README.md

---
license: cc-by-nd-4.0
language:
- en
library_name: pytorch
tags:
- eeg
- biosignal
- mamba
- state-space-model
- bidirectional
- foundation-model
- self-supervised
- masked-modeling
- efficient
- edge-ai
- neuroscience
datasets:
- TUEG
- TUAB
- TUAR
- TUSL
metrics:
- balanced_accuracy
- roc_auc
- pr_auc
thumbnail: https://raw.githubusercontent.com/pulp-bio/BioFoundation/refs/heads/main/docs/model/logo/FEMBA_logo.png
model-index:
  - name: FEMBA-Tiny
    results:
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Binary)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: binary
        metrics:
          - type: roc_auc
            value: 0.937
            name: AUROC
          - type: pr_auc
            value: 0.912
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multilabel)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multilabel
        metrics:
          - type: roc_auc
            value: 0.887
            name: AUROC
          - type: pr_auc
            value: 0.645
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multiclass-Multioutput)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multiclass-multioutput
        metrics:
          - type: roc_auc
            value: 0.893
            name: AUROC
          - type: pr_auc
            value: 0.504
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multiclass)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multiclass
        metrics:
          - type: roc_auc
            value: 0.918
            name: AUROC
          - type: pr_auc
            value: 0.518
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Slowing Classification
        dataset:
          type: TUSL
          name: TUH EEG Slowing Corpus (TUSL)
        metrics:
          - type: roc_auc
            value: 0.708
            name: AUROC
          - type: pr_auc
            value: 0.277
            name: AUC-PR
  - name: FEMBA-Base
    results:
      - task:
          type: time-series-classification
          name: EEG Abnormality Detection
        dataset:
          type: TUAB
          name: TUH EEG Abnormal Corpus (TUAB)
        metrics:
          - type: balanced_accuracy
            value: 81.05
            name: Balanced Accuracy (%)
          - type: roc_auc
            value: 0.8829
            name: AUROC
          - type: pr_auc
            value: 0.8894
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Binary)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: binary
        metrics:
          - type: roc_auc
            value: 0.949
            name: AUROC
          - type: pr_auc
            value: 0.932
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multilabel)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multilabel
        metrics:
          - type: roc_auc
            value: 0.909
            name: AUROC
          - type: pr_auc
            value: 0.634
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multiclass-Multioutput)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multiclass-multioutput
        metrics:
          - type: roc_auc
            value: 0.888
            name: AUROC
          - type: pr_auc
            value: 0.518
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multiclass)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multiclass
        metrics:
          - type: roc_auc
            value: 0.900
            name: AUROC
          - type: pr_auc
            value: 0.559
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Slowing Classification
        dataset:
          type: TUSL
          name: TUH EEG Slowing Corpus (TUSL)
        metrics:
          - type: roc_auc
            value: 0.731
            name: AUROC
          - type: pr_auc
            value: 0.289
            name: AUC-PR
  - name: FEMBA-Large
    results:
      - task:
          type: time-series-classification
          name: EEG Abnormality Detection
        dataset:
          type: TUAB
          name: TUH EEG Abnormal Corpus (TUAB)
        metrics:
          - type: balanced_accuracy
            value: 81.47
            name: Balanced Accuracy (%)
          - type: roc_auc
            value: 0.8856
            name: AUROC
          - type: pr_auc
            value: 0.8992
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Binary)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: binary
        metrics:
          - type: roc_auc
            value: 0.944
            name: AUROC
          - type: pr_auc
            value: 0.913
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multilabel)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multilabel
        metrics:
          - type: roc_auc
            value: 0.899
            name: AUROC
          - type: pr_auc
            value: 0.608
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multiclass-Multioutput)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multiclass-multioutput
        metrics:
          - type: roc_auc
            value: 0.878
            name: AUROC
          - type: pr_auc
            value: 0.516
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Artifact Detection (Multiclass)
        dataset:
          type: TUAR
          name: TUH EEG Artifact Corpus (TUAR)
          config: multiclass
        metrics:
          - type: roc_auc
            value: 0.915
            name: AUROC
          - type: pr_auc
            value: 0.521
            name: AUC-PR
      - task:
          type: time-series-classification
          name: EEG Slowing Classification
        dataset:
          type: TUSL
          name: TUH EEG Slowing Corpus (TUSL)
        metrics:
          - type: roc_auc
            value: 0.714
            name: AUROC
          - type: pr_auc
            value: 0.282
            name: AUC-PR
  - name: FEMBA-Huge
    results:
      - task:
          type: time-series-classification
          name: EEG Abnormality Detection
        dataset:
          type: TUAB
          name: TUH EEG Abnormal Corpus (TUAB)
        metrics:
          - type: balanced_accuracy
            value: 81.82
            name: Balanced Accuracy (%)
          - type: roc_auc
            value: 0.8921
            name: AUROC
          - type: pr_auc
            value: 0.9005
            name: AUC-PR
---

