BME MS Defense: Chien-Chun Yang

Predicting Biomechanical Strength of Proximal Femur Specimens through a Combination of Texture Feature Analysis and Machine Learning Techniques

Supervised by Dr. Axel Wismueller

Abstract:

Bone fragility and fracture caused by osteoporosis or injury are prevalent in adults over the age of 50 and can reduce their quality of life. Hence, predicting the biomechanical bone strength, specifically of the proximal femur, through non-invasive imaging-based methods is an important goal for the diagnosis of osteoporosis as well as estimating fracture risk. Dual X-ray absorptiometry (DXA) has been used as a standard clinical procedure for assessment and diagnosis of bone strength and osteoporosis through bone mineral density (BMD) measurements. However, previous studies have shown that quantitative computer tomography (QCT) can be more sensitive and specific to trabecular bone characterization because it reduces the overlap effects and interferences from the surrounding soft tissue and cortical shell.

This study proposes a new method to predict the bone strength of proximal femur specimens from quantitative multi-detector computer tomography (MDCT) images. Texture analysis methods such as conventional statistical moments (mean, standard deviation, etc), gray-level co-occurrence matrix (GLCM) and scaling index method (SIM) are used to quantify BMD properties of the trabecular bone micro-architecture. Combinations of these extracted features are then used to predict the femur specimens' strength through machine learning techniques such as multi-regression (MultiReg) and support vector regression with linear kernel (SVRlin). The prediction performance achieved with these feature sets is compared to the standard approach that uses the mean BMD of the specimens and multi-regression models using root-mean-square error (RMSE) and the coefficient of determination R2.

The highest prediction performance for GLCM features alone was obtained with a GLCM feature set and SVRlin (RMSE = 1.069 ± 0.145, R2= 0.533), which was significantly lower than the standard approach of using mean BMD and MultiReg (RMSE = 1.108 ± 0.141, R2 = 0.493, p