How can crisis management be optimised while taking available resources into account? Understanding a complex system and anticipating how it may evolve requires a combination of data, models and simulation. Digital twins enable organisations to explore different scenarios and assess the potential consequences of a decision.
In healthcare, a patient’s care may take place across several settings, from home to hospital, and involve a range of professionals. At each stage, different tools generate data that can support their care. How can we ensure that this information, and the models using it, can support patients throughout their care pathway?
This is one of the challenges addressed by the “From Chips to Healthcare Services” initiative, supported by the European Commission. It brings together several partners with a shared objective: to enable technologies, data and healthcare services that are currently developed separately to work together more effectively.
As part of this initiative, Coruscant, Willing’s applied AI department, is leading a demonstrator focused on patient digital twins. Together with its partners, the Coruscant team is exploring how a digital representation of a patient could be enriched and reused throughout their care pathway, rather than rebuilt at each stage.
Towards a shared representation of the patient
A digital twin is a virtual representation of a real-world system that evolves based on data. When applied to a patient, it can help monitor their condition, anticipate potential developments and support healthcare professionals in their decision-making.
Today, these tools are usually designed for a specific setting, such as home monitoring, a hospital procedure or rehabilitation. Each can be relevant within its own area, but the data and models they use do not easily carry over from one stage to the next.
The demonstrator led by Coruscant uses the stroke care pathway as its first case study and brings together three complementary contributions. Coruscant’s teams are working on long-term patient monitoring. inSteps company is developing a digital twin designed to support preparations for thrombectomy, a procedure performed in certain cases of stroke. Demcon Johan Sports is focusing on personalised monitoring of recovery and rehabilitation.
The objective is to connect these approaches so that information generated at one stage of the care pathway can be reused and enriched at subsequent stages. This capability remains to be developed and assessed as part of the project.
Coruscant’s contribution: a shared architecture for digital twins
As part of the demonstrator, Coruscant is designing the shared data foundation and architecture that will connect the partners’ contributions. This work complements its contribution to long-term patient monitoring.
This architecture needs to integrate different types of information, maintain a long-term view of the patient and enable secure data exchange between tools. To facilitate the exchange of health data, the project plans to use FHIR, an international standard for structuring and exchanging such data.
The models focusing on long-term monitoring, thrombectomy and rehabilitation address different questions. By connecting them, the partners aim to determine whether information generated at one stage of the care pathway could also be useful at subsequent stages. The ability of these tools to exchange and use this information is therefore central to the demonstrator.
Following the stroke care pathway, this approach will also be explored in relation to ageing, diabetes and cardiometabolic diseases. The aim is to establish a foundation that can be adapted to different care pathways and use cases.
A challenge that extends beyond healthcare
Data describing the same system is often spread across different tools, teams or sites. Bringing it together is not enough: it is essential to analyse how the data interacts, track how it evolves and determine how it can inform decision-making.
An organisation can examine the impact of increased demand on its resources and its ability to deliver a service. An industrial company may want to anticipate equipment wear and compare maintenance strategies. An energy provider can test how a network might respond to an incident or a peak in consumption.
Each digital twin therefore starts with a specific question. The data, models and scenarios are selected according to the reality to be represented and the decision they are intended to inform. Their relevance must then be verified and maintained as that reality evolves.
The European demonstrator illustrates Coruscant’s approach: combining modelling, artificial intelligence and data integration to build a representation that remains useful beyond a single tool or point in time. In healthcare, the ambition is to provide better support throughout the patient journey. In other sectors, it is to help organisations understand their systems, anticipate how they may evolve and make better-informed decisions.
To discuss the applications of digital twins and the questions they raise, please contact the Coruscant team.