Improving the usability of Medical Infrastructure Software: the redesign of a Linear Accelerator machine log

Medical Infrastructure Software in the National Health Service (NHS) is pervasive. This type of software is not regulated medical device software, it is behind the
scenes software that helps to coordinate the management of equipment that delivers or supports patient treatment.

This software is sometimes developed in-house by Healthcare Professionals such as Clinical Scientists & Technologists, in departments such as Medical Physics &
Clinical Engineering. However some of this software is not used as intended and exhibits a lack of use problem. Human-Computer-Interaction (HCI) is an area of Computer Science that aims to address these lack of use issues.

An example of this software is MachineLog, developed by Clinical Scientist & Technologists at the Northern Centre For Cancer Care (Newcastle upon Tyne NHS
Foundation Trust). MachineLog has diffused into The James Cook University Hospital (South Tees Hospitals NHS Foundation Trust) and it is intended that MachineLog
is used by Radiographers to record faults for medical equipment known as Linear Accelerators (Linacs), machines that are used to deliver radiation to a cancerous
tumour in cancer patients. However, in practice, MachineLog is almost never used and this creates a problem with recording and logging faults for the Linacs.

The work in this thesis aimed to understand the factors that influence usability and acceptability of Medical Infrastructure developed by Healthare professionals, and to re-design MachineLog as a Healthcare Professional who is familiar with software development but is not an expert in design or usability. To achieve this, the
literature was reviewed to explore the aspects of HCI that can make Medical Infrastructure Software usable in the NHS, leading to methods such as Contextual Design
for investigating the lack of use of MachineLog.

A Contextual Inquiry was performed for discovering the existing work of the Radiographers whilst they were using (or not using) MachineLog. The data from
this study was used to create personas, existing and new scenarios of use, and was further used to help communicate the requirements of a new prototype called
LinacLog. Furthermore, this data allowed us to conclude that the Radiographers will not use MachineLog in its current form because it does not align to their goals of treating patients and is a factor that influences acceptability. It was anticipated that LinacLog be a fully functional prototype, however, limitations due to Hospital
IT led to designing a low-fidelity prototype. LinacLog was then evaluated, despite encountering SARS-CoV-19 related difficulties.