JMIR Publications

ABSTRACT

Background:

Early detection of adolescent idiopathic scoliosis (AIS) is critical for timely intervention and optimal clinical outcomes. Conventional radiography, the current reference standard, is poorly suited for large-scale screening because of cumulative ionizing radiation exposure and concerns related to privacy and patient acceptability.

Objective:

This study aimed to evaluate the diagnostic accuracy of a millimeter-wave imaging system for scoliosis screening, using radiographic Cobb angle measurements as the reference standard.

Methods:

In this prospective diagnostic accuracy study, 132 consecutive pediatric outpatients (aged 6-23 years) with suspected scoliosis underwent a 2-second millimeter-wave scan of the back without removing clothing, followed by standard standing full-spine radiography. Scoliosis was defined as a Cobb angle of ≥10°. Millimeter-wave images were evaluated for established morphological indicators of spinal asymmetry, including shoulder height asymmetry, trunk lateral shift, waistline contour asymmetry, and lower limb height discrepancy. Diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated in accordance with STARD reporting guidelines. Participants older than 18 years were included to reflect real-world outpatient screening practice.

Results:

Radiographic assessment identified scoliosis in 98 of 132 participants (74.2%). Millimeter-wave imaging achieved an overall accuracy of 86.4% (95% CI 76.5-94.7), with a sensitivity of 85.7% (95% CI 75.1-96.5) and a specificity of 88.2% (95% CI 70.7-97.6). All scans were completed within 2 seconds and maintained full patient privacy.

Conclusions:

Millimeter-wave imaging is a feasible, rapid, and nonionizing modality for scoliosis screening. Its high sensitivity supports its use as a first-line screening tool in school and outpatient settings, enabling targeted referral for confirmatory radiography while adhering to the ALARA principle.