Abstract
Take a moment to consider how different life would be without your thumb. Imagine trying to pick up a pencil or tie your shoes. It turns out the thumb is credited with 40% of the hand’s functionality, and the loss of a thumb is equivalent to up to a 23% disability of an individual as a whole, according to official WPI (Whole Person Impairment) charts. Additionally, 45,000 partial hand and finger amputations take place in the United States alone, yet many prosthetics are either subpar in functionality or simply too expensive. Many prosthetic thumbs fail to replicate the CMC joint, which is responsible for making the thumb opposable. Those that do have full-functioning thumbs, namely myoelectric hands, range from $20,000-$100,000 or more, which most amputees simply can’t afford. Even more, functional prosthetic fingers for victims of partial-hand loss can cost around $15,000 per finger. The goal of OtherHand is to create a prosthetic that is not only functional but cost-effective for people with partial-hand loss.
Problem Statement
Despite 45,000 partial hand and finger amputations taking place in the United States each year, many prosthetics leave amputees without necessary dexterity, and the ones that do range from $20,000 to $100,000, making them inaccessible for the average amputee. OtherHand aims to provide partial-hand amputees around the world with a functional and comfortable prosthetic, giving them back their dexterity at a lower cost of production than alternatives.
Design Criteria
Based on our vision, we created actionable design criteria to keep our ideas and solutions connected to the original problem and to make sure that our future tests proved the usability of the device.
- Functionality
- Fingers fully open and close 90 degrees at each joint
- Thumb moves on multiple axes and can touch the base of the pinky
- Can hold 1 kg for 10 seconds
- Comfort
- Adjustable socket
- Material is compatible with skin
- User-defined scale
- Durability
- Materials that can stretch and bend where necessary
- Finger can be bent back and forth 100 times without damage
- Natural look
- Skin-toned components
- User-defined scale
- Cheap and accessible
- Below $250
- Open source CAD files
Ideation
To get started, we wrote all our design ideas onto sticky notes and organized them into three categories, form, feature, and function. We drew sketches of possible joints, listed possible materials, and added possible features such as a fingernail for prying stuff open.
ChatGPT’s AI played a minor role as we continued our research and ideation. We gave prompts asking about issues with current prosthetics and solutions that make use of engineering to improve them.
Design Alternatives and Screening
In order to create a final design that we could then move on to prototyping, we used decision matrices to rank ideas based on how well they fit our design criteria. We repeated this for different categories of our design to create a more concrete idea of what we were going to make, what we were going to make it out of, and what it would do.
Example of Decision Matrix:
For the materials to use, we listed out:
- Skin Color Filament
- Resin 3d print
- Carbon 3d print
- 3d printed dragon skin
- Flexible filament 3d print
- Full aluminum
- Silicon socket
- Fabric socket
- Grippy fingertips
- Elastic fabric covering
And we ended up with this list to work with:
- Carbon 3d print
- Flexible filament 3d print
- Silicon socket
- Grippy fingertips
- Elastic fabric covering
Our Solution
After narrowing down our ideas, this is what we decided on:
- Two-part design: Thumb and finger
- Thumb
- Electrically controlled by Myoware EMG (electromyography) sensors, which sense electrical activity produced by muscles
- 3D printed from flexible TPU filament and hard PLA filament
- Silicone sock to fit to hand
- Finger
- Body-powered; controlled by remnants of finger and linkage design
- 3D printed from PLA filament
- Attached to hand by wristband
- Thumb
- Dual-option business model:
- Manufactured and given to people in need
- Made at home by people with access to 3D printing
For the thumb design, we wanted to make a prosthetic thumb for people missing their thumb from its base; at or near the CMC joint. These people are greatly limited in their dexterity and ability to grab things, and many prosthetics for them lack the multi-directional movement of the thumb.
We decided for it to be electrically powered with a muscle sensor on account of the complexity of the thumb movements, which cannot be easily replicated by a body-powered prosthetic that is elbow or shoulder actuated, for instance.
For the finger design, we wanted to make a prosthetic finger forpeople missing their finger from the proximal phalange, since partial amputations of the finger are more common than full amputations. This design would be able to be duplicated for the other fingers.
We decided for it to be body-powered to keep costs down and because it is simple to do with a portion of the finger still remaining.
Prototyping - Thumb
Our first 3 prints; 2 misprinted
For the thumb, we started out by designing a simple thumb with 2 joints that would then be printed out of flexible TPU filament. After we printed it and tested it, we moved onto creating the thumb model, going through many different iterations until we ended up with our final design including 3 joints, a base to attach it to the servo, a hole for the tendon, and holes along the side for structural support pieces printed out of non-flexible PLA to prevent it from bending unnaturally. The thumb would bend when the tendon is pulled and straighten back due to the flexible material.
For the electronics, we went with a design that included a servo to rotate the thumb similar to the common movement plane of the CMC joint, as well as a small DC motor to control tendon actuation. We used Myoware 2.0 muscle sensors to detect muscle movement, which would then send a signal to control the motors. This way, you could simply flex your forearm as if you were using your thumb muscles to control actuation of the CMC joint. We soldered a Teensy 4.1 microcontroller to a board as well as a DC motor controller and our Myoware sensor, allowing us to control all of the systems from the microcontroller which stored our code.
