Within reach: Integrating robotics, biology improves functionality of prosthetic hand

Twenty-seven bones, 27 joints, more than 30 muscles and over 100 ligaments make up the human hand.

Coordinating these components with our brain and nervous system, hands are capable of doing amazing things. One thing scientists have not yet been able to do with their own brains and hands, however, is design an effective prosthesis for the thousands of people affected by upper limb loss.

“There are many (prosthetic) hands developed by research that are beautiful engineering masterpieces. The problem is, they often don’t get out of the lab,” says Marco Santello, a professor of biomedical engineering in the Ira A. Fulton Schools of Engineering at Arizona State University. “There is a huge gap between building something very elegant and sophisticated and something that someone would like to use in daily activities.”

Santello and a team of interdisciplinary researchers have been developing the SoftHand Pro — the first prosthetic to combine soft robotic technologies and the natural biomechanics of the human hand — to address the unmet needs of functionality, versatility and robustness in a hand prosthesis and improve people’s quality of life.

“If prosthetic users are not satisfied with how a prosthesis works, they rely on constantly using their intact limb,” Santello says. “However, there are a lot of activities we do with more than one hand. So, if they have a prosthetic hand they can rely on to function reliably and consistently, they can do (more) activities they care about.”

After years of proof-of-concept research and preliminary testing, Santello is collaborating with Kristin Zhao, director of the Assistive and Restorative Technology Laboratory at Mayo Clinic in Rochester, Minnesota, on a $3.2 million National Institutes of Health-funded R01 project to conduct a clinical trial to assess and compare the performance of the SoftHand Pro against a commercially available prosthetic hand.

“Clinical trials are a priority for us as translational researchers because they give us the opportunity to learn whether our novel technologies will meet the needs of patients,” says Zhao, the project's co-principal investigator alongside Santello. “Further, this clinical trial exemplifies a powerful team science approach that brings together teams with expertise in robotics, biomechanics, neural control of movement and clinical care for patients with limb loss.”

This research initially received one of three seed funding grants from Team Science — a partnership between ASU and Mayo Clinic in which a multidisciplinary team comprised of one ASU researcher and one Mayo Clinic researcher collaborate on innovative health care solutions — as well as additional National Institutes of Health R21 grant funding for early stage research.

Reinventing the hand

In the laboratory, hand prostheses are often very complex, require extensive maintenance or are prone to breakage. Many commercial prosthetic hands have their own drawbacks, such as their expense, weight and difficulty to repair.

“The biggest weakness is that users are often not satisfied (with commercial prostheses) because the range of functions you can perform is often limited,” says Santello, who is also the director of the School of Biological and Health Systems Engineering, one of the seven Fulton Schools.

With too few joints, a prosthetic hand is easier to control, but can’t do everything patients would like it to do. More joints increase functionality, but can make controlling the hand too complex. However, Santello says the solution needs to go beyond the prosthetic hand's mechanical structure.

During his postdoctoral work in neuroscience, Santello observed that even though the hand has many joints, biomechanical restrictions in the human hand and neural constraints in the central nervous system result in only three or four main coordination patterns of finger motion, known as synergies. These patterns, extracted from hand shapes used to grasp and use a large set of everyday objects, can be considered the fundamental “building blocks” that underlie all possible hand shapes.

This concept caught the attention of robotics researchers, including Antonio Bicchi, who is a professor of robotics at the University of Pisa, a senior scientist at the Italian Institute of Technology in Genoa, Italy, and an adjunct professor in the School of Biological and Health Systems Engineering at ASU; and Manuel G. Catalano, a researcher at IIT and an affiliate at Mayo Clinic. In his doctoral thesis, Catalano, advised by Bicchi and Santello, developed and investigated the design concept of the SoftHand Pro, which earned him international acclaim with the 2014 Georges Giralt PhD Award for the best PhD thesis in robotics.

“People who work in robotics thought this was a really cool concept because so far, we’ve been building robotic hands with as many motors as joints, which makes control very, very difficult,” Santello says. “But now, if you can design the (prosthetic) hand in a way that moves according to biological constrained patterns of finger motion, you dramatically simplify control without necessarily compromising function and versatility.”

By using the concept of hand synergies, Bicchi and his team designed and built the prototypes of what is now the SoftHand Pro.

