Telefono: 02.2399.8228
Skype: davide.felice.redaelli
E-mail: davidefelice.redaelli(at)

Methods and Tools for Product Design

Relatore: Prof. Giorgio Colombo

Tutor: Prof. Maurizio Vedani

Università di Provenienza: Politecnico di Milano - Ingegneria Meccanica

Titolo della Tesi: Innovation in the Production Process of Custom-Made Scoliosis Braces with Additive Manufacturing

Innovation in the Production Process of Custom-Made Scoliosis Braces with Additive Manufacturing

The work presents a new manufacturing process to produce patient-specific scoliosis braces by using the 3D printing technology


The scoliosis is the most common spine disorder and the treatment consists on a combination of physical therapy and use of back braces, which are used to support and help realigning the spine [1], [2].
The current manufacturing processes are based on thermoforming a plastic plate, requiring the production of a positive mold to be wrapped, which is then discarded as a waste. Moreover, the production is still very dependent on the technician’s manual operations reducing the repeatability of the process [3].


The main goal was the substitution of the thermoforming of the plastic plate with 3D printing of polymers.
Analyze the accuracy of 3D scanners and select the most appropriate for the application.
Verify the current modeling tools and provide an improvement by using the 3D patient-specific skeleton model.
Have a fast indication of the brace quality with static structural analysis on the computer.
Evaluate available filament materials and select the one that best compares to the thermoformed Polypropylene.
Define the best printing setup and test a fully 3D printed brace for post-processability and for comfort with a patient.


A set of 3D scanners was compared using standard reference objects required by the guidelines4 and mannequin parts, representative of the orthopedic application.
The fixed scanners performed better in terms of reconstruction quality. However, the long acquisition time makes them impractical for the orthopedic. Instead, the handheld scanners can compute a fast reconstruction by moving around the patient in few seconds.
Results showed that Artec Leo and Structure Sensor are the most appropriate for the orthopedic centers. The Artec Leo resulted in higher accuracy (about 0.2 mm), which could be required for very detailed parts, while the Structure Sensor can be stably used for the back- brace design, with an accuracy of about 0.6 mm.
The Structure Sensor was thus selected for the following activities with the real patient.

The most relevant gap of the current CAD tools is the absence of a 3D skeleton model that could be used as a visual reference. The technicians are limited to using bi- planar X-ray images to imagine the interaction with the patient’s body.
For this reason, I developed a pseudo-parametric skeleton model that could be morphed according to these two X-ray projections. This enables obtaining an approximated patient-specific skeleton model, which can be used as a reference when sculpting the brace in its low poly tessellated formulation.

A NURBS-surface formulation of the model, working in the background, was used in the numerical simulation tool to compute the interaction between the brace and the patient’s body.
A set of simplified simulations was performed to verify the usability of the developed model and the possibility to reach a fast result to validate the sculpted brace before 3D printing it.
The simulations converged in about one hour due to the non- linearity of the contact problem. The results show the interaction with the patient’s body and help the decisions about the material distribution in relation to the loading regions.

Regarding the Additive Manufacturing, the FFF technology was selected for the best compromise in terms of costs, speed and printing volume.
Different materials were initially considered, and the best results were obtained with the PETG, both for the mechanical properties and for the inter-layer adhesion, verified with a SEM analysis.
PETG was thus used for 3D printing a full brace that was successfully tested with a volunteering patient. The positive feedback regarded not only the patient but also the orthopedic physician and technicians involved in the experiment.


The main conclusion of the thesis is that the process is currently feasible in all the different steps.
First, the Structure Sensor can be used to acquire a reliable skin model of the patient. Next, the 3D skeleton model is needed as a reference for creating a consistent virtual model of the brace. Then, the FEM analysis helps validating and improving the brace shape, before 3D printing it. Finally, the brace can be produced using PETG and post-processed with common orthopedic tools..


[1] M. Aebi, The adult scoliosis, Eur Spine J 14, 925–948, 2005
[2] S.L. Weinstein, L.A. Dolan, J.G. Wright, M.B. Dobbs, Effects of bracing in adolescents with idiopathic scoliosis, New England Journal of Medicine, 369(16):1512–1521, 2013.
[3] D.F. Redaelli, E. Biffi, G. Colombo, P. Fraschini, G. Reni, Current and future manufacturing of chest orthoses, considering the case of osteogenesis imperfecta, In ASME IDETC-CIE, 2018.
[4] VDI. Optical 3d-measuring systems, Standard, VDI, Dusseldorf, Germany, December 2008.