AI has taken the world by storm in recent months, with the development of generative AI and new tools such as ChatGPT bringing the technology to great prominence.
What most people don’t realise however, is that AI and machine learning (ML) is already commonplace in many other areas of our lives. From morning until night we are already encountered by AI and it is a regular touchpoint for many of us – often without even knowing it. From unlocking your phone using FaceID, to using smart home devices, to the algorithms on social media and streaming devices, AI is a regular and everyday feature of our lives.
But what’s next for AI – and how is the healthcare industry in particular utilising AI to improve patient care and health resources? Again, great strides have been taken in recent months to introduce AI and machine learning to the health industry, and the results are positive and transformative. AI and machine learning is being used to diagnose diseases, provide personalised treatments, and improve patient outcomes and quality of care.
By leveraging digital technology, new patterns in patient data can be identified that may have been overlooked by traditional methods. Skin cancer and breast cancer are two areas which are already benefiting from analyses using AI which can spot anomalies better than humans can.
As with any new technology disrupting the status quo however, some people have concerns and queries about the impact of introducing new tools. This piece discusses the most common misconceptions around AI and aims to bust the five most common myths around its use in healthcare.
Busting the five biggest myths about AI / ML in healthcare
1. AI will replace human doctors
This is one of the most common fears about AI in healthcare. But the reality is that AI is not going to replace human doctors anytime soon. It might be the case that AI can perform impressive feats like diagnose diseases and recommend treatments – and we have seen this development in recent years – but AI/ML still lacks the ability to empathise with patients and provide the human touch that is so crucial in healthcare. Getting experienced and skilled clinicians to work with AI and leverage it to help patients is where the most potential lies.
2. AI will eliminate healthcare jobs
More generally, there is a fear that AI will replace other healthcare jobs, leaving many people unemployed. On the contrary, while it's true that AI has the potential to automate certain tasks that were previously performed by humans, such as administrative work and data analysis, it also has the potential to create new job roles in areas like AI development, data analysis, and patient care coordination. It’s more likely that AI will complement existing clinical staff and help to continuously enhance and improve the delivery of patient-centred care. Healthcare providers can then spend more time and resources caring, and less time on spreadsheets or administrative tasks that can be delegated to machines.
3. AI will make healthcare more expensive
Another common myth about AI is that it will make healthcare more expensive. This is simply not true. In fact, AI has the potential to reduce healthcare costs by improving efficiency and reducing the need for unnecessary tests and procedures. For example, AI-powered predictive analytics can help identify patients who are at high risk of developing a certain condition, allowing doctors to intervene before the condition becomes severe and costly to treat. This is HN’s model, which has demonstrated via a robust clinical trial that it can improve patient care while simultaneously reducing costs for the health system.
4. AI will make healthcare less secure
Some people are concerned that the use of AI in healthcare could compromise patient data and lead to security breaches. However, in practice AI often makes healthcare more secure, by detecting and preventing potential data breaches. For instance, AI algorithms can detect anomalous behaviour that could indicate a security breach. Moreover, AI can also minimise the use of healthcare data, by only utilising the most important and pertinent information from the patient record. In practice, companies such as HN that develop AI for health are strongly motivated to protect your data as much as possible. Regulation in the UK is stringent and imposes high minimum security and information governance requirements for any company that develops AI technologies.
5. AI is infallible and unbiased
Finally, AI is often perceived as a perfect and objective decision-maker. However, AI can only be as accurate and unbiased as the data and algorithms that are used to train it. If the data is biased, then the AI will be biased as well. This means that it's important for healthcare systems to use high-quality, unbiased data to train their AI algorithms, and further for these algorithms to be regularly validated and updated, to ensure both their performance and their fairness.
What’s next?
While there are certainly some myths and misconceptions about AI in healthcare, it's clear that AI has the potential to revolutionise the industry in many positive ways. However, to address concerns we must approach AI with a critical eye, and ensure that it's used ethically, transparently, and in the best interests of patients. With the pressures facing modern European health systems following the Covid-19 pandemic resulting in long waiting lists for care and staffing crises, it’s time to embrace this new technology and explore its full potential to improve lives.
AI evidence in healthcare
One area where AI and ML can be particularly helpful is in the application of ‘predictive care’. By analysing data from patients’ medical histories, AI tools can provide valuable insights that allow healthcare providers to anticipate and prepare for the patient's future needs. This can lead to more efficient and effective planning of care delivery, and ultimately, better patient outcomes.
HN (Health Navigator Ltd) are pioneers in applying AI and ML for predictive and preventative healthcare. HN can predict eight out of ten patients who will require unplanned care and prevent one in three unplanned care events for supported patients.
HN has carried out extensive research on its use of ML to help identify patients at high risk of clinical crisis and hospital care. HN’s AI-powered predictive analytics tool – HN Predict – is able to identify and prioritise people at risk of worsening health conditions in real-time by analysing routinely collected patient records. The algorithm hones in on individuals who are likely to have three or more unplanned hospital bed days within the next six months, or who will require an increase in GP-led care.
This helps to manage emergency care demand in a more proactive and predictive manner away from the front-door, to allow resources to support additional elective activity. Recent publications show promising results that visits to hospitals and ultimately deaths could be prevented, particularly by enabling early treatment of acute episodes of their long term conditions.
Find out more about HN’s AI capabilities here.