Emory BMI GSoC Project Ideas

Supporting code for the GX Dataset

Openwater's Open-Source Neuromodulation Software

Neuromodulatory Control Networks (NCNs), a novel LLM architectural modification inspired by the neuromodulatory systems in the vertebrate brain.

Openwater's Open-Source Neuromodulation Hardware

focused ultrasound toolbox

Multidigraph learning (MDGL) for training recurrent spiking neural networks

A 3D Slicer extension for Openwater’s OpenLIFU (Low Intensity Focused Ultrasound) research platform, providing an advanced interface to features found in the OpenLIFU app.

Python implementation of the NICE & SONIC models for multiple neuron types, along with command-line scripts for rapid simulations.

Python port of the PARRM algorithm for removing periodic artefacts from signals.

Image-to-image deep learning framework for MRI to porosity map translation

Graphical user interface for planning and controlling Open-LIFU sonications

ANML: A Neuromodulated Meta-Learning algorithm (using the `higher` framework)

Python+NEURON framework enabling the creation of morphologically-structured neuron models incorporating the SONIC paradigm.

A repository implementing a biologically inspired spiking neural network for psychological profiling. This project uses my own "QLIF-Neurons" by incorporating feedback and cross‐connections among key brain regions (e.g., prefrontal cortex, amygdala, hippocampus, thalamus, and striatum) and integrates text‐based emotion analysis.

Openwater Phantoms

open-LIFU TX Module Firmware for STM32L443

🧠 A biologically-accurate spiking neural network. Pure C99. No frameworks. No matrices. Neurons fire, learn, dream, and evolve. — Proprietary: Non-commercial use only.

Code used to generate the results in "A unified model library maps how neuromodulation reshapes the excitability landscape of neurons across the brain"

BBOP is a modular pipeline tailored for tFUS planning with BabelBrain. While optimized for use with BabelBrain integrated in Brainsight, it can also be used with the standalone installation. It integrates MRI preprocessing, MNI-based targeting, transducer positioning, pseudo-CT generation, and simulation preparation in a structured workflow.