Search

Background Accurate estimation of a burned area is crucial to decisions about fluid resuscitation, surgical options, nutritional support, and prognosis. Widely used clinical methods to estimate a burn area are two-dimensional. They do not consider age, sex, body mass, physical deformities, or other relevant factors. Computer-aided methods have improved the accuracy of estimating burned areas by including data analysis and reducing subjective differences. Threedimensional (3D) scanning allows us to determine body dimensions rapidly and reproducibly. We describe an individualized, cost-efficient, portable 3D scanning system, BurnCalc, that can create an individual 3D model and then calculate body surface area (BSA) and the burn area accurately and quickly. Methods The BurnCalc system was validated by verifying the accuracy and stability of BSA calculation. We measured 10 regular objects in experiment 1, using Student's t-test and the intraclass correlation coefficient (ICC) in the analysis. In experiment 2, artificial paper patches of known dimensions were attached to various parts of the body of 40 volunteers. Their sizes were then calculated using BurnCalc. The BurnCalc data were compared to actually measured values to verify accuracy and stability. Total BSAs of these 40 volunteers were also calculated by BurnCalc and compared to those derived from an accepted formula. In experiment 3, four experts using Chinese Rule-of-Nines or Rule-of-Palms methods calculated the percentages of the total BSA in 17 volunteers. Student's t-test and ICC, respectively, were used to compare the results obtained with the BurnCalc technique. Results Statistically, in experiment 1, p = 0.834 and ICC = 0.999, demonstrating that there was no difference between the BurnCalc and real measurements. Also, the hypothesis of null difference among measures (experiment 2) was true because p > 0.05 and ICC = 0.999, indicating that calculations of the total BSA and the burn area were more accurate using the BurnCalc technology. The reliability of the BurnCalc program was 99.9%. In experiment 3, only the BurnCalc method exhibited values of p > 0.05 (p = 0.774) and ICC = 0.999. Conclusions BurnCalc technology produced stable, accurate readings, suggesting that BurnCalc could be regarded as a new standard clinical method. [ABSTRACT FROM AUTHOR]