My Evolution over 29 Years with Prosthetic Hands
According to statistics from the Ministry of Civil Affairs, there are 85 million people living with disabilities in China, among whom there are 24 million who suffer from extremity disabilities. People with prosthetic limbs fitted to their bodies usually pay for them with work injury compensation or insurance payouts from at-fault parties in car accidents, while some are covered at their own expense. An average person can hardly afford to purchase a myoelectric degree-of-freedom prosthetic limb, the prices of which start in the hundreds of thousands. Furthermore, medical insurance is yet to cover prosthetic limbs.
In 2016, a team of high-caliber students from several reputable universities in China and the U.S. began paying close attention to the issue, and have been working since to help Chinese with extremity disabilities get access to both cheap and comfortable intelligent prosthetic hands that incorporate a number of powerful functions.
Now, after three years of hard work, their technological achievement is being commercialized: the BrainRobotics intelligent and flexible myoelectric prosthetic hand, the world's first 8-electrode commercial prosthetic hand, is expected to hit the market at the end of 2019. Users who have extremity disabilities will be able to use it to control every single finger of the prosthetic hand, making precise and total control of their prosthetic hands by the use of their brains a reality.
In the first part of this special, in-depth article for TMTPOST Image, we are going to introduce the myoelectric prosthetic arm technology, as well as the tech firm BrainRobotics. In the second part, we will tell the story of a fitness coach who lost both of his hands, and how he accomplished his amazing 'evolution', at the same time bearing witness to the adoption of a new technology.
**The Evolution of BrainRobotics
**
24 April, 2019. In his office in Shenzhen's High-Tech Zone, Huang Qi, team leader in charge of algorithm and mass production at the intelligent prosthetic limb project BrainRobotics, repairs a myoelectric prosthetic hand for display.
Headquartered in Boston, BrainRobotics' team consists of a dozen Ph.Ds and master's graduates in the fields of brain science, mechanical engineering and human-machine interaction.
Huang Qi is 30 years old and completed his Ph.D. at the National Key Laboratory of Robotics Technologies and Systems of the Robotics Research Institute at the Harbin Institute of Technology in 2018, devoting the entirety of his master's and Ph.D. degrees to the research of intelligent prosthetic hands. His master's thesis topic was on the myoelectric control and sensory feedback of bionic prosthetic hands, while his doctoral research centered on two-way interactive control of bionic prosthetic hands based on adaptive learning.
Huang Qi was first exposed to intelligent prosthetic hands during his master's study; he found the topic 'interesting', particularly the concept that such prosthetic hands could be controlled by electrodes attached to the arm. In February 2012, following his first contact with a patient with extremity disabilities, his view of the intelligent prosthetic limb industry changed from merely 'interesting' to 'relevant'.
"I brought the prosthetic hand from the lab to the Heilongjiang Recovery Assistive Products Center to look for patients to carry out testing," The first patient Huang Qi contacted was a man in his fifties who had lost all of his hands and feet as a result of a workplace injury. "I fitted the prosthetic hand to his arm, then guided him to recall his muscle movements and collect his muscle signals. Before long, he was able to grab a glass on the table and drink from it," Huang Qi recollects. He was overwhelmed at the time. Afterwards, Huang Qi met with more than 20 patients with extremity disabilities, in the process becoming more determined than ever to pursue a career in intelligent prosthetic limb research.
Before finishing his Ph.D., Huang Qi was faced with the choice of either joining a research institute or entering the corporate world. In February 2018, Founder of BrainCo & BrainRobotics, Dr. Han Bizheng of Harvard's Brian Science Center, visited Harbin to attend the annual Yabuli China Entrepreneurs Forum while paying a special visit to Huang Qi at the Harbin Institute of Technology, where the two exchanged views on intelligent prosthesis and its industrial applications. Han told Huang that the founding goal of BrainRobotics was to help more people with disabilities gain access to cheap and easy-to-use myoelectric prosthetic hands. Huang Qi was moved and decided to join the BrainRobotics team.
"Something that is both meaningful and in demand must have great commercial
prospects," stated Huang Qi to TMTPOST Image. Just as Han Bizheng, Huang Qi is
hoping that adoption of the technology will benefit more people with
disabilities and change their lives.
