Validation reliability and responsiveness of a new surface topography instrument

Aims of study:

Idiopathic scoliosis is an abnormal lateral curvature of the spine, generally first noticed as a change in back shape, notably an asymmetric rib hump on the back. The severity of the condition has traditionally been measured by radiography (via Cobb angle measurement) but this has two disadvantages. First, over a period of years, the patient will be exposed to a large dose of radiation and there is now clear evidence that this increases the risk of developing breast cancer. Second, it has been increasingly recognised that cosmetic factors such as rib hump play a large part in the requirement for treatment. The size of the rib hump is as important to the patient as the radiological position of the spine and yet is not measured at all by radiography. The availability of low-cost, radiation-free equipment to measure back topography would therefore be very useful to both surgeon and patient.


A new system developed to record and measure the shape of the back has been tested in regular use at the Nuffield Orthopaedic Centre in Oxford. It comprises an LCD projector and camera under the control of a computer. A striped pattern is projected onto the back and a single digital photograph is taken. Analysis of the distortion of the stripes relative to their pattern on a flat reference plane allows the three-dimensional shape of the back to be calculated. The main objective of the work covered by this grant was to validate the new automated system for the measurement of back shape in scoliosis. This included assessment of the reliability of the system for repeat measurements on a number of patients by the same and by different operators. The work also aimed to determine the natural variation that will occur in the calculated clinical parameters if a patient is measured several times. One of the parameters looked at, lateral asymmetry (a measure of the curve in the spine when looking at the patient from behind) can be compared with radiographic Cobb angle.

Outcome of research:

Validation and reliability testing of this new system was carried out by measuring the variability in repeat measurements of the same patient with a walk round the room between photographs (intra-rater) and repeat measurements where the stickers marking the bony landmarks are replaced by a second marker before the second photograph (inter-rater). The variability in lateral asymmetry was particularly examined because of the interest of most surgeons in having an equivalent parameter to Cobb angle without the risks that radiation poses. The intra-rater variability in lateral asymmetry was ±3.5º, lower than the variability in Cobb angle (±3º-9º). The interrater variability in lateral asymmetry was comparable to the upper range of interrater variability found in Cobb angle (±5º-10º). Lateral asymmetry cannot be used alone to estimate Cobb angle; in only 80% of cases, did lateral asymmetry measurements estimate Cobb angle to within 10º. However, lateral asymmetry may yet be used for monitoring the change in curvature of the spine and hence avoid unnecessary exposure to radiation. This is being followed up on. This new system has been installed in the scoliosis clinic at the Nuffield Orthopaedic Centre in Oxford. It is now in regular use, run by the radiographers. The database containing patient data is building up and a few patients have now been measured on three occasions.

An example of some results from the system together with a radiograph of the patient are shown in Figure 1.


Figure 1: Photograph of back with grid and bony landmark markers using the new system described above.


Inter-rater: In statistics, inter-rater reliability is the degree of agreement among raters.

Intra-rater: In statistics, intra-rater reliability is the degree of agreement among repeated administrations of a diagnostic test performed by a single rater.


Paper not published.


Share this page