Main Article Content

Abstract

Artificial disc replacement (ADR) is a surgical procedure during which degenerated intervertebral discs within the human cervical spine are replaced with artificial discs. ADR modifies the stress patterns in the bones, leading to abnormal bone growth called heterotopic ossification (HO). The motional behavior of the spine (i.e. range of motion) has to be investigated after such an ADR procedure. In this study, an anatomically accurate and validated finite element model (FEM) of a human sub-axial spinal column (C3-C7) with a Bryan disc placed at C5-C6 was simulated. Material properties for all spine components were taken from the literature. An algorithm that predicts HO formation from literature was coded using the FORTRAN subroutine and linked with ABAQUS. Python programming was used for the automation of a range of motion (ROM) computations, multiple job file execution, and data extraction. The FEM data such as ROM and load applied were extracted and used in a random forest machine learning regression model which predicted ROM for varying loads and disc positions (i.e. different FSUs). The results predicted by the machine learning model agreed strongly with the simulated results, as indicated by an R2 value of 0.94.

Keywords

cervical spine, heterotopic ossification, finite element, random forest, range of motion, automation.

Article Details

How to Cite
Shruthi, R., Parvathi, R., Kumar, S. D., Srinivasan, S., Babu, L., Jebaseelan, D., & Yoganandan, N. (2021). Automation and Machine Learning Framework using Finite Element Solution for Cervical Spine Applications. Trends in Biomaterials & Artificial Organs, 35(2), 173-176. Retrieved from https://biomaterials.org.in/tibao/index.php/tibao/article/view/440