Dr Amber Hill, Research Grid Ltd and QMUL (Queen Mary University of London)

More than 80% of medical research data is still captured by pen and paper. This number is rising because of the increase in decentralised clinical operations post pandemic. Academic research institutions, in particular, struggle with the resources to address these types of manual operational challenges. Timelines in medical research add to this burden. For example, it is mandatory that this type of data is manually entered into a computer by a designated time, which is usually carried out by over-qualified staff, such as senior research nurses, PhD candidates, or senior investigators. This leads to slow processing, data entry errors, long lead times, reduced efficiency, and timeline failures at a rate of 85% in clinical research. This project’s proprietary machine-learning algorithm will automatically extract unstructured paper-form medical assessments and surveys, to streamline them into easy-to-read digital electronic data forms and accelerate administrative research processes.