Paper Info
Title
The effect of augmentation and transfer learning on the modelling of lower-limb sockets using 3D adversarial autoencoders
Abstract
Lower limb amputation is a condition affecting millions of people worldwide. Patients are often prescribed with lower limb prostheses to aid their mobility, but these prostheses require frequent adjustments through an iterative and manual process, which heavily depends on patient feedback and on the prosthetist’s experience. New computer-aided design and manufacturing technologies have been emerging as ways to improve the fitting process by creating virtual models of the prosthesis’ interface component with the limb, the socket. Using Adversarial Autoencoders, a generative model describing both transtibial and transfemoral sockets was created. Two strategies were tested to counteract the small size of the dataset: transfer learning using the ModelNet dataset and data augmentation through a previously validated socket statistical shape model. The minimum reconstruction error was 0.00124 mm and was obtained for the model which combined the two approaches. A single-blind assessment conducted with prosthetists showed that, while generated and real shapes are distinguishable, most generated ones assume plausible shapes. Our results show that the use of transfer learning allowed for a correct training and regularization of the latent space, inducing in the model generative abilities with potential clinical applications.
Year
DOI
Venue
2022
10.1016/j.displa.2022.102190
Displays
Keywords
DocType
Volume
3D generative modelling,Adversarial Autoencoder,Lower limb sockets,3D scanning,Transfer learning,Data augmentation
Journal
74
ISSN
Citations 
PageRank 
0141-9382
0
0.34
References 
Authors
0
5
Name
Order
Citations
PageRank
Ana Costa100.34
Daniel Rodrigues200.34
Marina Castro300.34
Sofia Assis400.34
Hélder P. Oliveira56313.99