Add mosaic-intelligence README + pipeline(s) from accepted_submissions

mosaic-intelligence/README.md

@@ -0,0 +1,237 @@
+---
+name: "OpenH-RF — Mosaic Intelligence / NuevoSono IVUS"
+license: cc-by-4.0
+task_categories:
+  - image-segmentation
+tags:
+  - ultrasound
+  - rf
+  - openh-rf
+  - IVUS
+  - tracked-ultrasound
+---
+
+# Data Card — Mosaic Intelligence / NuevoSono in-vivo IVUS
+
+This data card consists of a collection of in-vivo intravascular ultrasound (IVUS) acquisitions from a porcine study. Each acquisition is saved in the *zea* file format.  The general dataset sections below (contributors, license, characterization, etc.) apply to all acquisitions, with individual subsections detailing each dataset's specific dimensions and which optional groups are present.
+
+There are 9 acquisitions in total:
+
+**Group 1: `15_*` acquisitions (untracked, counterclockwise rotation):**
+- `15_10_18_21`
+- `15_10_50_19`
+- `15_16_45_06`
+
+**Group 2: `22_*` acquisitions (tracked, linear encoder pullback, clockwise rotation):**
+- `22_12_10_52`
+- `22_12_29_46`
+- `22_12_38_45`
+- `22_13_10_16`
+- `22_13_56_43`
+- `22_14_29_54`
+
+## Dataset Description
+
+Each dataset is an IVUS acquisition collected in a porcine animal study. The source data consists of raw IVUS RF frames and a pre-computed (scan-converted) B-mode image. Six of the acquisitions have time-sampled linear encoder pullback positions of the IVUS probe at each frame. For each acquisition and for every frame, per-class segmentation masks are provided for the vessel lumen, intima-media, and guidewire.
+
+## Dataset Contributors
+
+Mosaic Intelligence Labs in collaboration with NuevoSono
+
+Primary points of contact:
+Brian Boitnott (1): brian@mosaicintelligence.xyz
+Ali Mackanic (1): ali@mosaicintelligence.xyz
+
+## Dataset Creation Date
+
+07/17/2026
+
+## License / Terms of Use
+
+All contributed data, labels, and metadata is released by Mosaic Intelligence Labs
+and NuevoSono under CC BY 4.0.
+Pre-existing hardware, software, simulation, and platform intellectual property
+remain the property of their respective owners. The team agrees to comply with
+OpenH-RF governance, publication, and data-sharing policies. The license is also
+recorded in each file's `metadata/credit` field.
+
+## Intended Usage
+
+Intended for IVUS tracking and segmentation applications, including lesion
+detection and image-guided intervention.
+
+## Dataset Characterization
+
+- **Data Collection Method:** Porcine (in-vivo animal study)
+- **Labeling Method:** derived tracking metadata, semi-automated labeling
+- **Acquisition System:** Single-element IVUS, center frequency 30 MHz,
+  sampling rate 1 GHz
+
+## Dataset Format
+
+All acquisitions are submitted in the *zea* file format. The RF data is stored as
+a rotational sequence of A-lines (`n_tx` transmits per frame, one element/channel);
+the accompanying B-mode `image` and `segmentation` masks are pre-computed,
+scan-converted Cartesian frames sharing a per-pixel coordinate grid.
