A Review of Recent Advances in Prosthetic Socket Fabrication: Integration of 3D Printing and Machine Learning Optimization

This review provides a comprehensive analysis of recent advancements in 3D-printed prosthetic socket materials, 3D printing techniques and data-driven machine learning (ML) approaches for prosthetic design and alignment. Furthermore, it outlines key technical limitations, material trade-offs, and future research directions necessary to advance socket development. The review was conducted through a systematic and critical analysis of a large body of literature obtained from widely recognized scientific databases. The findings indicate that the integration of 3D printing technology with ML enhances the optimization of fabrication parameters, improves alignment detection, and enables advanced movement control. Reported findings suggest that carbon fiber (CF)-reinforced composites offer favorable mechanical strength and rigidity for socket design, while nanoparticle-reinforced matrices like PLA- multi-walled carbon nanotubes (MWCNTs) represent promising alternatives under controlled test parameters. Regarding data-driven methods, ML models mainly process sensor data for alignment and motion tasks. Among these, Support Vector Machine (SVM) and Levenberg–Marquardt backpropagation architectures are widely investigated and show encouraging accuracy for specific classification and prediction tasks, though broader generalization requires further standardization across testing conditions. Despite these advancements, significant research gaps remain. Future studies should focus on the development and evaluation of novel materials to further improve socket performance and user comfort. Additionally, a broader range of ML algorithms should be explored for gait and alignment prediction, as well as for the optimization of the 3D-printing process itself.