A Review on Recent Advancements in Scoliosis Detection System

Abstract

Scoliosis, particularly Adolescent Idiopathic Scoliosis (AIS), represents a prevalent orthopedic condition characterized by a lateral curvature and axial rotation of the spine. Early detection and intervention are critical to prevent curve progression and associated complications. This review critically examines ten significant research studies spanning clinical diagnostics [1][2], genetic predisposition analysis [5][7], artificial intelligence (AI)-based detection methods [3][4][8][9], and large-scale epidemiological investigations [10]. The integration of deep learning algorithms into scoliosis screening has demonstrated substantial potential in enhancing diagnostic accuracy while minimizing reliance on radiographic imaging [3][4]. Genetic studies have elucidated loci such as LBX1 [5], suggesting hereditary susceptibility, although clinical translation remains limited [7]. Epidemiological data underline the need for structured screening initiatives [10], particularly during adolescence. Despite these advancements, challenges persist, including the external validation of AI models, standardization of clinical protocols, and integration of multimodal data for personalized care. This review synthesizes current knowledge, identifies key research gaps, and highlights future directions toward developing comprehensive, non-invasive, and predictive scoliosis management frameworks.