AI Supports Proactive Human-Centred Care

Healthcare faces an enormous challenge: how can we make healthcare both more efficient and human-centred? Because of ageing populations and increasing demand for personalised care, pressure is being put on existing systems. Here lies a major design challenge: how can we organise care processes in such a way that they add value for patients and enhance professionals’ job satisfaction at the same time?

In this interview, Maaike Kleinsmann (Professor of Design for Digital Transformation) and Geert-Jan Houben (Pro Vice Rector Magnificus AI, Data and Digitalisation) discuss how AI can contribute to digital transformation in healthcare. Can AI help pave the way from reactive to proactive, human-centred care? And how can we keep putting patients and healthcare professionals first in the process?

Focusing on care pathways

Maaike Kleinsmann is professor of Design for Digital Transformation at TU Delft (Faculty of Industrial Design) and Leiden University Medical Centre (LUMC). One of Maaike’s main research interests is the digital transformation of healthcare: “My approach focuses on care pathways, which are patient-centred and affect virtually all healthcare stakeholders. Care pathways connect the workflows of multiple healthcare providers: from GPs (primary care) to hospitals (secondary care) to academic hospital (tertiary care), as well as the patient journey (the 'route' the patient takes in a treatment pathway). Moreover, it is an effective way of managing healthcare costing, which is a key concern at the administrative level. You need to involve all these various layers to facilitate a real transformation while the support of digital technology will help optimise improvements to healthcare and patient health.”

Cross-fertilisation between sectors

Geert-Jan continues: “A human-centred digital transformation should benefit patients and healthcare professionals alike. And when we figure out personalised healthcare, we can explore whether the same underlying technology has potential applications in other transformations. It is very important for TU Delft researchers to look and compare across sectors, as cross-fertilisation between different disciplines will teach us a lot and will ultimately enable us to deliver a better end product. In an article by Maaike that I read, there was an excellent example illustrating how chatbots work. The same technology can be used in other sectors. The digitalisation solutions we discover here aren’t just specific: we also find and share generic solutions.”

“The same is true in reverse”, Maaike nods in agreement. Healthcare is a unique domain, but there’s a lot it can learn from other sectors, especially in terms of digitalisation. We’re in the midst of a transition that revolves around optimising synergy between people and technology. Actually implementing the technology is a complex matter  because care pathways involve a large number of people and technology changes how they work together.”  

More than a technological revolution

“AI can help change various healthcare processes. When used effectively, it can be used to make care more human-centred and efficient by enabling personalised services tailored to patients and caregivers. Without AI, digitising processes would be incredibly labour-intensive and cumbersome. In fact, the changes ahead are more than a technological revolution alone”, says Maaike. 

Remote patient management

Maaike gives an example of an application she is currently working on: Remote Patient Management. “That  means monitoring from a distance. Smartwatches, scales and blood pressure monitors can provide healthcare practitioners with a wealth of patient data without them ever having to see a patient in person, she explains. “While that’s a good thing, all that data collection would come to naught if the data went unused, and processing that much data is a massive job. That’s where AI comes in, because healthcare professionals have enough on their plate already without scaling a mountain of data. On the one hand, AI can flag which patients need what care and when, while saving a lot of time by eliminating check-ups, because so much data is already available. On the other hand, AI can also assist patients by providinginsights and tips on illness and recovery based on personalised and up-to-date data. It’s both highly complex and tremendously effective.”

AI and improved accessibility of care

“A basic example: patients with language difficulties are unlikely to benefit much from reports full of complicated medical jargon. There’s a lot of room for improvement there, as AI can easily draw up reports in language that patients can actually understand. This would be a big step towards more personalised healthcare, in a way that would be very difficult for doctors to do on their own. Besides, it would make it considerably easier for patients to play a bigger role in prevention and recovery”, Maaike explains.

Putting the patient first with AI

Maaike adds that the effective use of AI will put the patient first. “In the past, doctors had to make decisions based on history-taking and questionnaires. Now, they can look to data for support. Moreover, quality of care will also improve, as data is based on vital parameters. These data and insights can be used to make predictions but also to improve future care, which will benefit today’s patients, future patients and the healthcare system alike by lowering healthcare costs. On top of that, the ability to accurately monitor health data helps reduce the burden of disease for individual patients.” 

From reactive to proactive patients

What makes these technological developments so interesting is that they enable people to be proactive about their own health. Besides, data and suggestions are continuously personalised, leading to better and better solutions. Maaike notes that AI will help strike a better balance, giving both healthcare providers and patients much more control. “Technology and AI have myriad valuable, relevant applications in healthcare.” 

“In the next decade or so, patients will get a much bigger say in health-related decisions. They’ll be able to make well-founded decisions about their quality of life and what care they do and do not want. Ultimately, we will also have to accept that there are limits to what can be done”, Maaike notes. Geert-Jan adds that the goal is not just to transform healthcare, but everything around it. “We’re transitioning from a reactive approach to a proactive and preventative approach.”

Reactive versus proactive care

As Maaike explains, reactive care will still be the go-to approach in acute situations. “Chronically ill patients, for example, can monitor themselves proactively and can therefore often see incidents coming and act accordingly. This allows them to gain control of their own lives, which improves their quality of life. This will also make care more plannable, provided it is organised differently after the transformation.” Maaike even suggests that some jobs in healthcare may  have to be ended or scaled down: “How we work will change of course, but we’ll also have to avoid making healthcare even more expensive by mindlessly creating new possibilities. The advent of technology also means that we’ll have to cut back or even stop certain activities.”

Geert-Jan stresses the need to look at the bigger picture: “Some of the traditional tasks of the healthcare sector will shift to people’s immediate surroundings. AI and digitalisation are interesting because they effectively give patients access to a personalised healthcare assistant, and the same goes for other sectors. For instance, in the energy sector,  people can get personalised advice on their individual energy needs. In the end, this will only benefit society. AI is sure to change society in many ways, but it will also make key services more personal and human-centred in the process.”

Looking for more information on this topic?

More informatio on “Responsible design of human-centred AI and data-driven systems”.
More informatin on “AI for Health and Care”.

Or get in touch via ai-initiative@tudelft.nl