Specification of neck muscle dysfunction through digital image analysis using machine learning

Abstract

Everyone has or will have experience some degree of neck pain. Typically, neck pain is associated with the sensation of tense, tight or stiff neck muscles. However, it is unclear whether the neck muscles are objectively stiffer with neck pain. Some investigations documented higher stiffness of the neck muscles with neck pain, while others did not find differences compared to asymptomatic study participants. This is a cross- sectional, observational study that analyses shear wave elastography data obtained from 38 women. In this study, we trained machine learning models that can classify the shear wave elastography images at the level of an expert. Knowledge on a potentially increased objective stiffness of the neck muscles is important when related to diagnosis or therapeutic decisions. Moreover, such an automated approach enables a computed image analysis, which may provide new insights of the physiological properties of the neck muscles in individuals suffering from neck pain.