Web-based total body surface area burned (TBSA-B) estimation tool

Introduction An accurate burn care is related to the ratio of total body surface area burned (TBSA-B) to total body surface area (TBSA). The estimation of a patient’s total TBSA is challenging due to the varying size and body proportions during growth. Accurate burn care consumes time, money, and resources. The treatment is related to the ratio of total body surface area burned (TBSA-B) to total body surface area (TBSA). For example, the decision on how to treat shock is based on the quotient of these two values. The formula to calculate the amount of fluid is also based on this ratio. Thus, the precise estimation is crucial. Common estimation methods (e.g., Lund-Browder charts) and rules (e.g., rule of nine and the rule of palm) are widely used and their weaknesses are well known and published. As cited by Neuwalder et al. [2], a study in 1985 showed, that the use of common surface estimation methods, even experienced burn physicians tend to overestimate TBSA-B by 29%, and less experienced staff by up to 49%. In [3] the rule of palm tends to result in an overestimation factor of 10-20%. The rule of nines [5], seems rather exact for certain persons, but even with this method overestimation of TBSA-B occur. The most precise commonly used method is the Lund- Browder charts as cited by Haller et al. in [1]. The Lund-Browder chat takes the different body proportions associated with different ages in account. But even this method is not perfect, and an overestimation was discovered. In [4] proportionally correct 3D children models based on anthropocentric measurements done by Med- Campus IV were generated and provided in Surface 3D, a newly developed web-based application whose aim, e.g. is a thorough and accurate burn estimation. The top image shows the different groups. The bottom left image shows the different body regions color coded. The chart visualizes the proportions and body part surface area in the different groups. Some body-parts have a higher proportion impact on the groups, thus are more important for the generation process. Material and Methods Surface 3D provides the following features: »» Proportional correct 3D models created in a previous study. »» A web-based burn annotation and wound size estimation tool. »» Two different annotation methods: ◊ Polygon-Painting: annotation is done directly on the model’s polygons; drawing resolution is restricted by polygon-size. ◊ Texture-Painting: annotation is done on a separate texture, without restriction on polygon-size. This enables precise wound documentation. »» Different burn degrees can be distinguished by their colors and their sizes are calculated in real time. »» Photos can be superimposed on the 3D patient model, facilitating the visual verification of marked areas. Results Surface 3D is developed with Angular, a modern platform to build applications with the web. This approach allows a platform and location independent usage. Surface 3D improves the characterization quality by switching the polygon-based method to a texture-based. The polygon-based approach is restricted to the polygon size. Surface 3D is using texture for e.g. determination of burned surface on the 3D model which increases the resolution and the natural presentation. To enable, precise three-dimensional (3D) body surface area measurements, Surface 3D integrated 12 proportional correct 3D models. A set of 24 body measurements was obtained from 2529 pediatric patients, aged 0-17 years. A correlation cluster analysis revealed key proportions that were used to an age partition with high proportion stability. For each age range a correct 3D model was generated using an optimization plug-in for a human character modeling program. The cluster analysis was resulted in six age groups (0-1, 1-2.5, 2.5-4, 4-6, 6-11.5, 11.5-18 years) with stable proportions. The median body-measurements were used to generate six proportional correct 3D models for each sex. Conclusion & Outlook Surface 3D is a web-based application for annotation and estimation of different surface characteristics using modern web technologies like Angular and three.js. The application is developed modular. Thus, it could be easily adapted for other applications and derivations of Surface 3D providing specialized user interfaces depending on the use case. A modified application for wound and dermatology documentation is also proposed. This web-based solution benefits clinics, who only have a few burn treatments. Accurate burn size estimation needs to be trained, thus the quality with Surface 3D will increase. The accuracy depends on the precise and perfect fitting models. In a previous study [4] proportions correct models were generated obtained from pediatric patients, aged 0-17 years. This study could be adopted to patients, aged 0-99 years. Other proportional researches are needed, to generate correct obese or under- weighted models. Future research is directed towards automated model adaption. In additional to the standard patient parameters, like name, age, gender, weight and height, the pose and the shape of the patient should be adapted for generating a perfect fitting 3D model. Simple 2D RGB images are used, applying Deep Learning approaches.