<div align="center">
  <img src="https://raw.githubusercontent.com/pulp-bio/BioFoundation/refs/heads/main/docs/model/logo/FEMBA_logo.png" alt="FEMBA Logo" width="800"/>
  <h1>FEMBA: Foundational Encoder Model with Bidirectional Mamba for EEG</h1>
</div>
<p align="center">
  <a href="https://github.com/pulp-bio/BioFoundation">
    <img src ="https://img.shields.io/github/stars/pulp-bio/BioFoundation?color=ccf" alt="Github">
  </a>
  <a href="https://creativecommons.org/licenses/by-nd/4.0/">
    <img src="https://img.shields.io/badge/License-CC_BY--ND_4.0-lightgrey.svg" alt="License">
  </a>
  <a href="https://arxiv.org/abs/2502.06438">
    <img src="https://img.shields.io/badge/arXiv-2502.06438-b31b1b.svg" alt="Paper">
  </a>
</p>

**FEMBA** is a powerful and efficient foundation model for **EEG signal analysis**, built upon a **bidirectional Mamba** state-space architecture. It supports **self-supervised pre-training** via masked reconstruction and **end-to-end supervised fine-tuning** for multiple downstream tasks (abnormal EEG, artifact detection, slowing classification). By using linear-time state-space modeling instead of quadratic attention, FEMBA scales to long EEG sequences and constrained hardware while remaining performant.

---

## 🔒 License & Usage Policy (Weights)

**Weights license:** The released model weights are licensed under **Creative Commons Attribution–NoDerivatives 4.0 (CC BY-ND 4.0)**. This section summarizes the practical implications for users. *This is not legal advice; please read the full license text.*

### ✅ You may
- **Use** and **redistribute** the **unmodified** FEMBA weights (including in commercial settings) **with proper attribution** to the FEMBA authors.
- **Fine-tune / adapt** the weights **for your internal use** (research or production) **without redistributing** the modified weights.
- **Publish your code, configs, logs, and papers** describing experiments with FEMBA (please cite the paper).

### 🚫 You may not
- **Share, host, or redistribute any modified weights** (including LoRA/adapter/delta checkpoints or pruned/quantized variants). Any parameter set that encodes an adaptation is considered a derivative and cannot be shared under CC BY-ND 4.0.
- **Imply endorsement** by the FEMBA authors for any derivative or evaluation without our written permission.
- **Use the FEMBA name** in a way that suggests your modified model is an official FEMBA release.

### 🤝 How to contribute improvements (PR-gated releases)
We welcome community improvements via a **pull-request (PR)** workflow. If you believe your improvements should become an **official FEMBA release**:
1. **Open a PR** in the [BioFoundation repository](https://github.com/pulp-bio/BioFoundation) describing the change (architecture/head/training recipe, datasets, preprocessing, compute).
2. Include **reproducibility artifacts**: configs, seeds, scripts, environment details, training/validation logs, and the **evaluation protocol** (e.g., TUAB/TUAR/TUSL) with exact splits.
3. Provide **comprehensive results** (AUROC/AUPR/BA, FLOPs, memory) vs. the baselines reported in the FEMBA paper.
4. After **maintainer review**, approved changes will be **retrained/validated** and, if accepted, **released by the maintainers** as a new **official FEMBA** checkpoint under **CC BY-ND 4.0**.

> Rationale: CC BY-ND protects users from fragmented, lower-quality “FEMBA variants,” while still enabling internal fine-tuning and a path for the community to upstream improvements through review.

