Context-aware sensor solution for remote monitoring of adolescent idiopathic scoliosis brace treatment

Abstract

The medical condition of Scoliosis occurs when an individual’s spine develops curvature in adolescents. A Brace treatment is used to control the lateral curvature of the spine in scoliosis. However, brace treatment is a long and inconvenient process that demands strict compliance by the patients. In this work, we designed a wearable sensor solution to monitor the brace treatment compliance. The hardware is embedded into the patient’s brace. The custom designed hardware consists of a sensor board, multiple sensors. The force sensor collects the force being exerted on the patient’s back, while the motion sensor generates cues to determine the patient’s activities and context. We aim to evaluate monitoring of the effectiveness of the brace treatment pervasively based on fusion of continuous force and motion recordings. The proposed method evaluates the duration of brace wear through the process of segmentation and calculates the level of tightness of brace by estimating the baseline force per segment in the presence of different activities including sitting, standing, climbing, walking, running and lying. We investigated an experimental scenario in which, the patient performs a series of pre-defined activities at home during day long segments of brace wear, during pervasive sensor data recordings. The experimental results demonstrated that we achieved an overall accuracy of a 96.1% for unsupervised activity detection. Our trained model estimated a reduction in the level of tightness of brace by 30% during a period of 2 weeks while the compliance of brace treatment gradually increased.