After the outbreak of the pandemic, it is clearer than ever that COVID-19 has hit health systems that are unprepared to deal with major health crises. The pandemic unveiled already existing weaknesses, which added up to the struggles health systems found themselves with. COVID-19 will have profound implications on economic progress, trust in governments and social cohesion but, above all, on patients’ health.

In particular, the struggles healthcare systems have had have resulted in the care of millions of patients being postponed or reduced. More specifically, patients with chronic conditions have been disproportionately affected by this crisis and had to deal with increased risks of severe illness and long delays in accessing care. Therefore, there is an urgent need to transform European health systems to ensure that potential future pandemics will not cause as much harm as the current one. Moreover, after the pandemic, priority needs to be given to ensuring that the highest possible level of care is always delivered to patients in line with the latest innovations and technological updates.

AI can save time for healthcare professionals

Digital technologies have indeed the power to upgrade European health systems by creating new models of care, which are predictive and preventative. In particular, Artificial Intelligence (AI) technologies do have the potential to assist European health systems in responding to major challenges they face by improving the ways they function. We have been hearing it over and over: European health systems face shortages of critical medical professionals, long waiting times, rising demand for services driven by an ageing population as well as financial constraints.

The trustworthy application of AI has the power to ease some of these constraints by reducing the time healthcare professionals (HCPs) spend on repetitive tasks, in order to allow them to spend more time on high-value activities for the benefit of patients. In fact, evidence shows that AI applications have the potential to free up to 1.8 billion hours every year, which is the equivalent of having 500,000 additional full-time healthcare professionals. Here is how different AI tools could help us achieve this objective.

Virtual health assistance

Virtual health assistance, for example, could diminish the burden of administrative tasks on doctors and medical personnel by reducing documentation burdens, extracting medical information, helping doctors with general administrative tasks and answering routine patient questions.

Personalised applications

Personalised apps, including nurse applications, could be especially beneficial for the treatment of chronic diseases. For example, the patient is guided through the treatment process by a virtual nurse assistant, which personally monitors the patients’ health conditions, keeps track of doctor appointments, and predicts follow-up treatments. The virtual nurse assistant also provides patients with instant feedback and supports them with notification alerts, ensuring they adhere to their medication.

Robotics

Robotics could also significantly reduce costs and improve financial resources of hospitals in different ways. For example, AI-enabled robot hands can perform new surgical techniques by employing data from past operations. Auxiliary robots can take care of restocking supplies, transporting medical equipment, and cleaning and disinfecting patient rooms.

AI can save lives

Evidence shows that AI technologies can save approximately 400,000 lives yearly, which equals the population of a medium-sized city, or almost two-thirds of Luxembourg. Wearable AI applications have the largest impact, saving up to 313,000 lives, followed by AI applications in monitoring (42,000 lives) and imaging (41,000 lives). It is important to note that all the mentioned benefits that AI can have in healthcare greatly rely on citizens and patients donating their data, as health data are foundational for the below AI tools to be effective and save lives.

Wearables

Wearables can be used in a variety of settings, including fall prevention, diagnosing cardiovascular disease, treating diabetes or chronic obstructive pulmonary disease, remotely monitoring a patient’s vital signs, and tracking patient activity pre- and post-surgery to monitor recovery. Examples of wearables are accelerometer bracelets or smart belts, which combined with an AI algorithm, allow the accurate prediction of fall events pre- and post-impact. Another example is the use of a combination of smartwatches with electrocardiograms and AI algorithms to detect atrial fibrillation and cardiac arrhythmias, which are important indicators of heart failure. Finally, AI can improve the quality of life of diabetes patients and increase the accuracy of glucose measurements through AI-enabled continuous glucose monitors - wearable devices tracking blood sugar levels day and night to predict highs and lows and providing tailored health advice to patients.

Monitoring

Thanks to a wide variety of AI applications, patients can be followed remotely without having to be in a healthcare facility. This could enable the detection of health issues 24/7 and trigger a timely response when needed.

Imaging

AI can also have a tremendous impact on the interpretation of medical images in digital pathology by detecting, diagnosing and monitoring multiple pulmonary, cardiac and oncological pathologies. Furthermore, it can play a role in image acquisition and reconstruction, as well as in video processing to guide surgeons during procedures and 3D imaging. Another AI application in medical imaging is the early detection of coronary artery diseases as well as the screening, diagnosis and treatment of breast cancer.

Final thoughts

Although AI cannot constitute the solution to all the problems healthcare systems face, it does have the great potential of improving human wellbeing by enabling a shift to proactive disease prevention and by empowering patients and further engaging them in care decisions. Trust is fundamental to ensure a human-centric approach to AI: a governance framework for trustworthy AI is needed as well as trust in AI from citizens and policymakers. It goes without saying that the algorithms AI approaches need to be subjected to transparent standards in approval procedures. In addition, the patients’ health data, on which AI tools make decisions, must be complete, correct and protected. In this way, it will be possible for trustworthy AI innovation and adoption to bring high-quality care for European citizens and build patient-centred models of care, which are preventative and predictive. [1]

[1] The source of this blog post is the MTE-Deloitte report “The socio-economic impact of AI in healthcare”, October 2020.

By Michael Strübin, Director, Digital Health at MedTech Europe in collaboration with Giulia Nava, Digital Health at MedTech Europe