Software & Devices

Products for the mind, by the mind. Our devices and software divisions aim to help members develop the skills to design products for the future.

Purpose


Neurotechnology’s potential to improve lives is unmatched. Neurotech@Berkeley is committed to building innovative, impactful software and hardware solutions to big problems.

Our Devices Division is developing custom, non-invasive neural interfaces to experiment with new ways of interfacing with computers, augment human capabilities, and innovate new forms of prosthetics. Some of our devices, such as our EMG-based wristband (NeuroWrist) and our fully-customizable EEG headset (NeuroMorph), are used by the software division to collect data!
Our Software Division uses different forms of neural data, such as those from EEG, MRI, and fNIRS, to build software-based solutions to a neuroscience problems. We currently have a wide variety of projects, ranging from eye-tracking typing methods to brain-state-controlled music selection algorithms.




Current Projects

RobLES

Build and control an external limb using EMG (building off of Neurowrist work)

Wetware Computing

Bridging the gap between silicon and biology - the first collegiate initiative to compute and understand live neural circuits!

Neuromorph

Design and build a flexible EEG that is affordable and enables user-defined electrode position and count

fMRI Image Reconstruction Model

Reconstructing images that people see from fMRI data, i.e. “imagination reading” with a stable diffusion based machine learning architecture

MRI Cancer Detection Model

A computer vision project utilizing the U Net architecture to identify tumor cells in an MRI scan of a person’s brain.

EMG Gesture Prediction

A novel CNN and LSTM based machine learning project predicting hand gestures from EMG muscle movements




Recent Archives

Neurowrist

Utilize real-time processing EMG signals from the wrist and translate captured features into controls

Gazeware

Answering visual questions using image and word embeddings and recurrent attention algorithms

EMG Roboarm

Built machine learning classifier for EMG signals for operating a 3D printed prosthetic arm

EEG Mind-Control Car

Used EEG tracking with a Muse headset for hands-free vehicle control through blinks and winks

EEG Art

Used EEG signals collected through a Muse headset to generate real-time visual art based on the user's brainstate

Canine fNIRS BCI

Built an fNIRS headset for dogs for tracking oxygenation of their olfactory bulb to detect disease states of their owners

Tunable PCB for EMG/EEG

An EOG headset and application for monitoring productivity during work sessions. Analyzes EOG signals to measure focus, fatigue, and alertness and displays metrics in a dashboard.

EEG Music

EEG Music project on mood classification and playback control