Quantitative and rapid estimations of human sub-surface skin mass using ultra-high-resolution spectral domain optical coherence tomography.

Abstract

Non-invasive and quantitative estimations for the delineation of sub-surface tumor margins could greatly aid in the early detection and monitoring of the morphological appearances of tumor growth, ensure complete tumor excision without the unnecessary sacrifice of healthy tissue, and facilitate post-operative follow-up for recurrence. In this study, a high-speed, non-invasive, and ultra-high-resolution spectral domain optical coherence tomography (UHR-SDOCT) imaging platform was developed for the quantitative measurement of human sub-surface skin mass. With a proposed robust, semi-automatic analysis, the system can rapidly quantify lesion area and shape regularity by an en-face-oriented algorithm. Various sizes of nylon sutures embedded in pork skin were used first as a phantom to verify the accuracy of our algorithm, and then in vivo, feasibility was proven using benign human angiomas and pigmented nevi. Clinically, this is the first step towards an automated skin lesion measurement system. In vivo optical coherence tomography (OCT) image of angioma (A). Thin red arrows point to a blood vessel (BV).