01/01/2026
🧠⌨️ An intuitive, bimanual, high-throughput QWERTY touch typing neuroprosthesis for people with tetraplegia
What if people with paralysis could type as naturally and quickly as anyone else — just by attempting finger movements? This project brings that vision to life.
🧠 Highlights
• Provides a bimanual QWERTY keyboard interface controlled by attempted finger movements
• Uses a recurrent neural network (RNN) to decode neural signals into characters in real-time
• High throughput: up to 110 characters per minute, 22 words per minute, with a 1.6% word error rate
• Requires only ~30 calibration sentences to achieve accurate decoding
• Works for participants with ALS or high cervical spinal cord injury, including those unable to use eye-gaze or speech-based systems
• Typing is self-paced, intuitive, and familiar, leveraging memory of standard keyboard layouts
💡 Why it matters
This iBCI typing neuroprosthesis provides fast, free-form digital communication for individuals with tetraplegia. By mimicking natural finger movements, it offers a clinically viable, easy-to-learn, and high-accuracy solution, surpassing current state-of-the-art hand motor iBCI systems and offering independence and privacy in communication.
🔥 Inspired? Submissions for BCI Award 2026 are OPEN!
📅 Deadline: September 1st, 2026
💰 Total Prize: $6,000 USD
📖 Top 12 projects will be invited to publish in the BCI State-of-the-Art series by Springer
🔗 Details: www.bci-award.com/Home