10161342CogiMotion: AI-powered Motor-Imagery BCI Integration for Upper-Limb Robotic Rehabilitation SystemsActiveFeasibility StudiesInnovate UKCogiMotion will develop an innovative AI-powered brain-controlled interface for upper-limb stroke rehabilitation by integrating Cogitat's advanced Brain-Computer Interface (BCI) technology with established robotic rehabilitation systems. This project addresses a critical healthcare challenge: stroke survivors with severe motor impairment currently receive only passive rehabilitation with robotic systems, as they cannot generate sufficient muscle activity to initiate movement themselves. Our technology transforms this experience by enabling active, patient-initiated rehabilitation through direct brain-to-device communication. Current rehabilitation robotics rely on detecting physical movement or muscle activity to provide therapy. For patients with no residual movement, these systems are ineffective. While BCI technology could bridge this gap by detecting movement intentions directly from brain signals, existing solutions are impractical for clinical use, requiring 30-50 minutes total setup-time. As an Imperial College London spinout, Cogitat has developed breakthrough AI technology that requires no calibration and can be set up in under 5 minutes using any EEG device with sensors over the motor cortex. Our technology has been successfully validated in preclinical settings using lightweight wearable dry-sensor headsets. Our AI models are trained on proprietary large-scale EEG datasets, enabling unprecedented generalisation across different individuals and EEG devices. This technology, which won first place in a global EEG competition, can accurately detect when a patient imagines moving their arm, even if they cannot physically move it. This 12-month feasibility study will create a cloud-based platform and API architecture that allows robotic rehabilitation systems to be controlled by patients' thoughts. The scalable platform will securely host our validated AI models while enabling integration with multiple rehabilitation devices. We will optimize our AI-driven brain signal processing specifically for rehabilitation applications and build a platform architecture that inherently supports personalisation features for tailored rehabilitation pathways. The integrated system will be demonstrated at a leading UK rehabilitation centre, where stroke rehabilitation specialists will provide professional feedback on performance and usability. The project includes preparation for medical device certification with support from regulatory consultants. We will develop the necessary documentation and quality systems required for UKCA marking, enabling rapid market entry following project completion. By enabling AI-powered thought-controlled rehabilitation, CogiMotion will expand access to robotic therapy for stroke survivors with severe impairment. The technology will integrate seamlessly with existing rehabilitation equipment, allowing clinics to serve patients who previously could not benefit from robotic rehabilitation. This aligns with NHS priorities for innovative stroke care and supports the UK's leadership in AI-driven medical technology development.