---

## 🔎 Model Summary

- **Architecture:** Bidirectional Mamba encoder with a 2D-conv tokenizer (patching over channels × time), random masking (60%) for SSL, and either a lightweight linear head or a Mamba-enhanced head for downstream tasks. Hidden state size is fixed at 80 across variants.
- **Scaling:** Linear time & memory in sequence length (state-space model), enabling efficient long-context EEG modeling and on-device scenarios.
- **Pre-training data:** >21,000 hours of unlabeled clinical EEG from Temple University Hospital (TUEG). Subjects overlapping with TUAB/TUAR/TUSL are removed to prevent leakage.
- **Downstream tasks:** TUAB abnormal/normal (binary), TUAR artifact detection (BC/MC/MMC/MCC), TUSL slowing (4-class). TUAB uses its predefined split; TUAR/TUSL use 80/10/10 splits.
- **Optimization (typical):** Pre-training with Smooth L1 masked-patch reconstruction; fine-tuning with Adam (LR 1e-4), cosine decay, early stopping; layer-wise LR decay factor 0.75.

---

## 🚀 Model Variants

| Variant | Parameters | (num_blocks, embed_dim) |
| :--- | :--- | :--- |
| **FEMBA-tiny** | 7.8M | (2, 35) |
| **FEMBA-base** | 47.7M | (12, 35) |
| **FEMBA-large** | 77.8M | (4, 79) |
| **FEMBA-huge** | 386M | (20, 79) |

*Hidden state size is 80 for all variants; blocks correspond to Bi-Mamba layers in the encoder.*

---

## 🧠 Intended Use & Limitations

**Intended use.** Research on EEG representation learning and downstream classification (e.g., abnormal EEG detection, artifact detection, slowing classification). FEMBA is particularly useful when long sequences or limited compute/memory preclude quadratic-cost attention.

**Out-of-scope / limitations.**
- **Not a medical device.** Outputs are research signals and **must not** be used for clinical decision-making without appropriate validation and regulatory processes.
- **Domain shift.** Performance can degrade across cohorts (e.g., neonatal vs. adult EEG) and label protocols; domain adaptation is encouraged.
- **Class imbalance.** On some tasks (e.g., TUSL), AUROC may be strong while AUPR can trail attention baselines, highlighting sensitivity to class imbalance and protocol specifics.

---

## 🏗️ Architecture & Training Details

**Tokenizer & patches.** Raw EEG (C×T) is quartile-normalized per channel (IQR scaling) and tokenized with a 2D convolution over channel×time patches (e.g., 4×32) with learnable positional embeddings.

**Self-supervised objective.** Randomly mask **60%** of patches; reconstruct masked content with a lightweight decoder using **Smooth L1** loss (computed on masked patches only).

**Encoder.** Stacked **Bidirectional Mamba** blocks (forward + backward over a reversed copy), merged and residually connected; hidden size fixed to 80.

**Fine-tuning heads.**
- *Linear classifier:* small MLP (≈0.5M params).
- *Mamba-enhanced classifier:* adds one Mamba block before the linear layer (up to ≈0.7M params).

**Optimization notes.** Layer-wise LR decay (0.75); fine-tuning with Adam (initial LR 1e-4), cosine decay, early stopping; end-to-end updates (encoder + head).

---

## 📚 Training Data

- **Pre-training:** Temple University Hospital EEG (TUEG), ~21k hours, ~15k subjects; broad clinical coverage. Overlaps with TUAB/TUAR/TUSL removed to avoid leakage.
- **Downstream:**  
  - **TUAB** (abnormal vs normal; predefined split).  
  - **TUAR** (artifact detection, BC/MC/MMC/MCC protocols; randomized 80/10/10).  
  - **TUSL** (4-class slowing/seizure/complex/normal; randomized 80/10/10).

*See paper for dataset licenses and curation details; users are responsible for complying with source dataset terms.*

---

## 🔧 How to Use

FEMBA weights are organized by downstream task:

- **`TUAB/`** → base/large variants pre-trained on TUEG (excluding TUAB), for TUAB abnormal EEG.  
- **`TUAR/`** → tiny/base/large variants pre-trained on TUEG (excluding TUAR), for TUAR artifact detection.  
- **`TUSL/`** → variants pre-trained on TUEG (excluding TUSL), for TUSL slowing classification.

