There are no barriers when it comes to the applications of Artificial Intelligence (AI) in the modern world. With the introduction of AI as a first-hand utility tool such as; ChatGPT, the world has entered into a new era. Now people are convinced that enhancing efficiency is the prime objective.
Indeed, we are witnessing the remarkable impact of AI in various fields, from automated text solutions to cybersecurity, transportation, manufacturing, retail, finance, education, and particularly the field of healthcare.
Healthcare and AI
Healthcare has been extremely important forever, directly linked to human quality of life and welfare. AI has transformed modern healthcare diagnosis, treatments and patient care with its extensive applications in medical tests and treatments like; MRI, CT scans, X-rays, diagnostic algorithms, drug delivery, personalized medicine, robotics-assisted surgeries and so on.
One of the most promising and interesting applications of AI is in the field of prosthetics, providing smart and accessible solutions. Gone are the days when artificial limbs had to be static, uncomfortable and detached from the body’s sensational feedback.
What is Prosthetics?
Prosthetics are artificial devices designed to augment a missing or damaged body part such as hands, feet, limbs or a facial feature. According to World Health Organization (WHO), prostheses are artificial devices to replace missing body parts, while orthoses are supportive braces and splints to help damaged parts.
A good prosthetic should deliver both function and aesthetic pleasance, making the amputee feel independent, emotionally comfortable and complete.
Traditional Prosthetics and Challenges
The first prosthetic was a ‘toe’ used by an Egyptian woman around 3000 years ago as it was necessary to wear an Egyptian sandal. Later, during the Dark Ages, prosthetic limbs came into existence, which were mere rigid components with no functional value. Examples of such prosthetics are wooden or metal hands and pegleg, used by sea pirates.
Traditional prosthetics were neither aesthetically pleasant nor exhibited mobility or functional value. These early devices were typically constructed from heavy materials, resulting in bulky and uncomfortable designs. This not only impacted the physical comfort of the wearer but also took a toll on their emotional well-being. Individuals relying on traditional prosthetics often face challenges in carrying out daily tasks independently, leading them to seek assistance despite the use of artificial devices. These challenges prompted the development of modern-day prosthetics.
Modern Prosthetics and the Role of AI
The evolution of prosthetics has been nothing short of extraordinary, traversing centuries of innovation and advancements. From rudimentary wooden peg legs to intricately designed robotic limbs, the field of prosthetics has undergone a remarkable transformation.
A French war surgeon ‘Ambroise Pare’ developed the first functional prosthetics and switched to lighter materials. He used the principles of human physiology to mimic body movements like natural. These prosthetic designs are still used, with added improvements.
The year 1993 marks the introduction of intelligent prosthetics, opening a new direction for smart and sense-controlled prosthetics. Adaptive prosthetics emerged in 1998, which worked on microprocessor, pneumatic and hydraulic-based mechanisms. In 2006, OSSUR (Iceland-based company) developed the fully controlled AI-based power knee. Later, the same company developed the first bionic leg, which connects the mind and machine together.
The first fully integrated artificial limb was developed in 2015 by Blatchford. This used a total of 7 sensors and 4 CPUs to connect it with the body sensations and control. This AI-based system gives a more natural outcome when it comes to routine tasks like sitting, walking, and standing and gives independence to the user.
Today, AI is a necessary feature of modern prosthetics.
By harnessing the power of AI, we can create prosthetics that are more functional, intuitive, and personalized than ever before.
How AI Makes Prosthetics Smart
Artificial Intelligence (AI) is used in modern prosthetics on a similar human natural coordination system principle.
Just like the human sensory organs (eyes, nose etc.) coordinate with effectors (hands, limbs etc.) through the brain, the robotic prosthetics work similarly. These devices use cameras or radiation as sensors and motor devices, connected electrically as effectors. Using complex algorithms and formulas, a central unit acts as the brain.
This central unit (like the brain) is equipped with receiving and interpreting body sensations. It is the most important component, where AI is applied through two basic mechanisms:
SL (Symbolic Learning):
It helps to process images, symbols and the environment through a camera lens (computer vision).
ML (Machine Learning):
It helps to process the data input through sensors, storing some of it as memory to adapt and adjust to the user’s needs over time. It uses classifier and prediction algorithms to recognize speech and language, known as a ‘statistical learning mechanism’.
In addition, ML achieves sensory connection to the amputee’s body through a ‘deep learning mechanism’, depending upon CNN (Convolution Neural Network) and RNN (Recurrent Neural Network).
| Humans | AI Prosthetics |
| Vision, Speaking and Listening | Symbolic vision and statistical learning |
| Learning, recognition and memory | Artificial neural network and processor |
| Object and environment recognition | Machine learning through CNN |
Exciting Breakthroughs and Success Stories of AI-Powered Prosthetics
Artificial Intelligence (AI) is used in modern myoelectric prosthetics, where electrodes use muscle impulses to generate and amplify signals that translate into controlled movements. In addition, ‘Peripheral Nerve Interface’ (PNI) and ‘Brain Machine Interface’ (BMI) are employed to understand brain signals and connect smart prosthetics to human voluntary control. Some major breakthroughs in AI-driven prosthetics are:
AI-Based Myoelectric Hands and Bionic Limbs
Balanced voluntary control, jumping obstacles and climbing stairs are some of the most challenging activities for amputees who depend on prosthetics.