24 April, Shenzhen. Huang Qi demonstrates to TMTPOST Image how the prosthetic hand is controlled using myoelectric signals. This is a third generation BrianRobotics model. At the time, the fourth generation model was already being tested on the production line for trial mass production.
Prosthetic hands can be controlled by visual, electroencephalography(EEG) and myoelectric controls. Among these three types, myoelectric control has become the standard of the industry due to its strong controllability, ease of implementation and intuitive operations.
The movements of myoelectric prosthetic hands are enabled by collecting and
processing surface myoelectric signals produced by human muscle movements. Any
person who attaches external electrodes to their limb muscles while recording
muscle signals on the back-end system can be capable of controlling this
prosthetic hand.
26 April, Shenzhen. In a BrainRobotics' OEM workshop, a myoelectric prosthetic hand is about to be assembled. This is the fourth generation model; it is also the final trial version for mass production.
It took them three years to reach their first trial mass production target of 20.
The prototype of the BrainRobotics myoelectric prosthetic hand was born in the innovation lab at Harvard. Relevant technological research was initiated in 2016. From 2016 to February 2019, with the goal of mass production, the prosthetic hand experienced four iterations, or generations. The fourth generation model went through more than 2,000 big and small bug fixes, and adjustments of parts and details, in a short period of just over two months in its journey from prototype to mass production trial version.
In the mind of Huang Qi, the 2,000-plus fixes necessary constitute the most valuable of all experiences and are inevitable for the mass production of any prosthetic hand. "Even if some other teams get hold of our prosthetic hands, they will not be able to copy it. It's just like a plane. You can buy a plane then disassemble it to see how it is built, but to build it from scratch will be very difficult."
"Many laboratories at home and abroad are conducting research on intelligent prosthetic limbs and their research has attained a certain level but is yet to go further than prototypes," said Huang Qi to TMTPOST Image. Take sensory feedback technology as an example. The touch sensing gloves developed by MIT can be worn over prosthetic hands and used to identify objects by touch. "Many products such as this are still not yet ready to be mass produced."
In research, prototypes are developed to achieve certain functions in R&D; aspects such as reliability, cost or production difficulties are rarely taken into consideration. However, when it comes to mass production, all factors must be considered and a balance struck between different needs. This sometimes dictates the sacrifice of a number of cutting-edge functionalities in the prototype.
Huang Qi says that the trial mass production represents just the beginning; they plan to expand the mass production further in July. The latest prosthetic hand weighs less than 500 grams, has a maximum grip strength of 60 Newton (which converts to 3 kilograms-force) and is link rod driven.
Currently, the world's two most widely known and used multi-fingered, flexible myoelectric prosthetic hands are the Bebionic prosthetic hand by German maker Otto Bock (95 to 539 grams; maximum grip strength 75 Newton) and i-Limb from British bionic firm Touch Bionics (50 to 615 grams, maximum grip strength 136 Newton). Otto Bock was established in 1919. Touch Bionics was founded in 2005, then acquired in 2006 by famous prosthetic limb maker and industry giant Össur.
BrainRobotics' product, which is also link rod driven, is on par in terms of its weight-to-strength balance with both above-mentioned myoelectric prosthetic hands which have been on the market for years.
In terms of degrees of freedom and number of movements, Huang Qi says BrainRobotics' product has six degrees of freedom and is able to make 23 movements. It can be used to lift an object weighing up to 8 kilograms.
Degree of freedom can be understood as the movement of a hand. The contraction and extension of a finger is one degree of freedom. More degrees of freedom mean higher flexibility. A normal person's hand has 26 degrees of freedom. BrainRobotics' hand has a single-stroke speed of 0.8 seconds per one degree of freedom with just 300 milliseconds delay between the myoelectric signal and motion of the prosthetic.
The Bebionic and i-Limb hands hold their respective advantages. The Bebionic includes 5 degrees of freedom and can achieve 14 movement modes, while the i-Limb has 6 degrees of freedom with 10 predefined hand gestures; users can also self-define their own gestures using the i-Limb App.
