Prediction Method of Elastic Modulus of Trabecular Bone Based on SE-DenseVoxNet.
10.3969/j.issn.1671-7104.2021.01.002
- Author:
Yiwei CAO
1
;
Siyuan HE
1
;
Ping ZHOU
1
;
Lan LI
2
;
Nirong BAO
3
Author Information
1. School of Biological Science & Medical Engineering, Southeast University, Nanjing, 210096.
2. Nanjing Drum Tower Hospital, Sports Medicine and Adult Reconstructive Surgery, State Key Laboratory of Medical Biotechnology, Nanjing, 210008.
3. Jinling Hospital, Nanjing, 210002.
- Publication Type:Journal Article
- Keywords:
DenseVoxNet;
SENet;
elastic modulus;
trabeculae
- MeSH:
Biomechanical Phenomena;
Bone Density;
Cancellous Bone/diagnostic imaging*;
Elastic Modulus;
Humans;
Osteoporosis/diagnostic imaging*
- From:
Chinese Journal of Medical Instrumentation
2021;45(1):6-10
- CountryChina
- Language:Chinese
-
Abstract:
Osteoporosis is one of the common metabolic diseases, which can easily lead to osteoporotic fractures. Accurate prediction of bone biomechanical properties is of great significance for the early prevention and diagnosis of osteoporosis. Bone mineral density measurement is currently used clinically as the gold standard for assessing bone strength and diagnosing osteoporosis, but studies have shown that bone mineral density can only explain 60% to 70% of bone strength changes, and trabecular bone microstructure is an important factor affecting bone strength. In order to establish the connection between trabecular bone microstructure and bone strength, this paper proposes a prediction method of trabecular bone modulus based on SE-DenseVoxNet. This method takes three-dimensional binary images of trabecular bone as input and predicts its elastic modulus in the z-axis direction. Experiments show that the error and bias between the predicted value of the method and the true value of the sample are small and have good consistency.
