What are we trying to achieve?

The construction of Rehapiano isolates the actions of individual muscles so that manifestations of the underlying problem can be well localised. The acquired data are time-structured, so that appropriate algorithms can detect phenomena related to various medical conditions. These can range from reduced muscle force or asymmetric muscle forces among the extremities, usually present after injuries, to unusual muscle force oscillations, usually present when the neural system is affected. The system is also equipped to quantify the severity of the detected anomalies.

The vision of the project lies in the need to improve the quality of life of patients suffering from muscular and neuromuscular disorders and to provide innovative and cost-effective support to improve the quality of healthcare for these patients in the field of intelligent diagnostics and support of more effective rehabilitation interventions. The basis of this multidisciplinary research is the use of sensors and data analysis for early and accurate diagnostics, the application of artificial intelligence in the adaptation of the rehabilitation system and process to individual needs, data mining to support self-learning of the therapeutic and diagnostic systems, the use of gamification as a way to facilitate motivation in rehabilitation, and objective economic and medical validation of therapy in real-time.

Expected results of the proposed project is a set of AI models supporting selected medical diagnosis, enabling incorporation into relevant models or decision-making procedures, a set of enhanced AI-models offering sufficiently accurate but at the same time easier to understand diagnosis at different levels, and a set of best practices for more effective cooperation between data analyst and medical experts. All software components designed and implemented during the project will be universally applicable to different types of devices through a standard communication protocol. We are dealing with unique time-based pressure data and curating a distinct dataset; our goal is to establish protocols and standards for capturing this specialised type of data. Additionally, we aim to create a state-of-the-art approach to preprocessing and processing this unique data, specifically tailored for applications in the medical domain. These innovations will contribute to advancing innovative clinical medical diagnostics systems.