In terms of price, BrainRobotics' prosthetic hand is a lot more affordable
than both the Bebionic and i-Limb; the sales price on the Chinese mainland
ranges from 300,000 to 600,000 RMB. "Our product is a lot cheaper than theirs
even though all products are smart, flexible prosthetic hands that enable the
individual control of every single finger," Huang Qi said to TMTPOST Image.
Ni Mincheng, a newly recruited prosthetic hand tester, aged 39, originates from the prefecture-level city of Jinhua in China's southeastern Zhejiang province. At the age of just 10, he lost both of his hands in an accident that resulted from explosives set off by villagers, originally intended to kill wild boars. On 22 May this year, in Shanghai, Ni Mincheng had gypsum mold cast to his arm to obtain an accurate limb model in order to prepare the cavity for a new prosthetic hand.
The cavity is an indispensable component in connecting the residual limb with the prosthetic hand. For technology companies, the cavity is an area of expertise that is at once both well-known and yet unfamiliar. This is a field that necessitates a crucially high level of experience and technical expertise.
The cavity consists of both an internal and external cavity. The internal cavity fits directly onto the residual limb, while the wall of the internal cavity integrates the electrodes that read the myoelectric signals. On top of the internal cavity sits the external cavity wherein is contained other components such as the battery and switch.
The human body is elastic and, as a result, distorts when impacted by external
impact or stress. Therefore, the extent to, and manner in which, the skin of a
residual limb adheres to the electrodes become important factors that will
affect the level of comfort of the prosthetic hand and the accuracy of its
signal readings.
A finished internal cavity. Eight electrodes are mounted onto the cavity and act as sensors that collect myoelectric signals.
Currently, mass-produced myoelectric prosthetic hands have only two electrodes - therefore, two channels. These two electrodes are usually attached to the location of a pair of muscle flexors and extenders inside of the cavity to recognize a user's intention of motion. A two-channel prosthetic hand can identify four separate kinds of movement: relax, flex, extend and co-contract.
The myoelectric information readable by the prosthetic hand is limited under the scenario of only two electrodes. In order to make more movements a reality, it is necessary to group the movements. The user is then able to switch back and forth between different groups either by pressing buttons on the prosthetic hand or by applying a corresponding muscle action.
"In other words, the movement made by the prosthetic hand may not be what the user imagined. It's not like whatever the residual limb is doing, the prosthetic hand will simply follow." Huang Qi explained to TMTPOST Image. If there is no corresponding relationship between the intention of motion and actual movement, a patient will find it difficult to treat the prosthetic hand as a part of his/her own body, rather they will feel it more like a machine.
"We have produced the world's first commercial eight-electrode prosthetic hand," said Huang Qi, who added that eight-electrode prosthetic hands are capable of acquiring far more precise and detailed information than two- electrode ones. Users can directly control every single finger without the need to switch between groups for more movements when myoelectric signals are insufficient which they must do with a two-electrode prosthetic hand.
So why eight electrodes, but not nine or seven? In 2011, Erik Scheme, an
authoritative Canadian scholar in the myoelectric field, pointed out in a
thesis that the statistically drawn curve between the number of electrodes and
recognition effect when identifying finger movements indicated that a
prosthetic hand with eight electrodes performs better. His conclusion has
become one of the most important points of reference in the field of
myoelectric prosthetic limb research. When doing his Ph.D. at the Robotics
Research Institute of the Harbin Institute of Technology, Huang Qi and his
laboratory colleagues focused on 8-electrode prosthetic hands.
Ni Mincheng puts on the cavity for the recording of myoelectric signals to test the prosthetic hand. The eight curves on the computer screen represent the myoelectric signals read by the eight electrodes.
In comparison to the two-electrode prosthetic hands chosen by mainstream prosthetic limb manufacturers, an eight-electrode hand means a higher production cost of the cavity, and demand more complex algorithms, higher ability of generalization and more computing power. "A better-performing chip than the one used for two-electrode prosthetic hands is needed."
