Quantitative and kinetic evolution of wound healing through image analysis.

To define a healing function based on parameters measured on digitized images of wounds, and to use it to compare the rate of healing of two skin graft donor sites, one treated with petrolatum gauze (Pg) and the other with a topical preparation containing alginates (A). Digital photographs of donor sites (depth 0.6 mm) taken every two days between day 6 and day 12 were analyzed blind using the same algorithm, following changes in color and homogeneity. Analysis of variance was used to identify those parameters that changed during healing. The healing function was constructed using measurements made in six patients (group 1) randomly chosen from ten requiring skin grafts, and was applied and validated using data from the remaining four patients (group 2). The results given by this healing function were compared with those provided by principal component analysis. The most significant healing parameters were those measuring wound homogeneity, and our healing function reflects how these change with time. The time-dependent curves of the function calculated for groups 1 and 2 matched well enough to be considered as being derived from the same set of measurements. The results given by this healing function explained, by analogy, the meaning of the first principal component of principal component analysis. From day 6 to day 12, the healing function followed the same time-course for the Pg and A treatments, but healing was achieved significantly earlier (4 days, p < 0.03) with A. This suggests that the effect of A on wound healing is achieved in the first six days, before the visual changes from epidermalization are analyzable.