Cloud machine learning with compressed data and unlabelled partial data

1) Brief description of the context of the research including potential impact

The research will focus on the deployment of machine learning models on the cloud to be used in real-time. The data can be compressed or incomplete as well as unlabelled. This research will be applied in the field of endoscopy where it can have a huge positive impact on the deployment of medical devices which are designed to work in real-time and in all cases, including the lack of data.

2) Aims and Objectives -The specific objectives are to:

Study the influence of using compressed and unlabelled partial data for machine learning. Study the difficulties in deploying a real-time model on the cloud. Study how these two problems can link together and propose different solutions to face these. 3) Novelty of Research Methodology

Doing machine learning with compressed and unlabelled partial data is unusual, especially in medical imaging where we often lack data. Combining this with cloud-deployed models is something that hasn't been effectively done yet so this research aims to know what is the best way to do it. 4) Alignment to EPSRC's strategies and research areas

This research is related to artificial intelligence technologies and medical imaging which are both EPSRC's research areas and aim to address a challenging problem in current endoscopy procedures. 5) Any companies or collaborators involved Odin Vision, a medical imagery company specialized in endoscopy.