Date of Award

Spring 1989

Document Type

Thesis

Degree Name

Master of Science in Biomedical Engineering - (M.S.)

Department

Biomedical Engineering Committee

First Advisor

Stanley Martin Dunn

Second Advisor

Peter Engler

Third Advisor

David S. Kristol

Abstract

The need for a reliable and accurate method for assessing the surface area of burn wounds currently exists in the branch of medicine involved with burn care and treatment. The percentage of the surface area is of critical importance in evaluating fluid replacement amounts and nutritional support during the 24 hours of postburn therapy. A noninvasive technique has been developed which facilitates the measurement of burn area. The method we shall describe is an inexpensive technique to measure the burn areas accurately.

Our imaging system is based on a technique known as structured light. Most structured light computer imaging systems, including ours, use triangulation to determine the location of points in three dimensions as the intersection of two lines: a ray of light originating from the structured light projector and the line of sight determined by the location of the image point in the camera plane. The geometry used to determine 3D location by triangulation is identical to the geometry of other stereo-based vision system, including the human vision system.

Our system projects a square grid pattern from 35mm slide onto the patient. The grid on the slide is composed of uniformly spaced orthogonal stripes which may be indexed by row and column. Each slide also has square markers placed in between time lines of the grid in both the horizontal and vertical directions in the center of the slide. Our system locates intersections of the projected grid stripes in the camera image and determines the 3D location of the corresponding points on the body by triangulation.

Four steps are necessary in order to reconstruct 3D locations of points on the surface of the skin: camera and projector calibration; image processing to locate the grid intersections in the camera image; grid labeling to establish the correspondence between projected and imaged intersections; and triangulation to determine three-dimensional position. Three steps are required to segment burned portion in image: edge detection to get the strongest edges of the region; edge following to form a closed boundary; and region filling to identify the burn region. After combining the reconstructed 3D locations and segmented image, numerical analysis and geometric modeling techniques are used to calculate the burn area. We use cubic spline interpolation, bicubic surface patches and Gaussian quadrature double integration to calculate the burn wound area.

The accuracy of this technique is demonstrated The benefits and advantages of this technique are, first, that we don’t have to make any assumptions about the shape of the human body and second, there is no need for either the Rule-of-Nines, or the weight and height of the patient. This technique can be used for human body shape, regardless of weight proportion, size, sex or skin pigmentation.

The low cost, intuitive method, and demonstrated efficiency of this computer imaging technique makes it a desirable alternative to current methods and provides the burn care specialist with a sterile, safe, and effective diagnostic tool in assessing and investigating burn areas.

Share

COinS