Funding outcome

Adolescent idiopathic scoliosis (AIS) is a condition where the spine curves to the side, resulting in a C or S-shaped curve. AIS can produce consequences such as back discomfort, breathing difficulties, and poor self-image. Despite the fact that both genders are susceptible, females are more likely to develop scoliosis. Additionally, females with AIS have a higher risk of worsening of the spinal curve during their growth period. Currently, the standard method for diagnosing and monitoring AIS is the Cobb angle, which is measured from two-dimensional radiographs. This method has limitations, including exposure to ionizing radiation, and does not reflect the curvature of the spine in a three-dimensional space. Moreover, the bending test, the most common tool to detect scoliosis in children early relies on the accuracy of the examiner and has a high referral rate, exposing children to unnecessary radiation. An alternative tool for screening and monitoring scoliosis without harmful side effects is the 3D markerless surface topography (ST). The ST technique method identifies areas of asymmetry of the torso and the intensity through a deviation colormap (DCM) image. However, the ST method has focused only on broad classification of curve severity, and it cannot predict the minimum threshold of scoliosis. Our study aims at integrating artificial intelligence (AI) in the ST analysis technique to detect the condition and improve the prediction of curve severity. The ST approach can be an impactful tool for improving the health outcomes of woman and children by reducing the reliance on radiographs and developing an effective tool for assessing and monitoring AIS.