NL
DE
“With the software we developed, the design process for prosthetics becomes much simpler. This allows local staff at Masanga Hospital in Sierra Leone to independently produce prosthetics. A prosthesis truly makes a difference for someone who has undergone an amputation. Moreover, they can be made more quickly, enabling rehabilitation to start sooner,” says Technical Physician Merel van der Stelt from Radboud university medical center (Radboudumc). On Wednesday, October 22, she will defend her PhD thesis on this topic.
Globally, there is a significant unmet need for prosthetics. The production method developed by Radboudumc for Masanga Hospital is now also being used to support people with amputations in Ukraine. In her dissertation, Van der Stelt demonstrates that integrated AI algorithms can be used to train local staff in Sierra Leone to independently produce prosthetics. This enables people with amputations to rejoin society and regain independence.
Only one in ten people with amputations in low- and middle-income countries has access to a prosthesis
A well-fitting prosthesis is essential for people with amputations. Worldwide, an estimated 35 to 40 million people need a prosthesis, and this number is expected to rise due to aging populations, increasing rates of diabetes, and more trauma from accidents and armed conflict. According to the World Health Organization, only one in ten people with amputations in low- and middle-income countries currently has access to a prosthesis. This is largely due to a shortage of trained prosthetists and the high cost of production.
3D-printed prosthetics
Initial studies by Van der Stelt showed that people appreciated 3D-printed arm prosthetics because of their lifelike appearance, which boosted confidence and social participation. She later focused on leg prosthetics, which were in high demand in Sierra Leone. Together with local colleagues, she evaluated materials suitable for long-term use and assessed the quality of locally produced prosthetics.
AI predicts prosthetic socket fit
One major challenge remained: designing the prosthetic socket—the part that connects directly to the residual limb—was too complex for local staff to perform independently. The solution came through artificial intelligence. Van der Stelt and her team developed user-friendly software with integrated AI algorithms trained on data from experienced prosthetists. These algorithms can predict the ideal socket shape, allowing local staff to design them independently.
After intensive training, the local team produced custom prosthetics for 34 people with lower-leg amputations within a year, with minimal supervision. Follow-up showed that the prosthetics were worn for an average of nine hours per day and that most users achieved their rehabilitation goals. To date, over 80 prosthetics have been produced entirely independently.
About the PhD defense
Merel van der Stelt conducted her research at the 3D Lab of Radboudumc under the supervision of Prof. Thomas Maal, the first professor of 3D technology in healthcare, Dr. Ruud Leijendekkers (Rehabilitation Medicine), trauma surgeon Dr. Lars Brouwers, and Prof. Martin Grobusch (Amsterdam UMC). She collaborated with colleagues from Radboudumc’s 3D Lab and the Research Unit at Masanga Hospital in Sierra Leone.
Prior to her PhD at Radboudumc, Merel van der Stelt studied Technical Medicine at the University of Twente. During her graduation internship—also at the 3D Lab of Masanga Hospital in Sierra Leone—she laid the foundation for her doctoral research. For her internship research, she received the Vooijs Award for the most clinically relevant study. Merel’s work highlights the growing importance of the close collaboration between Radboudumc and the University of Twente within the HealthTech Nexus.
Her PhD defense will take place on October 22, 2025, at 14:30. Title of the dissertation: 3D Technology in Prosthetic Care: Enhancing Accessibility in Sierra Leone. The defense can be followed via livestream.
-
Want to know more about these subjects? Click on the buttons below for more news.
