Early detection of skin cancer by using mobile devices at point of care
A project led by Moi Hoon Yap at Manchester Metropolitan University.
Our aim is to use existing electronic health records (EHR) in early detection of skin cancer originated in different parts of the body. This relies heavily on patient vigilance to observe the changes in skin lesion, and on the completeness of EHR. To automate the monitoring process and speed up the analysis of EHR, we use mobile devices to enable documentation at the point of care. Our goal is to develop an automated framework using Artificial Intelligence (AI) for early diagnosis of skin cancer occurring in different anatomic sites based on the analysis of electronic health records. The algorithm includes the use of global patterns and deep learning method. For AI development, the ISIC 2019 dataset, which consists of over 23,000 images with metadata, will be used to train the machine. We will run the model on a mobile device to demonstrate its efficacy and efficiency for use at point of care. Finally, it will be evaluated on the clinical record at DermNet New Zealand (remotely) and Manchester dermatology clinics (subject to ethical approval). The computerised technique will be integrated into a mobile application to enable social distancing and reduce the spread of Covid-19 in clinics.
This work and its findings are presented in a video by the Project Investigator here: