Uploaded June 2026 | Updated September 2026, 1 day ago
ECoG Neural Decoding for Hand Gesture Classification | Clinical BCI & Machine Learning | BR41N.IO Hackathon 2026
What can a team build in just 24 hours using intracranial Brain-Computer Interface data and advanced machine learning?
At the BR41N.IO BCI & Neurotechnology Hackathon, Team ECoGXCogs developed a high-performance neural decoding pipeline for hand gesture classification using electrocorticography (ECoG) recordings from an epilepsy patient. The project focused on analyzing intracranial brain signals associated with Rock, Paper, Scissors hand movements and improving classification accuracy through advanced feature extraction, machine learning, and deep learning approaches.
ECoG provides direct recordings of neural activity from the cortical surface, offering higher spatial resolution and signal quality than non-invasive methods. These signals are widely used in clinical neuroscience, functional brain mapping, neuroprosthetics, and next-generation Brain-Computer Interface research.
Building on previous work that achieved 98.9% classification accuracy, the team optimized preprocessing, feature extraction, and classification pipelines while exploring multiple machine learning and deep learning models. Their approach included spectral whitening, multi-time-window neural decoding, ensemble classification methods, and outlier detection to improve robustness and performance.
The project achieved classification accuracies of up to 100% for three-class hand gesture decoding and up to 97.74% for the more challenging four-class problem that included a resting state. The team identified important challenges involving overlapping motor cortex activity, resting-state classification, and epileptiform signal artifacts, demonstrating the complexities of real-world clinical Brain-Computer Interface development.
This project highlights the growing role of machine learning, neural decoding, intracranial recordings, and computational neuroscience in advancing Brain-Computer Interfaces. The work demonstrates how ECoG data can be used to decode motor intentions with remarkable accuracy, supporting future developments in neuroprosthetics, assistive technologies, and clinical neurotechnology.
Similar clinical neuroscience workflows are enabled by g.HIamp, CORTIQ, and the broader g.tec ecosystem for high-performance neural recording, functional brain mapping, and Brain-Computer Interface research.
More about BR41N.IO: https://www.gtec.at/hackathon/
ECoG Neural Decoding for Hand Gesture Classification | Clinical BCI & Machine Learning | BR41N.IO Hackathon 2026
What can a team build in just 24 hours using intracranial Brain-Computer Interface data and advanced machine learning?
At the BR41N.IO BCI & Neurotechnology Hackathon, Team ECoGXCogs developed a high-performance neural decoding pipeline for hand gesture classification using electrocorticography (ECoG) recordings from an epilepsy patient. The project focused on analyzing intracranial brain signals associated with Rock, Paper, Scissors hand movements and improving classification accuracy through advanced feature extraction, machine learning, and deep learning approaches.
ECoG provides direct recordings of neural activity from the cortical surface, offering higher spatial resolution and signal quality than non-invasive methods. These signals are widely used in clinical neuroscience, functional brain mapping, neuroprosthetics, and next-generation Brain-Computer Interface research.
Building on previous work that achieved 98.9% classification accuracy, the team optimized preprocessing, feature extraction, and classification pipelines while exploring multiple machine learning and deep learning models. Their approach included spectral whitening, multi-time-window neural decoding, ensemble classification methods, and outlier detection to improve robustness and performance.
The project achieved classification accuracies of up to 100% for three-class hand gesture decoding and up to 97.74% for the more challenging four-class problem that included a resting state. The team identified important challenges involving overlapping motor cortex activity, resting-state classification, and epileptiform signal artifacts, demonstrating the complexities of real-world clinical Brain-Computer Interface development.
This project highlights the growing role of machine learning, neural decoding, intracranial recordings, and computational neuroscience in advancing Brain-Computer Interfaces. The work demonstrates how ECoG data can be used to decode motor intentions with remarkable accuracy, supporting future developments in neuroprosthetics, assistive technologies, and clinical neurotechnology.
Similar clinical neuroscience workflows are enabled by g.HIamp, CORTIQ, and the broader g.tec ecosystem for high-performance neural recording, functional brain mapping, and Brain-Computer Interface research.
More about BR41N.IO: https://www.gtec.at/hackathon/