The algorithm is the core. BrainRobotics' algorithm engineers in Boston spent two years to achieve the final formulation of the world's first deep learning myoelectric signal identification algorithm integrated into an embedded system. Huang Qi told TMTPOST Image that the "highly robust" algorithm can identify the movements of every finger of a person with a disability. Furthermore, he says, once a new user undergoes his or her first training session to record myoelectric signals, no more training is required.
In order to improve the accuracy of the collection of myoelectric signals, it is important to ensure that the core algorithm is robust and the electrodes inside the cavity adhere well to the skin. In addition, the electrode must have excellent electrochemical stability and resistance to corrosion.
"We will continue to improve until we find a best solution." In Huang Qi's opinion, there has not been a single myoelectric prosthetic product so far that successfully combines comfort, functionality and affordability. This integration is what they are pursuing.
Evolution of a Paralympic Veteran
As a boy, Ni Mincheng often practiced calligraphy with his grandfather. "Every Chinese New Year, my grandpa would write couplets for the villagers and I learned by watching him." After he lost both of his hands, he practiced calligraphy with the use of only his arms for a short period of time in high school, however never picked up a calligraphy brush since.
In 2017, when BrainRobotics founder Han Bizheng visited Ni Mincheng and
introduced to him a prosthetic hand, he asked him what he would most like to
do if he had the use of a hand. The word 'calligraphy' burst from his lips.
Ni Mincheng practices calligraphy.
Ni Mincheng has loved sport since he was very young. In high school, he was selected as a member of the Zhejiang provincial disabled sports team on account of his outstanding talent, and even once won the 400-meter dash in China's National Paralympic Games.
In 2007, due to his age (27), Ni Mincheng retired from track and field competition. However, in 2008, he returned to sports when he became a member of Zhejiang's first ever Paralympic bicycle team. In August 2011, having retired from the bicycle team, Ni Mincheng maintained his passion for sports by taking over a Giant bicycle franchise, running the business until 2016.
"In 2016 I went to a bicycle expo and realized that the market had cooled," recalls Ni Mincheng. "So I sold the shop." He then came up with the idea of learning to be a fitness instructor and opening a gym. In pursuit of this goal, he did 3,000 pull-ups and 10,000 stomach crunches every day. In less than a year of self-training, he felt fit enough to travel to Beijing and attend a fitness instructor college there. In 2017, he joined the national bodybuilding and fitness team to train for a semester, working on all of his muscles.
During those two years, besides bodybuilding, Ni Mincheng also diligently
studied the art of public speaking. In 2018, Ni Mincheng opened his own
fitness studio in Jinhua, launching his career as a personal trainer and
spending all his days, in his own words, "instructing, training and reading".
In June 2019, he expanded the size of his studio from a few dozen square
meters to several hundred.
Shanghai. Taking a moment of rest from fitting the cavity, Ni Mincheng reads a book.
Since losing his hands at the age of 10, Ni was forced to face the momentous pressure of people's strange looks and their gossip. He dared not go to crowded places, nor to wear short sleeves during the summer.
"In all the years I spent growing up since losing my hands, the negativity and shocks I received every day from people made me feel depressed inside but also made me headstrong," Ni Mincheng told TMTPOST Image. When he first opened his bicycle shop, customer retention was difficult due to his 'weirdness' factor, to the extent that he even found communication with his shop assistants became problematic.
The turning point came in 2016. Ni started to read and contemplate, gaining the wisdom and awareness that he should look nowhere else for the problem but in himself. He also realized that a person's attitude towards him was a reflection of their own inner state.
"For instance, when meeting people from the lower social classes, I would usually hear either words of pity or astonishment at how anyone could live without hands. In the past, it would put me in a foul mood, and I would feel only pity for myself." Later, Ni Mincheng discovered the cause of their pity. "It's because they themselves feel a sense of hopelessness in their own lives, so they cast that weakness and pity onto me instead."
Ni Mincheng said that in his life, he has discovered that those with better education or successful careers are far better at seeing the positive side of his condition and finding his strength. Such people often heap praise on him, such as "You are just awesome," or "You have made so many interesting and wonderful achievements."
"I am a mirror, a reflection of every person I meet," Ni Mincheng remarked.
Ni Mincheng has gained a lot from his collaboration with BrainRobotics. His biggest dividend has come from his experience in the U.S. city of Boston. During that time, he experienced for the first time what it was like to be a "normal person".