Luckily, myoelectric and bionic prosthetics, advanced interventions in the area of prosthetics, have solved this problem by enabling the user to conduct comfortable voluntary actions. This makes use of ‘Electroencephalography’ (EEG) and ‘Electromyography’ (EMG) signals generated through implanted electrodes to take direct nervous messages from the user. Machine learning and memory help make repeated movements smoother and seamless.
Johnny Matheny’s Success with Myoelectric Prosthetic Limb
Johnny Matheny had his arm amputated due to cancer in 2008. He did not give up but collaborated with the APL research team to have the best prosthetic designed. This journey led to the development of advanced MPL (Modular Prosthetic Limb)
This fully self-controlled limb takes signals from the brain through neural networks and device electrodes. Interestingly, this advanced AI-based prosthetic enabled Johnny to play ‘Amazing Grace’ on a piano.
AI and 3D Printing
The amazing technology of 3D printing combined with AI holds great promise towards the user-centred manufacturing of prosthetics. These wearable parts can be made with greater precision, less time and less budget, while AI helps to meet the individual needs of the wearer’s body shape, weight and perfect sizing.
In addition to enhancing user comfort, 3D printing promises to improve the quality of life worldwide, especially in the poorest areas where affordability is one big challenge.
3D Printed Prosthetic Paw and the Success Story of Millie, a Greyhound
Millie, a greyhound puppy, was adopted by an Australian couple. Sadly, it had a missing front paw, greatly impacting its daily life and emotional well-being. However, the couple’s determination to provide Millie with a comfortable and fulfilling life made them choose an AI-based 3D-printed paw as the last yet best resort.
The Inspiring Story of Adrianna Haslet, a professional dancer
A ballroom dancer from Boston, Adrianna suffered a Marathon bombing and later a car accident which made her lose her lower limb and injure her arm badly. Despite the emotional baggage the incident brought, Adrianna stayed put and hopeful for the best. Thanks to AI, Haslet is able to resume her physical fitness and running after years through a prosthetic leg which adapts to her body needs and movements.
Artificial Skin, the E-dermis
Another intelligent feature added to modern prosthetics is the outer skin-like layer to add the real touch further. In addition to voluntary control, modern AI prosthetics provide the sensation of touch, pressure and pain just like natural skin receptors do. This artificial skin is called an ‘e-dermis’, made of rubber and fabric material, connected through a Peripheral Nerve Interface (PNI) to generate sensations of touch, pain and temperature.
AI-Based Exoskeletons
Exoskeletons are outerwear prosthetics like external coverings or suits. When equipped with AI modulators and networks, these work like a charm. Bionic limbs also use a similar approach where AI makes it possible to keep user intent and control integrated through robotic processors.
Angel Giuffria and her Bionic Limb
An actress and model, Angel Giuffria, was born without a left hand. Trying outdated minimal functional, bulky prosthetics in her childhood made her look for better options. Today, she uses an advanced Bionic left limb with a myoelectric hand, enabling her to perform her favourite activities of biking, archery, yoga and workouts.
Current Challenges and The Future of AI-Driven Prosthetics
Pioneering the future of prosthetics and healthcare, there is no doubt, AI is making the quality of life better for everyone. However, mishandling of anything may cause trouble; such is the case with AI. The most common challenges with this technology are:
- Keeping a delicate balance between fully automated AI systems and the human touch will help keep room for correcting malfunctioning. The lack of human touch is one sensitive problem of depending upon computerized and digitalized gadgets or AI. This can put a person at risk in case the automated system malfunctions and there is no option for humans to intervene to correct the error in time.
- Information theft or biohacking and ethical concerns of data sharing and data leakage can make the devices malfunction or freeze and invade users’ privacy. This may be done intentionally by competitor companies or unintentionally by system errors or weak security locks.
- Accessibility of technology The 3D printing of customizable prosthetics seems to be a great opportunity for the world, especially in poverty-stricken areas, but this technology is not yet freely accessible.
- Having sufficiently trained personnel in this area is another challenge. Dealing with the latest technology of smart prosthetics is not everyone’s piece of cake. There is a need for sufficiently trained staff who are aware of technical aspects, troubleshooting and other sensitive areas of the smart prosthetics they are to deal with.
- Overpricing and availability of AI prosthetic components, especially the chips that work as a main processing unit.
- Aesthetically Pleasant Designing to ensure maintaining user’s self-confidence, emotional well-being and independence in society. This can be done by making these devices look more natural, adding similar texture and colour as the body part it replaces.
- Lack of sufficient coordination among engineers, technicians and researchers results in prosthetics lacking in one or more areas.
