Louise Flintoft, Associate Director of Onyx Health, examines the role of artificial intelligence in healthcare in the future, following a report from Health Education England which found that 56 new technologies are expected to be deployed at large scale in the NHS next year.
Artificial intelligence (AI) has seen a post-pandemic boom. But what does this mean for its future use in healthcare? Will we be talking to robot doctors and nurses in the years to come? How will AI fit into the future of healthcare?
The use of AI in healthcare is increasingly important. While not a new phenomenon, it is one of many healthcare trends accelerated by the pandemic.
In response to this acceleration, Health Education England (HEE) has released the UK’s first roadmap for the use of AI in the National Health Service (NHS). He revealed that there are 56 AI technologies planned for large-scale deployment next year, 77% of which will be used in secondary care.
Digital transformation is also a priority area identified in the NHS long term plan, where AI will act as a catalyst for change. This will involve greater use of technology like chatbots and virtual assistants to change the way services are delivered.
AI chatbots and digital dialogue
Chatbots can perform a variety of traditional healthcare functions, including booking appointments, ordering prescriptions, checking symptoms, and providing basic medical information and advice.
Babylon HealthA striking example is the use by an AI chatbot of a consultation service based on the patient’s medical history and standardized medical knowledge. Patients can first report symptoms on an app, which is matched against an official medical database, using voice recognition technology. They will then receive a recommended course of action based on the information provided. The success of their work led them to partner with the NHS to develop a digital triage system for referrals and the distribution of health advice during the COVID-19 crisis.
There is evidence to suggest that AI chatbots could be used to eliminate stigma with sensitive health conditions like sexually transmitted infections (STIs). Research from the University of Westminster suggests that patients prefer talking to an AI chatbot rather than a GP about their sexual health. In addition to the transformation of service delivery, there is also evidence to suggest that AI is increasingly being used in NHS diagnostics.
Using AI to deliver a diagnosis
the British Heart Foundation recently reported that AI is being used as a tool to improve patient care by detecting heart disease faster and more accurately. The imaging technology works by analyzing heart MRIs while the patient is still in the scanner. It only takes 20 seconds for the scan results to be analyzed, compared to 13 minutes for a doctor to manually evaluate the images. The new tool can also identify problems with heart structure and function with 40% better accuracy than the human eye. AI technology like this can play a particularly important role in helping time-strapped clinicians as health services struggle to recover from the critical care backlog caused by the pandemic.
AI imaging technology is also central to the National Optimal Stroke Imaging Pathway (NOSIP). It is used to speed up the clinical decision-making process by providing real-time interpretation of scanned images faster and more accurately. Diagnostic and treatment interventions are very urgent for stroke patients, and technology like this could help save lives.
Fight against health inequalities
AI-led innovation is also being used to address various structural systemic issues related to healthcare access. NHSX AI Lab (now part of the NHS Transformation Directorate) and the Health Foundation are currently undertaking joint project work using AI to tackle racial and ethical inequalities in healthcare in the NHS.
Several technological innovations are in the works, such as I-SIRch, which uses AI technology to facilitate the identification of clinical factors that lead to adverse maternity incidents among ethnic minority mothers. Black women are five times more likely to die in the UK due to pregnancy complications than white women. The I-SIRch will assess how different factors combine to cause these maternity problems and help design better forms of intervention.
AI diagnostics for ethnic minorities are also being developed to improve the accuracy of screening for diabetic retinopathy. Recent studies have shown that people of Indian, Pakistani, Bangladeshi and Caribbean ethnic groups are at increased risk of developing diabetic retinopathy compared to Caucasians and receive poorer diagnosis due to their different retinal composition. In addition to revolutionizing diagnostic methods, AI technology is also being used in health services to address some of the underlying healthcare inequities exposed by the pandemic.
An increased role for AI in a post-pandemic healthcare system
Whatever the future of AI in the NHS, it will clearly play an increased role in the post-pandemic healthcare system. The latest innovations in AI have the potential to transform the way healthcare and diagnostic services are delivered and address some of the underlying healthcare inequities exposed by the pandemic.