During that short three-month period, Ni Mincheng ran back and forth between his lodging, the gym and the office. He found that, whether on the street or in the gym, everyone treated him just as they would a normal person. "People greeted me and communicated with me just as if I was normal. No-one once gave me a strange look or thought I should be pitied or helped."
Ni Mincheng said that he was also able to see wild animals in the vicinity of where he lived. He experienced people's respect for life in a completely different social environment. As a consequence, he started to understand the respect that people provide in their communication with others.
Why are people with disabilities invisible on the street in China? Ni Mincheng knows the reason well from his own experiences: a person with disabilities has to deal with verbal confrontations that are simply beyond the comprehension of normal people.
"This is the status quo for the disabled in China. They dread going out. I
want to tell them all about my experiences and thoughts, and give them a
helping hand to step outside and face the world with a positive outlook."
20 June, Jinhua, Zhejiang Province. Ni Mincheng gives a private lesson to a member of his gym.
In the past couple of years, Ni Mincheng has fallen in love with public speaking and enrolled in a number of public speaking training courses. He has also taken part in book readings in order to share his experiences. His public speaking has increased the popularity of his gym and brought him a lot of new fans.
Having listened to his stories, narrated by Ni himself, the audience and media have labeled him an "inspiration".
"I aspire to live as a normal person and integrate into the society, becoming
one of the mass like any other," Ni Mincheng told TMTPOST Image. He does not
want to be labeled. "Labels pigeonhole people." Neither does he think he is
really inspirational. He simply believes that he does what he wants to do and
pursues what he wants to pursue, just as any normal person can.
Ni Mincheng's girlfriend is a painter. 19 June, Jinhua, Zhejiang Province. On a rainy day, Ni Mincheng visits his girlfriend's studio to help her move her green plants outside onto the balcony to be watered by the rain.
It was Ni Mincheng's strong passion for bodybuilding that propelled him to open his own gym. He believes it is the best of both worlds, since he now has a place for training on top of a profitable business. He explained that to inspire others was never part of his original intention.
On one occasion, a media reporter, having heard his story, asked Ni Mincheng how he came to be so resilient. Speechless, he felt unable to respond.
'Inspiration' and 'resilience' are labels that separate Ni Mincheng from
normal people. "People find me inspirational only because they view disabled
people differently to others. The fact is that there are many normal people
who have achieved a lot more than I have. Only because I belong to this group
of people referred to as disabled, do they consider me inspirational."
Ni Mincheng trains in his gym.
Ni Mincheng's next goal is to give a two-hour public speech to an audience of over a thousand people at the end of this year. His theme for the speech is 'Hope', and will incorporate three segments including a drama scene. He plans to showcase different groups of people that he has met and share his personal experiences.
"I'm trying to radiate some light. Although it isn't much, I will do my best
to bring hope to more people." Ni told TMTPOST Image that before the public
speech, he plans to improve his physique so to make it more impressive.
20 June, Jinhua, Zhejiang Province. Donning the BrainRobotics prosthetic hand, Ni Mincheng poses in his gym for a picture by TMTPOST Image.
Ni Mincheng has brought the prosthetic hand gifted to him by the BrainRobotics team to his hometown of Jinhua.
Ni Mincheng has witnessed the evolution of the BrainRobotics prosthetic hand, while in the process slowing becoming the "normal person" that he has always dreamed of.
-----------------------------------------------
This article was edited by @Dido Pang and David Joseph. The original Chinese version was written by Zack Chen. Please note source and hyperlink when reproduce.
Follow us on Twitter @tmtpostenglish, Medium @TMTPOST and Facebook @TMTPOST.
更多精彩内容,关注钛媒体微信号(ID:taimeiti),或者下载钛媒体App
作者暂无likerid, 赞赏暂由本网站代持,当作者有likerid后会全部转账给作者(我们会尽力而为)。Tips: Until now, everytime you want to store your article, we will help you store it in Filecoin network. In the future, you can store it in Filecoin network using your own filecoin.
Support author:
Author's Filecoin address:
Or you can use Likecoin to support author: