Merging computational intelligence and wearable technologies for adolescent idiopathic scoliosis: a quest for multiscale modelling, long-term monitoring and personalized treatment

Adolescent idiopathic scoliosis (AIS) is a dynamic progression during growth, which requires long-term collaborations and efforts from clinicians, patients and their families. It would be beneficial to have a precise intervention based on cross-scale understandings of the etiology, real-time sensing and actuating to enable early detection, screening and personalized treatment. We argue that merging computational intelligence and wearable technologies can bridge the gap between the current trajectory of the techniques applied to AIS and this vision. Wearable technologies such as inertial measurement units (IMUs) and surface electromyography (sEMG) have shown great potential in monitoring spinal curvature and muscle activity in real-time. For instance, IMUs can track the kinematics of the spine during daily activities, while sEMG can detect asymmetric muscle activation patterns that may contribute to scoliosis progression. Computational intelligence, particularly deep learning algorithms, can process these multi-modal data streams to identify early signs of scoliosis and adapt treatment strategies dynamically. By using their combination, we can find potential solutions for a better understanding of the disease, a more effective and intelligent way for treatment and rehabilitation.