<- Back to digest

Journal of Pediatric Orthopaedics - 2026-10-01 - Journal Article; Multicenter Study

A Preliminary Report: Screening Idiopathic Scoliosis (IS) With a Smartphone Application.

Martrenchar P, Caille A, Dina A, Odent T

prospective cohortLOE IVn = 236 (154 IS, 82 controls)N/A — cross-sectional diagnostic accuracy study

Topics

pediatricsspine
PMID: 41747211DOI: 10.1097/BPO.0000000000003228View on PubMed ->

Key Takeaway

A smartphone app using AI-analyzed 3D surface topography detected idiopathic scoliosis (Cobb >10°) with 100% sensitivity and 89% specificity across 236 pediatric patients aged 8–18.

Summary Depth

Choose how much analysis to show on this article page.

Summary

This multicenter prospective diagnostic study from France and Canada evaluated a smartphone application using AI-processed 3D surface topography video scans against standard radiographic Cobb angle measurement in pediatric patients aged 8–18. Using a >10° Cobb angle threshold, the app achieved 100% sensitivity and 89% specificity for IS detection. The study establishes preliminary diagnostic accuracy but does not assess reproducibility, progression monitoring, or performance across curve severity subgroups.

Key Limitation

The absence of reproducibility or inter-rater reliability data for the AI scan acquisition process is the most critical gap, as real-world deployment by non-specialist users could substantially degrade the reported sensitivity and specificity.

Original Abstract

INTRODUCTION

Idiopathic scoliosis (IS) is a 3-dimensional spinal deformity that affects ∼3.1% of the population. Current strategies for IS screening and management remain limited resulting in delayed detection and inconsistent follow-up, which can result in missed opportunities for conservative treatment and an increased likelihood of requiring surgical intervention. Early and accessible screening tools may help address these gaps. This study aimed to evaluate the performance of a smartphone-based application that leverages 3D surface topography for IS screening, comparing its accuracy with radiographic assessment in patients with confirmed diagnoses and a cohort of control subjects, which had normal clinical examination excluding spinal pathology.

METHODS

A multicenter prospective diagnostic study was conducted involving pediatric patients from 2 reference hospitals in France and Canada. The participants included both iodiopathic scoliotic patients and healthy controls, all aged 8 to 18 years. The smartphone applications (SA) was used to capture video scans of the patients, which were then analyzed using AI to estimate the major coronal curve, blinded of the radiographic measurement. The results were compared with standard radiographic measurements to validate the accuracy of the SA.

RESULTS

A total of 236 participants were included in the study, comprising 154 (65%) childrens and adolescents diagnosed with idiopathic scoliosis (IS) and 82 (35%) control subjects. When used to screen for scoliosis, defined as a major coronal curve angle >10 degrees, the smartphone application demonstrated a sensitivity of 100% and a specificity of 89%.

DISCUSSION

This preliminary report showed that the SA is a promising tool for IS screening. By integrating this technology, the proportion of conservative treatments could increase, reducing the need for surgical interventions, minimizing X-ray exposure and it would decrease the associated socioeconomic costs. Despite the study's limitations, such as selection bias and the lack of reproducibility assessment, the SA shows significant potential for improving IS screening and management. Future studies will focus on using the SA to monitor disease progression.

LEVEL OF EVIDENCE

Level IV-diagnostic study, prospective, and multicentric.