+
+### Shared per-sample schema
+
+Fields that are in every acquisition:
+
+| Group / field | Shape | Dtype | Units | Description |
+|---|---|---|---|---|
+| `data/raw_data` | `[n_frames, n_tx, n_ax, 1, 1]` | float32 | source RF units | Raw IVUS RF frames (single element / channel) |
+| `data/image/values` | `[n_frames, H, W]` | uint8 | — | Pre-computed (scan-converted) B-mode image |
+| `data/image/coordinates` | `[H, W, 3]` | float32 | m | Per-pixel Cartesian positions `(x, y, z)`; cross-section lies in the x-y plane, `z = 0` |
+| `data/segmentation/values` | `[n_frames, H, W, n_labels]` | bool | — | Per-class boolean segmentation masks (one channel per label) |
+| `data/segmentation/coordinates` | `[H, W, 3]` | float32 | m | Same per-pixel grid as `image/coordinates` |
+| `data/segmentation/labels` | `[n_labels]` | str | — | Channel names: `background`, `lumen`, `intima_media`, `guidewire` |
+| `scan/*` | — | mixed | Hz / s / rad / m | Scan metadata (`sampling_frequency`, `center_frequency`, `demodulation_frequency`, `polar_angles`, `t0_delays`, `tx_apodizations`, `focus_distances`, `transmit_origins`, `initial_times`) |
+| `probe/name` | scalar | str | — | Probe name (`NuevoSono IVUS`) |
+| `probe/type` | scalar | str | — | Probe geometry type (`custom`) |
+| `probe/probe_geometry` | `[1, 3]` | float32 | m | Element position (single element at the origin) |
+| `probe/probe_center_frequency` | scalar | float32 | Hz | Nominal center frequency (30 MHz) |
+| `metadata/subject/{id, type}` | scalar | str | — | Subject identifier (`porcine_<acq>`) and type (`animal`) |
+| `metadata/credit` | scalar | str | — | Attribution: `Mosaic Intelligence - NuevoSono IVUS porcine dataset (CC BY 4.0)` |
+| `metadata/text_report` | scalar | str | — | transducer rotation direction |
+| `metadata/rotation/{samples, sampling_frequency, start_time_offset}` | `[1]` / scalar / scalar | float32 | — / Hz / s | Transducer rotation: `samples = +1` clockwise, `-1` counterclockwise |
+| `metadata/pullback_position/{samples, sampling_frequency, start_time_offset}` | `[n_frames]` / scalar / scalar | float32 | m / Hz / s | Linear encoder pullback position per frame (**tracked acquisitions only**) |
+
+## Per-dataset details
+
+One data card per contributed sub-dataset. All fields follow the shared schema above. `n_ax = 8192`, `n_el = 1`, and `n_labels = 4` (`background`, `lumen`,
+`intima_media`, `guidewire`) for every acquisition.
+
+### `15_10_18_21` (untracked, counterclockwise)
+
+- Dimensions: `n_frames = 100`, `n_tx = 540`, `H = 985`, `W = 986`
+- Frame rate: ~3.33 Hz
+- Rotation: `metadata/rotation/samples = -1` (counterclockwise); `metadata/text_report = "Transducer rotation direction: counterclockwise."`
+- Tracking: **none** — no `metadata/pullback_position` group (source has no `frameAttributes.txt`)
+- Total size on disk: ~1.45 GB
+- Note: the image/mask grid is non-square (`985 x 986`)
+
+![15_10_18_21 reconstruction overview](../outputs/15_10_18_21/overview_5_frames.png)
+
+### `15_10_50_19` (untracked, counterclockwise)
+
+- Dimensions: `n_frames = 60`, `n_tx = 360`, `H = 985`, `W = 986`
+- Frame rate: ~3.33 Hz
+- Rotation: `metadata/rotation/samples = -1` (counterclockwise)
+- Tracking: **none** — no `metadata/pullback_position` group
+- Total size on disk: ~0.59 GB
+- Note: the image/mask grid is non-square (`985 x 986`)
+
+![15_10_50_19 reconstruction overview](../outputs/15_10_50_19/overview_5_frames.png)
+
+### `15_16_45_06` (untracked, counterclockwise)
+
+- Dimensions: `n_frames = 100`, `n_tx = 540`, `H = 985`, `W = 986`
+- Frame rate: ~3.33 Hz
+- Rotation: `metadata/rotation/samples = -1` (counterclockwise)
