Comparison of two booths for the early detection of skin cancer.
Comparison of two booths for the medical diagnosis of skin cancer in whole body photo imaging.
Category: Computer Assisted Diagnosis
Technologies: Image Processing, Color Analysis.
The purpose of this project is the comparison of two cabins based on their background colors. These cabins are used for the comprehensive study of the skin over the whole body, for the early detection of potentially deadly skin cancers. We have analyzed the impact of the cabin background in Image Processing tasks. In this summary we present the results with two patients (next figure) and a single cabin.
The next figure shows the color distribution of the two patients (see last figure) in the HSL space (Hue, Saturation, Luminance). This figure shows the variability of the HSL diagrams with patients. The first patient produces moderately saturated colors, while the second patient produces colors with low saturation, in addition to color shifts toward blue. A greater variability is expected for the universe of all patients and a statistical study of these diagrams using images of many patients would define the patient region on the HSL diagram.
The next figure shows the interior of a booth used in our experiments and the projection on the HSL space.
The next figure shows the superposition of patients and cabin in the HSL space.
The next figure shows the result of a color segmentation. We have achieved a good segmentation thanks to the color separation between patients and cabin, as shown in the previous HLS diagrams.
Elastic Registration of two Images Spaced in Time using Trac...
Elastic Registration of two images using Natural Skin Marker...