Use Cases & Demos

The use case focuses on enabling a Remote Concurrent Preoperative Planning Framework (RCPPF) that allows surgeons and engineers to collaboratively perform advanced surgical planning at a distance. It combines analysis of 2D medical images, creation of 3D anatomical models, and further processing such as implant or resection guide design. The framework supports real-time collaboration through shared interfaces (screen, keyboard, mouse, webcams) and extends this into a multi-user, immersive environment using VR tools for 3D model interaction. It is designed to be open and interoperable with a wide range of certified medical software and to support specialised surgical domains. The system also enables participation from external experts, trainees, and advisors, and is applicable in both planned interventions and emergency or disaster medicine scenarios where remote access to imaging data from ambulances or point-of-care devices is required.

This use case is needed because current preoperative planning practices are constrained by the need for physical co-location of specialised personnel and by limited videoconferencing tools that do not adequately support complex medical data interaction. As a result, surgeons often perform interventions without full 3D preparation or patient-specific planning support, increasing workload and risk. The lack of high-performance remote collaboration tools also restricts access to expertise in peripheral or underserved hospitals and creates inefficiencies due to travel requirements for engineering support. In emergency and disaster contexts, these limitations become even more critical, as timely, lossless, and high-quality data exchange is required for rapid decision-making. The proposed framework addresses these gaps by enabling lossless, real-time, multi-user collaboration with strong requirements on latency, reliability, and bandwidth, ultimately improving surgical precision, reducing risk and costs, and increasing equitable access to advanced medical planning services.

The use case focuses on enabling a remote rehabilitation assessment framework that allows patients to perform therapeutic exercises at home while being monitored in real time by healthcare professionals. It removes the need for wearable sensors by relying on camera-based motion tracking, where AI models analyse video streams to assess movement quality and provide immediate feedback. When deviations are detected, the system can either suggest corrective actions directly to the patient or escalate to a clinician for remote video consultation. The solution integrates AI, extended reality (XR), and augmented reality (AR) to create an immersive rehabilitation experience, including applications such as virtual prosthesis mirroring for phantom limb pain therapy. Clinicians can remotely monitor progress, adjust treatment plans, and intervene when necessary, while the system continuously collects and labels data to improve model performance over time.

This use case is needed because traditional rehabilitation services require frequent in-person visits, which create significant physical, economic, and logistical burdens for patients, particularly those with limited mobility or living in rural or underserved regions. At the same time, healthcare systems face increasing demand for rehabilitation services, with limited availability of specialised staff, equipment, and appointment capacity, leading to long waiting lists and reduced access to care. Current solutions are not scalable enough to provide continuous, personalised monitoring outside clinical environments, and often rely on cumbersome sensor setups that reduce usability and adoption. This use case addresses these limitations by enabling scalable, home-based rehabilitation supported by AI-driven assessment and immersive feedback technologies, improving accessibility, reducing healthcare system pressure, and allowing clinicians to focus on more critical cases while maintaining quality of care.

Qualtek’s UC focuses on an e-Health/Telemedicine platform that features two key points.

The first point is about controlled utilisation of federated medical data on the grounds of trusted code to data patterns. Medical data may include patient records gathered and managed by institutions such as Hospitals, diagnostic centres or even doctors’ offices. The data concept expands also to AI-developed diagnostic models due to the protection requirement against inference attacks. In principle the UC aims at demonstrating adequate sovereignty over personal or owned information and the value potential of
it.
The second point is the provision of a trusted environment for data access and processing, that ensures protection (including confidentiality) in the management of medical information (which is considered sensitive personal data, as per GDPR [R2016/679]). This trust shall be provided by technical enforcement of the consent options of the individual, as stipulated by policies regarding provisions and prohibitions. The concept of the trusted environment expands to cater for protection requirements
relating to data products (e.g. AI models) and credible attribution of data provenance and ownership to the providers of data and developers of AI models.
The UC focuses on exploiting data without the need for centralised architectures and data
movement/duplication. This is achieved by introducing a data plane dimension that is integrated with and across the network (data-aware slices) and makes possible the utilisation of assets in various workflows and scenarios. The data plane dimension supports advertisement, discovery, usage negotiation and orchestration of trusted artifacts for the envisaged usage scenarios.

The UC challenges this pattern by introducing requirements aiming at automating enforcement of policies as well as guaranteeing that the overall data management is limited to what is foreseen by the will of the individual. Additionally, the UC invests significantly on the preservation of data value by ensuring that any exploitation of the primary or secondary products can trace back to the contributors, thus ensuring a proper share (claim of such share is not the subject of this UC, however).
The UC aims for data to remain where they are created, with the 6G network adopting of the trusted code- to data paradigm. It also aims to turn the use of AI-generated diagnostic models into a collaborative task with all participating nodes retaining fundamental rights, such as in/out opting and value share. Processing and training workflows not only shall address connectivity but also require active and trusted participation of the network in the processing plane. A new secure network slice pattern is to be created, which is enabling the individual to exploit/combine network connectivity capabilities with built-in privacy protection (e2e slice between owner’s assets, e.g. phone or home camera). The envisaged components and patterns may allow for future intelligence orchestration by the network, instead of the participating business/application/service