For the programming, we designed a code that could interpret the analog output of the Myoware. This analog output was then translated to a voltage and mapped to a serial plotter for the purpose of testing. The analog value was also mapped to a servo value for rotation. As the analog output rose, i.e. the muscle flexed, the servo would rotate accordingly. Once the analog output reached a certain threshold it would activate the DC motor, which would turn the spool and bend the thumb inwards due to its mechanics.
Prototyping - Finger
For the finger design, we first had to understand the 4-bar linkage, which is a system of four bars connected by four joints. We experimented with lengths and anchors to replicate the bending of a finger using linkage simulators and laser-cut linkages. As we progressed, we started combining two 4-bar linkages and translating 2D designs to 3D designs.
ㅤ
ㅤ
ㅤ
ㅤ
ㅤ
Sliding knuckle piece
Our initial prototype tried out a double 4-bar linkage with a sliding mechanism to drive the fingertip movement. However this didn’t prove to be successful and actually ended up over-complicating the mechanism.
New knuckle piece
We ended up with the design below, which instead relies only on rotational movement. The prosthetic finger bends down when the ring is pulled down by the remaining section of the user’s finger, effectively acting as an extension of their finger. This gives the amputee better reach and dexterity than they had with their missing finger(s). This prosthetic finger can also be rotated about the base knuckle and features adjustable locations for the ring.
ㅤ
ㅤ
Challenges
One of our biggest limitations in this project was the Myoware sensor we used to collect data on muscle movement. While the Myoware sensor was the best available option for controlling the servo in the CMC joint, its unreliability and difficulty of use made it a massive limitation for progress in the thumb. After hours and hours of testing, calibrating, and using different electrode placement, we were only able to get it producing usable data two times, which halted our progress in calibrating the movement of the thumb.
Another challenge we faced was with designing the linkage mechanism for the finger. While we did use a linkage simulator for visualization, the simulator couldn’t take into account factors like gravity and the 3D nature of our parts, which means that when we translated our 2D linkages into 3D models, the 3D finger often bent less than we wished or had other issues. This caused confusion as we designed our initial prototypes.
Conclusion
Our team may not have perfected our prosthetic in our 3-week timespan, but we are proud of what we did accomplish:
- Designed a prosthetic finger and thumb that can enable amputees to regain crucial opposable motion in their hand
- Designed a durable prosthetic thumb that can restore thumb functionality on multiple axes once perfected
- Designed and built an adjustable prosthetic finger that can be duplicated and scaled for multiple missing fingers
- Maintained costs significantly lower than market alternatives
- Inexpensive materials: 3D printing filament, nuts and screws, string, $40 Myoware sensor, minimal electronics
- Entire thumb assembly costs around $84 to make, while the finger costs around $2.
Project Video
Moving Forward
- Connect the thumb to a glove with a silicone socket
- Compact electronics and skin-toned components to make the prosthetic more inconspicuous
- Continue to adjust index finger design for grip and strength
- Further testing and assembly of the prosthetic
- More features like touchscreen sensitivity or solar power
Sources
Tee-Melegrito, R. A. (2023, February 17). What is the role of the thumb in hand anatomy?
https://www.medicalnewstoday.com/articles/is-a-thumb-a-finger
Admin. (n.d.). Movements of the thumb.
https://orthopaedicprinciples.com/2017/10/movements-of-the-thumb/
WPI Estimates by Injury. (2020). In Bradford & Barthel. Bradford & Barthel, LLP.
https://bradfordbarthel.com/wp-content/uploads/2021/06/ReserveEstimates-7-21-20.pdf
Enquirer, K. M. (2017, February 17). College students build 3D-printed prosthetic hands for kids. Cincinnati Enquirer.
Global trends of hand and wrist trauma: a systematic analysis of fracture and digit amputation using the Global Burden of Disease 2017 Study. (2020).
Injury Prevention, 26(Suppl 2)
https://doi.org/10.1136/injuryprev-2019-043495
Kawaiah, A., Thakur, M., Garg, S., Kawasmi, S. H., & Hassan, A. (2020). Fingertip Injuries and Amputations: A Review of the literature. Curēus.
https://doi.org/10.7759/cureus.8291
3D-printed custom-designed prostheses for partial hand amputation: mechanical challenges still exist. (2020). In Journal of Hand Therapy.
https://www.jhandtherapy.org/article/S0894-1130(20)30086-7/abstract
Maduri, P., & Akhondi, H. (2023, August 8). Upper limb amputation. StatPearls – NCBI Bookshelf.
https://www.ncbi.nlm.nih.gov/books/NBK540962/
Current rates of prosthetic usage in upper-limb amputees – have innovations had an impact on device acceptance? Disability and Rehabilitation, 44(14),
3708–3713.
https://doi.org/10.1080/09638288.2020.1866684
Zhang, L., Azmat, C. E., & Buckley, C. J. (2023, January 30). Digit amputation. StatPearls – NCBI Bookshelf.
https://www.ncbi.nlm.nih.gov/books/NBK538153/
Click outside images for source
Meet the Team
Anoushka
Gabe
Nolan
Rishi
Sylvia
Finger design, 2D/3D modeling, laser cutting, digital design
Thumb design, 3D modeling, electrical, programming, video editing
Thumb design, electrical, wiring/soldering, portfolio website
Thumb design, 3D modeling, wiring/soldering, manufacturing
Finger design, 2D/3D modeling, laser cutting, portfolio website
Photos
