ScolioClass: Data-driven Development of a New Classification Tool to Evaluate Adolescent Idiopathic Scoliosis Optically Diagnosed

Saša Ćuković^1*Mihai Neghina²Radu Petruse²Vanja Luković³Dijana Stojić³Yiying JoJo Zou⁴

• ¹ETH Zürich, Zurich, Switzerland
• ²Universitatea Lucian Blaga din Sibiu, Sibiu, Romania
• ³Univerzitet u Kragujevcu, Kragujevac, Serbia
• ⁴The Hong Kong Polytechnic University China Research and Development Network, Hong Kong, Hong Kong, SAR China

Adolescent idiopathic scoliosis (AIS) is traditionally assessed and classified using ionizing radiographic methods that rely on Cobb angle measurements and qualitative curve modifiers, exposing patients to repeated X-ray radiation and offering limited sensitivity to subtle three-dimensional deformities. We developed ScolioClass, a non-invasive, data-driven classification tool that harnesses 3D optical scanning and continuous indices, capturing curvature severity, directionality, and sagittal balance, to evaluate spinal deformities in 94 AIS patients. By comparing ScolioClass descriptions against the established Lenke classification, we observed a statistically significant association (χ²≈29.0, df=6, p<0.001) with 72.3% overall agreement, and between sagittal modifiers and ScolioClass kyphosis–lordosis categories (χ²≈48.4, df=3, p<0.0001) with 68.1% agreement. Notably, ScolioClass detected mild curves and lordotic patterns that were often overlooked by Lenke criteria. These findings demonstrate that ScolioClass provides a quantitative, three-dimensional assessment of AIS without radiation exposure, offering finer granularity for clinical decision-making and potential for automated analysis. With further validation and integration into clinical workflows, ScolioClass may enhance personalized treatment planning and reduce radiation risks in long-term scoliosis management.