- Improving AI prosthetics efficiency greatly depends on collecting data from different populations which are able to test the products before giving feedback. This can help with individual variability and better adaptation to improve upcoming AI prosthetics. However, there is still a gap in collecting sufficient high-quality and diverse data.
Resolving these challenges may be our primary target, stepping ahead into a brighter and better future for AI prosthetics.
AI Prosthetics – The Journey Ahead
The past decade has marked a remarkable journey in the field of AI-based healthcare products, especially the innovation of smart prosthetics. These products have made considerable improvements in human quality of life, both physically and emotionally. Combining the principles of ML, SL, EEG and 3D printing, we have achieved AI-integrated prosthetics which deliver more comfort, sensory value and enhanced functionality.
Current research in the area of smart prosthetics focuses on integrating AR (augmented reality) and improved neural coordination networks, making prosthetics work and look just as natural as the actual body part. In addition, advancements in 3D printing technology are being made to make it the ultimate cost-effective and user-specific solution for the world. This can solve the problem of overpricing and accessibility.
Another area to work on is distant/remote monitoring and maintenance of AI prosthetics. This may help detect any error or abnormal behaviour in time through integrated sensors, which can quickly channel this information to the right department. This may also help avoid any risk of biohacking.
The invention of electronic skin ‘e-dermis’ has helped in adding a natural touch to artificial prosthetics. However, the present prototype, made of rubber sheets and electrodes, looks different from the natural skin appearance. Better collaboration among biologists, physiologists, engineers and concerned departments may help design optimized solutions which are also aesthetically pleasing and look more natural. The selection of the right quality materials for prosthetic manufacturing plays a vital role in this matter.
Better collaboration of information is being made possible through an integration of ‘IoT’ (Internet of Things). This helps make data exchange and remote control more coordinated and seamless.
While the demand for advanced prosthetic solutions continues to grow, entrepreneurs and innovators have a unique opportunity to capitalize on the potential of AI in this field.
A Concluding Note
As we journey forward, let’s stay dedicated to advancing prosthetics through continuous development, innovation, and research. Together, we can empower individuals with limb loss, helping them regain independence and improve their overall well-being.
In the future, AI-powered prosthetics will empower individuals with limb loss to redefine possibilities and embrace a life of newfound independence.
References:
- Hassan, M. (2023, January 23). How AI is helping power next-generation prosthetic limbs. Wevolver. https://www.wevolver.com/article/how-ai-is-helping-power-next-generation-prosthetic-limbs
- World Health Organization. (2017). WHO standards for prosthetics and orthotics.
- Nayak, S., & Das, R. K. (2020). Application of artificial intelligence (AI) in prosthetic and orthotic rehabilitation. In Service Robotics. IntechOpen.
- Safonova, O. (2020, December 15). Bioprosthetics: Can They Really Be Controlled With Our Minds? Wevolver. https://www.wevolver.com/article/bioprosthetics-can-they-really-be-controlled-with-our-minds-
- Ghazaei, G., Alameer, A., Degenaar, P., Morgan, G., & Nazarpour, K. (2017). Deep learning-based artificial vision for grasp classification in myoelectric hands. Journal of neural engineering, 14(3), 036025.
- [John and Marcia Price College of Engineering]. (2019, October 29). Utah Bionic Leg [Video]. YouTube. https://www.youtube.com/watch?v=GHTbK3zJ6OY&ab_channel=UtahCOE
- Smith, M. (2021, October 6). Breakthroughs in Prosthetic Technology Promise Better Living Through Design. Redshift. https://redshift.autodesk.com/articles/prosthetic-technology
- Kim, M. S., Kim, J. J., Kang, K. H., Lee, J. H., & In, Y. (2023). Detection of Prosthetic Loosening in Hip and Knee Arthroplasty Using Machine Learning: A Systematic Review and Meta-Analysis. Medicina, 59(4), 782.
- Kulkarni, P. G., Paudel, N., Magar, S., Santilli, M. F., Kashyap, S., Baranwal, A. K., … & Singh, A. V. (2023). Overcoming Challenges and Innovations in Orthopedic Prosthesis Design: An Interdisciplinary Perspective. Biomedical Materials & Devices, 1-12.
- Patel, R. (2021, February 10). A Glimpse Into the Future of Prosthetics: Advanced Sensors, E-Skin, and AI. ALL ABOUT CIRCUITS. https://www.allaboutcircuits.com/news/glimpse-future-prosthetics-advanced-sensors-e-skin-ai/
Note: This article is written with the assistance of Dr Muhammad Mustafa, who is an Assistant Professor at Forman Christian College University (FCCU), Lahore. His main interest in research is Cancer metastasis and the impact of psychological factors on cancer progression. He is known for his work as a faculty trainer and science communicator.
Also Read: GENETICS IN THE SPOTLIGHT: PERSPECTIVES FROM A SCIENTIST AND COMMUNICATOR DR. ALEX DAINIS
Gull-e-lalah is a former intern at Scientia’s Science Writing Internship (cohort 1) and later, working as a team member. She is a scientist, working in the field of Biotechnology and Molecular Pathology, and has a strong passion for science writing, demonstration and teaching.
