Skin cancer is the most common cancer in the world. Clinical diagnosis of skin cancer is made by dermatologists using an instrument called a dermatoscope which enables close-up visualization of the skin. Dermatoscopes are expensive and require clinical expertise for use in daily practice. Sharing dermatoscopic images remotely can allow non-dermatologists collaborate with experts elsewhere, but the cost and limited availability of digital dermatoscopes prevents this from happening routinely. We will develop DeepSkin, a hand-held, portable and easily understood digital dermatoscope that is cheap, simple and capable of storing and transmitting dermatoscopic images.
Gold standard skin imaging techniques use multispectral and polarimetry data to supplement plain image analysis, but the equipment to do this is expensive and bulky. DeepSkin will be able to analyse skin lesions in multiple spectra from ultraviolet to near-infrared and in multiple polarities. Despite its advanced capabilities this will cost orders of magnitude less than legacy equipment used at present.
Steve Smith, email@example.com
A clinical dermatologist with experience in computational image analysis and modeling. Project lead and coordinator, will put the project in a clinical context as well as coding, building and designing the dermatoscope.
Youssef Badr, firstname.lastname@example.org
Reading Chemical Engineering in the Department of Chemical Engineering and Biotechnology. Will lead on the electronics work and provide engineering expertise.
Joseph Wu, email@example.com
PhD candidate researching the early diagnosis of cancer with research background studying the genetic regulation of skin cancer. Will ensure the project is consistent with recent advances in the biology of skin cancer and formally analyse cost-effectiveness and diagnostic accuracy.
Project report and documentation on Github
Bill of Materials