MindtheNoise

Artifact detection and clean-data retention in Muse 2 EEG

Research Lead Zach GoldwynTerm 2026–2027Research Team Benjamin DeMayo, Samuel Fakhouri, Rushil Ghosh, Namit Jain, Shrinithi Shaurya

.00001% of the human brain has been mapped in full detail, according to Harvard and Google scientists Viren Jain and Jeff Lichtman. For reference, the ocean is considered one of the most mysterious and unexplored domains in the world, and has 27% mapped. With this lack of information, we can’t understand some of the most fundamental human operations like consciousness, memory, attention, decision making, emotion, and learning.

One of the key solutions to this problem is collecting massive amounts of electroencephalography (EEG) data, which record the brain’s electrical activity through electrodes on the scalp. However, collecting clean EEG readings is incredibly difficult, often only having < 30% clean data per reading, which is why so little of the brain is understood. Mind the Noise sets out to develop novel methods to collect and clean EEG data with the Muse 2 headset.

The methods developed in this research has the potential to:

  • Bridge AI and neuroscience sectors like cognitive neuroscience, clinical neurophysiology, and more, by finally providing models with clean, quality data that can be used in training sets.
  • Accelerate research across neuroscience, medicine, and psychology by uniquely leveraging one of the most affordable EEG reading technologies, the Muse 2 headset.
  • Inspire one of the first at-home solutions to diagnosing neurological issues which can currently only be addressed through hospital grade hardware and methods.

Literature Review

Works Cited

  • Arpaia, Pasquale, et al. “A Systematic Review of Techniques for Artifact Detection and Artifact Category Identification in Electroencephalography from Wearable Devices.” Sensors, vol. 25, no. 18, 2025, article 5770. doi.org/10.3390/s25185770.
  • Delorme, Arnaud. “Automated Data Cleaning for the Muse EEG.” 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), IEEE, 2021. doi.org/10.1109/BIBM52615.2021.9669415.
  • Nippon Foundation–GEBCO Seabed 2030 Project. “Seabed 2030 Announces Millions of Square Kilometres of New Seafloor Data on World Hydrography Day.” Seabed 2030, 21 June 2025. seabed2030.org.
  • Shapson-Coe, Alexander, et al. “A Petavoxel Fragment of Human Cerebral Cortex Reconstructed at Nanoscale Resolution.” Science, vol. 384, no. 6696, 2024. doi.org/10.1126/science.adk4858.
  • Witteveen, Jolanda, et al. “Comparison of a Pragmatic and Regression Approach for Wearable EEG Signal Quality Assessment.” IEEE Journal of Biomedical and Health Informatics, vol. 24, no. 3, 2020. PDF.