AUTHOR=Boché André , Tabard-Fougère Anne , Bazin Ludmilla , Severyns Mathieu , Dayer Romain , Vendeuvre Tanguy 

TITLE=Time-of-flight camera achieves high diagnostic accuracy in adolescent idiopathic scoliosis: a promising radiation-free alternative to radiography

JOURNAL=Frontiers in Bioengineering and Biotechnology

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2025.1629872

DOI=10.3389/fbioe.2025.1629872

ISSN=2296-4185

ABSTRACT=BackgroundThe close monitoring of the adolescent idiopathic scoliosis (AIS) population during the growing years is necessary and requires repetitive X-rays. This study aimed to evaluate the validity and test characteristics of the time-of-flight (TOF) camera, a novel radiation-free tool, for assessing trunk asymmetry in patients with AIS.MethodsIn this prospective diagnostic accuracy study, 94 AIS patients (10–18 years) underwent standardized TOF camera imaging (seated position). Among them, 81 also received an angle of trunk rotation (ATR) assessment using a scoliometer (forward-bending position). The average major Cobb angle (CA) in the cohort was 21.4°. The global trunk asymmetry (GTA) provided by the TOF camera and the scoliometer-based ATR were compared with the gold-standard major Cobb angle from 2D radiography (standing position) using Pearson correlation. Receiver operating characteristic (ROC) analysis evaluated the diagnostic accuracy of GTA for AIS diagnosis (CA > 10°) and brace indication (CA > 20°). Sensitivity (Se), specificity (Sp), and area under the curve (AUC) values were reported and compared for GTA and ATR.ResultsGTA was significantly correlated with the radiographic CA (p < 0.001). For scoliosis diagnosis (CA ≥ 10°), the AUC was 0.87 (0.79–0.96) for GTA (threshold = 7°, Se = 80%, and Sp = 80%) and 0.95 (0.9–1.0) for ATR. The AUC values of GTA and ATR were not significantly different (p = 0.129), but sensitivity was significantly lower for GTA (76%) than for ATR (91%). For brace indication (CA ≥ 20°), the AUC was 0.92 (0.86–0.97) for GTA (threshold = 7.87, Se = 93%, and Sp = 76%).ConclusionThe TOF camera appears to offer promising test characteristics for AIS diagnosis, with a high correlation to radiographic CA and competitive diagnostic accuracy using a scoliometer. Although the seated positioning differs from standard radiographs, this approach enhances reproducibility and patient compliance. The high sensitivity and specificity of the TOF camera for scoliosis diagnosis highlight its potential as a safe, fast, and reliable alternative to X-ray imaging in routine clinical settings. Further investigations (assessing solid screening characteristics and inter- and intra-individual repeatability and validity) are needed before it can replace repetitive radiographs for monitoring AIS progression in growing patients.