🧠 Homeostatic Dynamic Mean Field Model (DMF_ISP)

This repository contains Python code for simulating a homeostatic dynamic mean field (DMF) model with synaptic plasticity and generating synthetic BOLD signals. The model implements biologically inspired excitatory-inhibitory dynamics using a structural connectivity (SC) matrix derived from human brain data. It supports noise, homeostatic inhibitory plasticity, and simulation of functional connectivity (FC).

---

📂 Repository Contents

DMF_ISP/
├── BOLDModel.py              # Module for simulating BOLD signals from firing rates
├── DMF_ISP_numba.py          # Core DMF model with inhibitory synaptic plasticity (Numba-accelerated)
├── run_DMF_ISP.py            # Example script to run DMF simulation and generate FC matrix
├── SCmatrices88healthy.mat   # Structural connectivity matrix averaged across 88 healthy participants

---

🧬 Scientific References

Structural Connectivity Source

Škoch, A., Rehák Bučková, B., Mareš, J., et al. (2022).
Human brain structural connectivity matrices–ready for modelling.
Scientific Data, 9, 486. https://doi.org/10.1038/s41597-022-01596-9

DMF Model (Preprint)

Mindlin, I., Coronel-Oliveros, C., Sitt, J. D., Cofré, R., Luppi, A., Andrillon, T., & Herzog, R. (in preparation).
A homeostatic dynamic mean field model: enhanced stability and state repertoire.

---

⚙️ Installation

1. Clone the repository:

git clone https://github.com/your-username/DMF_ISP.git
cd DMF_ISP

2. Install the required Python dependencies:

pip install numpy scipy matplotlib numba

---

🚀 Usage

To run a basic simulation:

python run_DMF_ISP.py

This will:

• Load the structural connectivity matrix
• Run the DMF model with or without plasticity
• Generate synthetic BOLD signals
• Compute and display a functional connectivity matrix

---

🧠 Model Description

• Nodes: 90 brain regions
• Inputs: Structural connectivity (SC), noise
• Outputs: Firing rates, BOLD signals, FC matrix
• Plasticity: Homeostatic inhibitory plasticity targeting a fixed mean firing rate
• Numerical Integration: Euler-Maruyama with Gaussian noise
• Optimization: Numba-accelerated for performance

---

📊 Outputs

• BOLD signals: simulated low-frequency hemodynamic response
• FC matrix: pairwise Pearson correlations between regions
• Plots: connectivity matrix visualization

---

📦 Dependencies

• numpy
• scipy
• matplotlib
• numba

---