Effects of smartphone sensor characteristics on dermatoscopic images: a simulation study

Abstract

Dermatoscopes are commonly used to evaluate skin lesions. The rising incidence of skin cancer has led to a wide array of medical imaging devices entering the market, some of which provide the patient the ability to analyze skin lesions themselves. They usually come in the form of smartphone attachments or mobile applications that leverage the optics of the smartphone to acquire the image; and in some cases, even give a preliminary diagnosis. In this digital age these devices look to ease the burden of having to visit a dermatologist multiple times. While these attachments are no doubt very useful, the image sensors used within smartphones are limited in terms of how much information they can process and effectively output to the user. Smartphone sensors are also very small which can result in a less detailed image as opposed to one from a professional camera. Our work is focused on the analysis of the information lost due to the known limitations of smartphone sensors, and its effect on the image appearance. This analysis has been performed using a virtual simulation pipeline for dermatology called VCT-Derma, which contains a module for a proprietary dermatoscope whose optical stack parameters will be adapted to the smartphone sensor specifications mentioned in this manuscript. This manuscript also describes the necessary sensor parameters required for adapting the simulation model, the software used along with any assumptions made, perceived differences in the resulting images, as well as the direction of the ongoing work.