa­tients to reach and grasp foam ball tar­gets. Both par­ti­ci­pants re­ported suc­cess at this task at lev­els sig­nif­i­cantly better than ex­pected by chance, the re­search­ers said, though the robotic arms did not prove as ag­ile as real arms on healthy peo­ple.

One of the pa­tients, a 58-year-old wom­an, used the sys­tem to reach and grasp a bot­tle of cof­fee, drink from it through a straw, and put it back on the ta­ble in four out of six at­tempts, Donoghue’s team re­ported. It was the first time in 14 years that the pa­tient was able to br­ing any drink­ing ves­sel to her mouth and drink from it solely from her own will, they said.

* * *

A new study reports that two paralyzed people were able to control a robotic arm through brain signals picked up by a computer system, and use it to make point-to-point reaches and graps. One patient was able to use the device, which reads electrical nerve signals in the brain through a tiny implant, to pick up a bottle fitted with a straw and drink from it. The work, published this week in the research journal Nature, illustrates considerably more complex robotic control than shown in previous studies of this nature, according to scientists. Researchers are hoping to use thes devices, called neural interface systems, to restore lost arm function in everyday tasks for people with paralysis. The devices work by translating brain activity directly into control signals for assistive devices. It had previously been shown that paralyzed people could use such technology to control computer cursors. Monkeys were also found to be able to control a robotic arm using the devices. In the new work, John Donoghue of Brown University in Providence, R.I. and colleagues fitted neural interface systems to two patients with long-term paralysis due to brain stem stroke. They used two types of robotic arms operated by user-driven, brain control of hand movements. The system decoded reach-related brain signals to continuously update an estimate of the participants’ intended hand movement and turn them into hand actions. The arms were not directly attached to the users’ bodies but mounted on separate supports. The team first tested abilities in a research setting by asking the two patients to reach and grasp foam ball targets. Both participants reported success at this task at levels which were significantly higher than expected by chance, the researchers said, though the robotic arms did not prove as agile as real arms on healthy people. One of the patients, a 58-year-old woman, used the system to reach and grasp a bottle of coffee, drink from it through a straw, and put it back on the table in four out of six attempts, Donoghue’s team reported. It was the first time in 14 years that the patient was able to bring any drinking vessel to her mouth and drink from it solely from her own will, they said.