Maria Fossati, a designer on the European SoftHand Pro team who also works at the Italian Institute of Technology in Genoa, demonstrates the SoftHand Pro at Cybathlon 2020, an international competition in which people with physical disabilities compete in events with state-of-the-art assistive technologies. Photo courtesy of IIT-Istituto Italiano di Tecnologia

Soft, robust and versatile

By combining soft robotics, biomechanics and neuroscience, the SoftHand Pro has emerged as a flexible prosthetic with 19 joints controlled by a single motor that manipulates the joints according to the natural movement patterns of the hand. Using only one motor to control so many joints is a unique feature compared to other prosthesis designs. The SoftHand Pro also has more joints, also known as degrees of freedom, than other commercially available prosthetic hands, and is closest to the degrees of freedom of the human hand.

The SoftHand Pro works with myoelectric technology that uses the electrical signals generated in the remaining tissue of a person’s forearm muscles to control the prosthetic hand. Santello says that all the user has to do is think about closing the prosthetic hand and, by contracting the forearm muscles, it will mold to an object using the concept of constrained movement patterns, not unlike what happens when using an organic hand.

“When you use your own hand, you’re not thinking that the thumb needs to flex 10 degrees and then this finger 25 degrees and the next finger five degrees — fortunately, you don’t!” Santello says. “Often, you have an approximation of the shape of the object in the shape of your hand, and you use the compliance of the hand to do the rest for you and stop closing as soon as you see or feel like you have secure contact. You don’t need a complex algorithm to choose a given hand shape for a specific object, nor does the SoftHand Pro user need to be aware (of the motion of the fingers).”

In the case of a prosthesis, the user practices this motion so it becomes second nature to see an object and grasp it with the prosthetic.

Even with such a simple control mechanism, the SoftHand Pro offers versatility for this motion. Its flexibility allows the hand to conform to many shapes and permits movements like picking a small object off of a table to taking a book from a shelf — motions other prosthetics would struggle with because of their rigidity or limitations in their mechanical design.

Qiushi Fu, one of Santello’s former doctoral students and postdoctoral researchers who is now an assistant professor at the University of Central Florida, helped conduct early testing of the SoftHand Pro with control subjects who have intact hands. He also assisted in the development of the SoftHand Pro’s bio-inspired robotic control algorithms and protocols that expanded and enhanced its capabilities.

“The SoftHand Pro prototype was a relatively new concept, and it provided opportunities to explore different directions. I was investigating bio-inspired robot control (as a doctoral student), and it helped to expand my expertise into prosthetics and human-machine interfaces,” Fu says. “Demonstrating that our control algorithm can help subjects transport fragile objects was an important achievement.”

In addition to providing increased functionality, the compliant nature of the SoftHand Pro is also more resistant to breakage. By being able to flex, extend and mold to the environment, the soft robotic mechanical joints are less likely to break during movements performed by organic joints.

Validating the design

Santello has brought the basic scientific foundation of the technology to the SoftHand Pro and worked with Bicchi and others in Italy to test the device. Now Zhao and Mayo Clinic’s biomechanical and clinical expertise will help the team evaluate if the prosthesis will lead to greater improvement in grasping and manipulation performance than a commercially available, multi-digit prosthetic hand called the i-limb by orthopedics company Össur.

“When the opportunity to work on a prosthetic project came up, Mayo Clinic was my top choice,” says Santello, who has fostered collaborations between the institution and his school for the past decade.

The ASU and Mayo Clinic team, with the support of the Italian Institute of Technology team, coordinated by Bicchi and Catalano, is now taking on a complex new phase of the project through an extensive, multi-site clinical trial that started in August 2021. They will work with 36 individuals with upper limb loss over five years to compare the SoftHand Pro to the commercial alternative. The subjects will participate in laboratory testing with both prosthetics, and use each prosthetic at home for daily activities for eight weeks.

Comparing the use of two prostheses over a long period of time in real-life scenarios outside of the lab “adds a more rigorous reality check,” Santello says. “That’s what excites us. There’s an opportunity to significantly impact quality of life” for individuals with upper limb loss who use prostheses.

This phase of research is made possible by working with Mayo Clinic in Rochester, Minnesota, which will provide access to a team of experts in prosthetics, occupational therapy and physical rehabilitation in addition to a prosthetic and amputee rehabilitation clinic, the Hanger Clinic.

“The Mayo Clinic team will leverage its integrated clinical practice, research and educational expertise to help carry out this pivotal clinical trial that has our core value at its center — to respond to the needs of the patient,” Zhao says. “We are excited to partner with Drs. Santello and Bicchi’s teams on this important effort.”

“It’s a high-risk, high-reward project,” Santello says. “We strongly believe that there is an opportunity to improve acceptance of prostheses, improve lives and enable prosthesis users to be more functional, more independent and able to do more things in a better way. We have to wait for the results, but based on what we know so far, there is a high likelihood that the outcomes of our project can make an impact.”