**Example:** fine-tune TUAR with the base checkpoint:

```
TUAR/base.safetensors
```

Open `run_train.py` from the [BioFoundation GitHub repository](https://github.com/pulp-bio/BioFoundation.git) and configure:

```python
# Set paths (example)
os.environ['DATA_PATH'] = "/path/to/dataset"
os.environ['CHECKPOINT_DIR'] = "/my_directory/TUAR/base.safetensors"
```

Then launch fine-tuning (Hydra):

```bash
python -u run_train.py +experiment=FEMBA_finetune
```

**Environment variables**  
- `DATA_PATH`: directory of the fine-tuning dataset.  
- `CHECKPOINT_DIR`: path to the chosen task-specific checkpoint.

---

## 📊 Results (Key Highlights)

**TUAB (Abnormal EEG Detection)**  
- **FEMBA-Huge:** **81.82%** balanced accuracy, **0.892** AUROC.

**TUAR (Artifact Detection)**  
- **Binary (BC):** **FEMBA-Base** AUROC **0.949**, AUPR **0.932**.

**TUSL (Slowing Classification, 4-class)**  
- **FEMBA-Base:** AUROC **0.731**.

> Full metrics, protocols, and comparisons—including MC/MMC on TUAR and multiple FEMBA sizes—are reported in the paper.

---

## ⚡ Efficiency

FEMBA provides strong accuracy with reduced compute/memory relative to Transformer baselines:

- **FEMBA-Huge (386M):** ~**58.7B FLOPs**, ~**30% less** memory than comparable Transformer baselines, with competitive TUAB accuracy.
- **FEMBA-Tiny (7.8M):** ~**1.31B FLOPs**—substantially fewer than large Transformer baselines—while achieving strong TUAR MCC performance.
- **FEMBA-Base (47.7M):** ~**7.52B FLOPs**, markedly lower than many attention-based baselines.

See the paper for details on measurement setup and baseline references.

---

## ✅ Responsible AI, Risks & Biases

- **Clinical safety:** This model is **not** a certified medical device and should **not** be used for diagnosis. Human oversight is required.  
- **Bias & drift:** Clinical EEG cohorts vary by device, montage, age, and pathology. Expect distribution shift and validate locally; consider domain adaptation (e.g., neonatal vs adult).  
- **Artifacts:** While artifact detection is strong, rare/complex artifacts may still be misclassified; use robust pre-processing and QC procedures.

---

## 🔗 Sources

- **Code:** https://github.com/pulp-bio/BioFoundation  
- **Paper:** FEMBA: Efficient and Scalable EEG Analysis with a Bidirectional Mamba Foundation Model (arXiv:2502.06438).

---

## 📜 Citation

If you use FEMBA in your research, please cite:

```bibtex
@misc{tegon2025fembaefficientscalableeeg,
      title={FEMBA: Efficient and Scalable EEG Analysis with a Bidirectional Mamba Foundation Model}, 
      author={Anna Tegon and Thorir Mar Ingolfsson and Xiaying Wang and Luca Benini and Yawei Li},
      year={2025},
      eprint={2502.06438},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2502.06438}
}
```

---

## 🛠️ Maintenance & Contact

- **Issues & support:** please open a GitHub issue in the BioFoundation repository.

---

## 🔗 Related Models

- **[LuMamba](https://huggingface.co/PulpBio/LuMamba)** — Direct successor. Combines FEMBA's bi-Mamba encoder with LUNA's channel-unification cross-attention for topology-invariant EEG modeling, plus LeJEPA pre-training.
- **[LUNA](https://huggingface.co/PulpBio/LUNA)** — Transformer-based topology-agnostic EEG foundation model (NeurIPS 2025). Companion architecture focused on channel-heterogeneity rather than sequence-length efficiency.
- **[TinyMyo](https://huggingface.co/PulpBio/TinyMyo)** — Tiny foundation model for flexible EMG signal processing at the edge.
---
## 🗒️ Changelog

- **v1.0:** Initial release of FEMBA model card with task-specific checkpoints and instructions.