The Approach for Quantification Assessment for Port-Wine Stains in Three-Dimensional and Color Space

摘要

Quantification evaluation and outcomes analysis of photodynamic therapy for port-wine stains (PWS) is usually rely on the clinician subjective assessment. The aim of this study is focused on an objective assessment model for port-wine stains during the successive treatment sessions of photodynamic therapy. The assessments of the outcome were assessed in the clearance of the lesion area before and after treatment and the trace of the lesion across the entire color space in this paper. Firstly, 3D point clouds containing the lesion coordinates and color was achieved. Then, the obtained unorganized point cloud is projected onto a specified plane to generate an organized grid, and the depth image on that plane can be also obtained. Thirdly, the lesion region is separated by using the SLIC superpixel segmentation combined with the color feature of the lesion, and the lesion was re-projected to the original three-dimensional point cloud. In the end, depending on the color and location information of the 3D lesion skin points, triangulation reconstruction was performed by the greedy projection triangulation algorithm to calculate the irregular surface area of the lesion skin. The result show that the lesion skin area based on three-dimensional space are stable and reliable regardless of the scanning angle or distance difference and the chrotnaticity distribution in the color space can show the recovery trend of the lesion skin at different treatment stages, so three-dimensional scanning of patients can be an approach to monitor the progress of an individual's PWS following treatments accurately and objectively.