A computer-based method for precise detection and calculation of affected skin areas

The aim of this study was to describe and validate a method to obtain reproducible and comparable results concerning extension of a specific skin area, unaffected by individual differences in body surface area.

A phantom simulating the human torso was equipped with three irregular areas representing the increasing extension of an affected skin area over time. A large sheet of flexible calques paper was placed at the phantom, and five clinicians copied the three irregular shapes two times, resulting in 60 copies. Subsequently, a digital photograph was taken of the calques papers with a clinical ruler placed at the margin. The images were postprocessed and measured in the program 'ImageJ' by two observers. An exact area measurement of the three irregular shapes was performed for comparison.

We found an interobserver variation of 0·36% when comparing the measurements of all three areas. Comparing observer measurements with the exact areas size, we found an underestimate of 2·52%. We observed a tendency that the discrepancy in measurement increases when the measured area decreases.

We find this method accurate, reproducible and easy to use. The presented method can be of help when documenting psoriasis and other dermatologic conditions as well as when exploring the effects of new types and variations of ultrasound-guided peripheral nerve blocks - especially in study volunteers.