---
pretty_name: SSL Ultrasound Representation (Stanford RF MultiFocal)
license: other
task_categories:
- feature-extraction
tags:
- ultrasound
- self-supervised
- rf-data
- mae
- jepa
size_categories:
- n<1K
---

# SSL Ultrasound Representation Dataset

Beamformed multi-focal ultrasound RF data from the Stanford Ultrasound RF Channel
dataset, aggregated for self-supervised pre-training (MAE / JEPA / contrastive).

## Files

| File | Description |
|------|-------------|
| `ultrasonic_dataset.zarr.zip` | 6D float32 zarr array, Blosc-LZ4 compressed (2.4 GB). |
| `ultrasonic_dataset.metadata.json` | Per-frame provenance (subject, file, subfolder, indices). |
| `README.md` | This card (auto-generated by `push_to_hf.py`). |

## Array layout

- **Name:** `rf_iq`
- **Shape:** `(14, 192, 192, 3, 2, 215)`
- **Axes:** `['frame', 'transducer', 'scanline', 'focal_zone', 'iq', 'sample']`
- **Chunks:** `(1, 32, 32, 1, 2, 215)` (one frame per chunk on the leading axis — aligned with DataLoader workers).
- **Dtype:** `float32`
- **IQ index:** `0` = in-phase, `1` = quadrature.

## Processing pipeline

1. Source: Verasonics Vantage `.mat` files with `MultiFocal` in the filename
   (each file holds a 2-frame cine via `Resource.RcvBuffer.numFrames = 2`).
2. Sub-aperture combination + IQ demodulation via `pymust.rf2iq`.
3. Fast-time axis truncated at **860 samples**
   (everything beyond is noise/zeros), then decimated by
   **4×** → **215** output samples per signal.
4. Stacked into a single 6D array with chunks optimised for 3D MAE patching.

See `scripts/dataset_scripts/build_mae_dataset.py` in the source repo for the
exact pipeline; see `scripts/dataset_scripts/visualize_beamformed_pymust.py` for
the reference beamformer.

## Frame counts

By subject:

| Subject | Frames |
|---------|--------|
| `rat1` | 8 |
| `rat2` | 6 |

By subject × subfolder:

| Subject | Subfolder | Frames |
|---------|-----------|--------|
| `rat1` | `—` | 6 |
| `rat1` | `exposed_liver` | 2 |
| `rat2` | `—` | 6 |

## Loading

```python
from src.dataset import build_split_datasets

splits, meta = build_split_datasets(
    "benbarkow/test-us-ssl",
    split_strategy="file",  # or "subject"
    cache_dir="/tmp/hf_cache",
)
train_ds = splits["train"]
```

The split helpers (`split_by_subject`, `split_by_file`) avoid cross-split leakage:
both frames of the same 2-frame cine always end up in the same split, and (for
`split_by_subject`) no subject appears in two splits.