by Christian Thomay and Benedikt Gollan (Research Studios Austria FG) and Erich Heil (tech2People)

The physical therapy centre of the future contains a variety of robotic assistance systems, allowing an increasing range of patients access to individualised rehabilitation therapy. However, novel, holistic approaches are needed to combine heterogeneous data from such different devices into a consistent digital twin that encompasses a patient’s circumstances and needs. tech2people and the Research Studios Austria FG aim to realise such a data-driven, digital therapy centre where data is combined and analysed using Big Data and AI methods, allowing for precise insight into, and forecast of, therapy progress for both therapists and patients.

"Get up and walk every day. Something that most of us take for granted, but for people who suffer from neurological diseases and are in a wheelchair, it is probably what they want most.” Gregor Demblin (Co-Founder of tech2people) is paralysed from the waist down; he has been in a wheelchair since a swimming accident in 1995. Walking and training with robotic assistive technology systems has given him a new degree of freedom and changed his life: he hardly needs to take any medication; bladder infections and abrasions on the body have become fewer; and after "sitting for 20 years" he now feels "like a new person". Studies have shown [1] that his personal experience is measurable. Motivated by these results he founded tech2people, with the goal to set up the most modern physical therapy centre in Central Europe and specialise in data-driven robot-assisted therapies for everyone.

In past studies, exoskeletons and other robotic assistive systems have been shown to contribute to recovery of body functionality, increasing recovery speed and therapy effectiveness [2]. An additional advantage of robotic therapy is the wealth of data that the assistive systems generate, for instance, an Eksobionics EksoNR robotic exoskeleton [L1] records data from hip and knee angle sensors, pressure sensors on toes and heels, and assistive motor force at a rate of up to 500 Hz. Similarly, other advanced assistive systems such as the Hocoma Lokomat [L2] – a robot-assisted treadmill system – and a range of devices from Tyromotion [L3] – which target different regions of the body such as upper extremities, finger/hand, and arm/shoulder – allow for detailed insight into the therapy process and the state of the patient.

However, much of this bounty of data is either not fully utilised yet – or not utilised at all. While data interfaces exist and feedback may be given to patients, there is no holistic methodology that allows for cross-examining the data between different devices and therapy types, thereby enabling a detailed analysis of this data towards studying a patient’s progress on their journey to rehabilitation. The need for such multimodal fusion has been concluded in studies into the effectiveness of exoskeleton therapy [3], as well as the need for therapy individualisation, but such systems have not been realised yet.

tech2people, in a scientific cooperation with the Research Studios Austria FG (RSA FG), intends to change that. Figure 1 illustrates the vision of the therapy centre of the future: the aim is to realise a fully digital therapy environment where different assistive systems and sensor technologies are all running on a shared platform, allowing for a holistic view of the patients and their therapies. This heterogenous data is processed using multi-sensor fusion, culminating in a digital twin model of the patient. This digital twin represents the state of relevant parts of the patient’s body and is constructed from data obtained from different sensor devices and assistive systems.

Figure 1: Data-driven therapy is supported by different devices, combining the data to allow for individualised approaches.
Figure 1: Data-driven therapy is supported by different devices, combining the data to allow for individualised approaches.

Such a digital twin serves two main purposes. It offers the therapists detailed insight into the patient’s progress and individual requirements, allowing the therapists to create customised therapy schemes tailor-made for individual patients, and to adjust therapy schemes over time. The digital twin also allows for an intuitive visualisation of their progress for the patients themselves: summarising complex information in accessible 3D models and colour schemes, patients can see how they did in a given therapy session, making the process more transparent and engaging.

However, visualising and evaluating the data on a session-to-session basis is only the first step. By deriving generalised metrics that offer insight into relevant parts of the rehabilitation process, therapy progress can be evaluated over time, showing trends and long-term developments that individual session data might not reveal. Furthermore, the wealth of information created throughout all therapy sessions comes together in an incrementally growing knowledge base. Together, these aspects form the foundation for predictive modelling: using Big Data and AI-based methods, therapy forecasts can be given, contributing to an optimal choice of therapy for each individual patient.

Due to the sensitive nature of confidential patient data, data security and privacy are of utmost importance. To that end, data encryption and anonymisation schemes will be applied, ensuring no outside access to patient data or identification of individuals. Furthermore, it is a key priority that the patient themselves always remain the focus; all results derived from the data will form the basis of recommendations to the therapists, but decisions will only be made in a dialogue between patient and therapist, ensuring a respectful usage of their data.

tech2people is currently working on realising its vision for the therapy centre of the future, located in Vienna, Austria. The ongoing research endeavour in partnership with RSA FG aims to establish the methodology to evaluate data from a range of assistive systems, creating a digital twin of the patient that allows both therapists and patients detailed insight into the rehabilitation process. These analyses of heterogeneous data are founded on the therapy expertise of tech2people, together with the data science and sensor fusion know-how of RSA FG, which in combination will contribute to making robotic physical therapy more effective, transparent, and user-friendly. The therapy centre aims to offer access to advanced therapy technology that was previously not available, increasing the range and accessibility of rehabilitation for patients that might otherwise not have access to it, and offering better, and more individual, therapy for anyone who requires it.


[1] A. D. Karelis, et al.: “Journal of rehabilitation medicine”, 49.1 (2017): 84-87.
[2] A. A. Frolov, et al.: “Use of robotic devices in post-stroke rehabilitation”, Neuroscience and behavioral physiology 48.9 (2018): 1053-1066.
[3] D. Shi, et al.: “A review on lower limb rehabilitation exoskeleton robots”, Chinese Journal of Mechanical Engineering 32.1 (2019): 1-11.

Please contact:
Christian Thomay
Research Studios Austria FG, Austria
+43 (1) 5850537-314
This email address is being protected from spambots. You need JavaScript enabled to view it.

Next issue: July 2023
Special theme:
"Eplainable AI"
Call for the next issue
Image ERCIM News 130
This issue in pdf


Image ERCIM News 130 epub
This issue in ePub format

Get the latest issue to your desktop
RSS Feed
Cookies user preferences
We use cookies to ensure you to get the best experience on our website. If you decline the use of cookies, this website may not function as expected.
Accept all
Decline all
Read more
Tools used to analyze the data to measure the effectiveness of a website and to understand how it works.
Google Analytics
Set of techniques which have for object the commercial strategy and in particular the market study.
DoubleClick/Google Marketing