A Network Resource-Aware Hyperspectral Imaging-Based Telemedicine Architecture for Rural, Remote, and Underserved Communities

The limited access of rural, remote, and underserved communities to quality specialist healthcare services is a major obstacle to the attainment of the United Nations Sustainable Development Goal 3 of good health and well-being for all by 2030. Telemedicine can help mitigate this challenge by enabling specialist healthcare services to be provided to patients in rural communities, from urban healthcare centres, over the Internet. Hyperspectral imaging is a new medical imaging modality that facilitates specialist healthcare such as skin cancer diagnosis and, potentially, latent tuberculosis screening. Rural and remote communities tend to be plagued with poor telecommunications infrastructure which is a key component of any Telemedicine system. Thus, we designed a hyperspectral (HS) imaging-based Telemedicine architecture that compresses the HS images based on the available network capability of a rural healthcare centre, minimizing transmission time and image quality loss. We tested our Telemedicine architecture at the Carnarvon hospital, located in a rural, remote, and underserved community in the Northern Cape province of South Africa. Tests with our architecture revealed 50% HS image compression to be optimum for maximising the limited network resources in Carnarvon across four South African mobile networks. Our results present further evidence of the usefulness of Telemedicine as a viable and effective modality to increase healthcare access to rural, remote, and underserved communities, thereby contributing towards the attainment of the UNSDG3 goal of equitable healthcare access for all by 2030.