The practical considerations of applying AI in healthcare settings

Digital innovation will drive the healthcare sector forward, but its practical next steps will be the true catalyst for change.

In September 2021, The National Artificial Intelligence (AI) Strategy was launched by the UK government, marking the start of a step-change for AI in the UK. It recognised the power of AI to increase resilience, productivity, growth, and innovation across multiple sectors including healthcare.

That milestone was the result of many years of developing, testing, and trialling AI technology. The goal: to make a positive impact on patient care from diagnosis through to treatment, as well as to help improve innovation and productivity within the healthcare setting.

Take lung cancer as an example.

It has the highest mortality rate of all cancers in the UK, so the NHS has set the ambition of diagnosing three in four cancers at stage 1 or 2. Ground-breaking AI software has been developed to help meet these targets by detecting and helping diagnose lung cancer earlier on in the patient care journey.

Not only might this technology crucially help to increase survival rates, but it also reduces unnecessary invasive procedures on healthy patients, helping to free-up NHS time and resources for patients who need it most.

It is a promising time following years of progress in AI research and development. However, to make early diagnosis targets a reality, AI needs to be widely adopted across multiple NHS clinical settings.

The next challenge is to take the time and preparation needed to ensure healthcare systems and operations are set up to truly maximise the impact of AI technology.

Harnessing the value of data

With the support of AI, clinicians gain access to new data to make better informed patient care decisions and improve patient outcomes. Learning how to extract the value from the analytics and insights presented to them by AI software is the first step of integration into the healthcare setting.

For example, Optellum’s lung cancer AI software gives clinicians a Lung Cancer Prediction score based on imaging AI that has been trained to distinguish benign lung nodules from malignant ones.

Widespread training is required to enable the clinicians to interpret the score and integrate the software within existing care procedures. Doing so will empower optimal decision-making so diagnosis and treatment can be provided before the disease has developed.

Spotting early diagnosis needs an integrated approach to care

To intercept lung cancer at the early stages, there needs to be a faster pathway from referral to diagnosis between GPs in local health and care systems and clinicians within hospital settings. This requires a new upstream approach to diagnosis, which begins as early on as seeing a GP.

In addition, a key opportunity for catching lung cancer sooner is in the hundreds of thousands of Computed Tomography (CT) scans NHS patients already have each year for other conditions, such as heart scans.

CT scan data needs to be shared and communicated across healthcare settings and systems as well as integrated with AI decision support software to maximise the opportunities to diagnose lung cancer sooner.

This unified approach requires the healthcare workforce to change how it currently works across settings so that a timely lung cancer diagnostic pathway can be created to intercept cancer earlier where treatment is likely to be more successful.

It requires the healthcare workplace to update existing job roles and descriptions, create new daily tasks and communication channels, as well as provide further support and training.

Networking and collaboration

The integration of care settings will also require greater communication between primary and secondary care professionals. The sharing of CT scans and patient tracking data will help facilitate information sharing to progress patient care as needed for early detection and diagnosis.

Co-location of professionals across care settings may be a beneficial structural change to help foster collaboration and progress patient care.

Looking ahead

With a combination of clinically trialed AI technology that has proven results, and the right practical support in place across healthcare settings, early detection, and diagnosis of illnesses such as lung cancer will have a life-changing impact over the long term.