Fig 1.
The US measurement on a scoliotic subject.
(a) the subject being scanned using SonixTABLET medical ultrasound system, (b) the US coronal image with the measurements of vertebral rotation, (c) the overlaid US coronal image on the previous radiograph, and (d) the US transverse image at apical level (T12).
Table 1.
The curve information and the comparison of measurements between this study and the previous study[14].
Table 2.
The US measurement results in different curve severities, curve locations, subjects’ weight status and US acquisition experiences.
Table 3.
The US measurement results in the different AVR groups.
Fig 2.
The comparison between radiographic and US measurement on Cobb angles.
a) the regression line between radiographic and US proxy Cobb angles, b) Bland-Altman plot between the measurement difference on the US proxy Cobb minus the radiographic Cobb angle versus the average Cobb angles between US and radiographic measurements.
Fig 3.
The US images from a patient with the maximum AVR of 16°.
(a) the coronal image, and (b) the transverse image.
Fig 4.
The schematic of ultrasound beam propagating from the transducer to a tilted vertebra.
Fig 5.
The transverse US images from the same US scan on a 16-year-old girl with BMI 20.5 at different vertebral levels, and the two arrows indicate the distances between skin surface and vertebral lamina, i.e. the muscle thickness covering on the vertebra.
(a) T7, (b) T12 and (c) L3.
Fig 6.
The US coronal images from 4 female subjects scanned by the two operators.
(a) Subject A (age:12.1, BMI:17.8, Maximum Cobb angle:12°, Maximum AVR: 1°) scanned by the experienced operator, (b) Subject B (age:12.0, BMI:17.0, Maximum Cobb angle:12°, Maximum AVR: 2°) scanned by a trainee, (c) Subject C (age:15, BMI:24.6, Maximum Cobb angle:26°, Maximum AVR: 11°) scanned by the experienced operator, (b) Subject D (age:16.6, BMI:23.6, Maximum Cobb angle:30°, Maximum AVR: 10°) scanned by a trainee.