Statistical prediction of bone microstructure degradation to study patient dependency in osteoporosis. (October 2022)

• Record Type: Journal Article
• Authors: Famouri, Seyedfarzad; Baghani, Mostafa; Sheidaei, Azadeh; George, Daniel; Farahani, Maryam Mazraehei; Panahi, Masoud Shariat; Baniassadi, Majid
• Other Names: Bersani Alberto Maria guest-editor.; Cazzani Antonio M guest-editor.; Eremeyev Victor guest-editor.; Giorgio Ivan guest-editor.; Spagnuolo Mario guest-editor.
• Is Part Of: Mathematics and mechanics of solids. Volume 27:Number 10(2022)
• Page Start: 1987
• Page End: 2001
• Publication Date: 2022-10
• Subjects: Osteoporosis -- two-phase recovery -- optimization -- two-point correlation function -- bone density prediction; Materials -- Mechanical properties -- Periodicals; Solids -- Periodicals; Materials science -- Mathematics -- Periodicals; 620.11205
• DOI: 10.1177/10812865221098777
• Languages: English
• ISSNs: 1081-2865

Abstract

Numerical prediction of osteoporosis evolution is a challenging objective in medicine, particularly when one desires to account for patient dependency. The use of statistical methods to reconstruct bone microstructure distribution could be a helpful tool for this prediction, as they are able to provide different types of microstructures that can be optimized to fit with each patient. An initial bone sample was obtained from high-resolution X-ray computed tomography (HRμCT). Its microstructure evolution in time using a previously developed degradation model was used as the ground truth. Statistical bone microstructures were reconstructed at different stages of this evolution using two-point correlation functions (TPCFs). A blind search approach is used to find the optimized statistical microstructures, and the optimized coefficient showed less than 2% TPCF error between the statistical reconstruction and the degraded model. The statistical models also showed less than 13% error in the corresponding mechanical properties. The results showed a good correlation between the developed approach and the ground truth. The method could be extrapolated to account for the physical characterization of patient dependency to predict bone density loss over time.