Implementing and evaluating mobile and artificial intelligence technologies in UK secondary healthcare
File(s)
Author(s)
Aggarwal, Ravi
Type
Thesis
Abstract
The deployment of mobile and artificial intelligence (AI) technologies has the potential to revolutionise healthcare delivery in the UK. However, despite significant interest from large technology companies, healthcare organizations in the UK have been slow to adopt novel digital health technologies (DHTs). Barriers to the widespread adoption of next-generation technology in the NHS include a lack of effective implementation strategies and regulatory standards, as well as a paucity of high-quality evidence regarding the effectiveness, safety, and utility of these technologies in clinical practice. To address these barriers, this thesis seeks to perform a comprehensive evaluation of the development and implementation of a novel clinician facing mobile application called Streams, which was developed in collaboration with industry partners, at a multi-site UK hospital; the first of its kind in the NHS.
An introduction to mobile and AI technologies is first presented, along with a summary of the challenges encountered with the implementation and evaluation of these technologies in healthcare settings. Pre-clinical evaluation of the performance of AI technology in healthcare was then examined through a large systematic review and meta-analysis. Findings from this review included significant deficiencies in design, conduct and reporting of AI studies, leading to high risk of bias and potential overestimation of algorithmic performance in the literature to date.
The environment and readiness for the implementation of mobile and AI technology in UK hospitals was then explored by assessing the perspectives of patients and end-users using a mixed-methods approach. From the patient survey, a lack of knowledge about AI was identified amongst patients, however the potential benefits and appetite for novel digital technologies in healthcare was shown amongst both stakeholder groups...
An introduction to mobile and AI technologies is first presented, along with a summary of the challenges encountered with the implementation and evaluation of these technologies in healthcare settings. Pre-clinical evaluation of the performance of AI technology in healthcare was then examined through a large systematic review and meta-analysis. Findings from this review included significant deficiencies in design, conduct and reporting of AI studies, leading to high risk of bias and potential overestimation of algorithmic performance in the literature to date.
The environment and readiness for the implementation of mobile and AI technology in UK hospitals was then explored by assessing the perspectives of patients and end-users using a mixed-methods approach. From the patient survey, a lack of knowledge about AI was identified amongst patients, however the potential benefits and appetite for novel digital technologies in healthcare was shown amongst both stakeholder groups...
Version
Open Access
Date Issued
2022-10-06
Date Awarded
01/03/2023
License URL
Advisor
Martin, Guy
Ashrafian, Hutan
Darzi, Ara
Sponsor
National Institute for Health Research (Great Britain)
Publisher Department
Department of Surgery & Cancer
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)