+- Tracking: **none** — no `metadata/pullback_position` group
+- Total size on disk: ~1.45 GB
+- Note: the image/mask grid is non-square (`985 x 986`)
+
+![15_16_45_06 reconstruction overview](../outputs/15_16_45_06/overview_5_frames.png)
+
+### `22_12_10_52` (tracked, clockwise)
+
+- Dimensions: `n_frames = 150`, `n_tx = 540`, `H = W = 2048`
+- Frame rate: ~3.33 Hz
+- Rotation: `metadata/rotation/samples = +1` (clockwise)
+- Tracking: includes `metadata/pullback_position` `[150]` (linear encoder position per frame)
+- Total size on disk: ~2.41 GB
+
+![22_12_10_52 reconstruction overview](../outputs/22_12_10_52/overview_5_frames.png)
+
+### `22_12_29_46` (tracked, clockwise)
+
+- Dimensions: `n_frames = 100`, `n_tx = 540`, `H = W = 2048`
+- Frame rate: ~8.33 Hz
+- Rotation: `metadata/rotation/samples = +1` (clockwise)
+- Tracking: includes `metadata/pullback_position` `[100]`
+- Total size on disk: ~1.66 GB
+
+![22_12_29_46 reconstruction overview](../outputs/22_12_29_46/overview_5_frames.png)
+
+### `22_12_38_45` (tracked, clockwise)
+
+- Dimensions: `n_frames = 150`, `n_tx = 360`, `H = W = 2048`
+- Frame rate: ~8.33 Hz
+- Rotation: `metadata/rotation/samples = +1` (clockwise)
+- Tracking: includes `metadata/pullback_position` `[150]`
+- Total size on disk: ~1.65 GB
+
+![22_12_38_45 reconstruction overview](../outputs/22_12_38_45/overview_5_frames.png)
+
+### `22_13_10_16` (tracked, clockwise)
+
+- Dimensions: `n_frames = 100`, `n_tx = 540`, `H = W = 2048`
+- Frame rate: ~8.33 Hz
+- Rotation: `metadata/rotation/samples = +1` (clockwise)
+- Tracking: includes `metadata/pullback_position` `[100]`
+- Total size on disk: ~1.54 GB
+
+![22_13_10_16 reconstruction overview](../outputs/22_13_10_16/overview_5_frames.png)
+
+### `22_13_56_43` (tracked, clockwise)
+
+- Dimensions: `n_frames = 100`, `n_tx = 540`, `H = W = 2048`
+- Frame rate: ~8.33 Hz
+- Rotation: `metadata/rotation/samples = +1` (clockwise)
+- Tracking: includes `metadata/pullback_position` `[100]`
+- Total size on disk: ~1.54 GB
+
+![22_13_56_43 reconstruction overview](../outputs/22_13_56_43/overview_5_frames.png)
+
+### `22_14_29_54` (tracked, clockwise)
+
+- Dimensions: `n_frames = 150`, `n_tx = 360`, `H = W = 2048`
+- Frame rate: ~3.33 Hz
+- Rotation: `metadata/rotation/samples = +1` (clockwise); `metadata/text_report = "Transducer rotation direction: clockwise."`
+- Tracking: includes `metadata/pullback_position` `[150]`
+- Total size on disk: ~1.71 GB
+
+![22_14_29_54 reconstruction overview](../outputs/22_14_29_54/overview_5_frames.png)
+
+## Dataset Quantification
+
+- Acquisitions: 9 (3 untracked `15_*`, 6 tracked `22_*`)
+- Frames per acquisition: 60–150 (1010 frames total across all acquisitions)
+- Frame rate: ~3.33 Hz or ~8.33 Hz depending on acquisition (see per-dataset table)
+- Train / validation / test split: N/A
+- Total size on disk: ~14.0 GB across all 9 acquisitions
+
+| Acquisition | Frames | `n_tx` | `H x W` | Frame rate | Rotation | Tracking | Size |
+|---|---|---|---|---|---|---|---|
+| `15_10_18_21` | 100 | 540 | 985 x 986 | ~3.33 Hz | counterclockwise | untracked | ~1.45 GB |
+| `15_10_50_19` | 60 | 360 | 985 x 986 | ~3.33 Hz | counterclockwise | untracked | ~0.59 GB |
+| `15_16_45_06` | 100 | 540 | 985 x 986 | ~3.33 Hz | counterclockwise | untracked | ~1.45 GB |
+| `22_12_10_52` | 150 | 540 | 2048 x 2048 | ~3.33 Hz | clockwise | tracked | ~2.41 GB |
+| `22_12_29_46` | 100 | 540 | 2048 x 2048 | ~8.33 Hz | clockwise | tracked | ~1.66 GB |
+| `22_12_38_45` | 150 | 360 | 2048 x 2048 | ~8.33 Hz | clockwise | tracked | ~1.65 GB |
+| `22_13_10_16` | 100 | 540 | 2048 x 2048 | ~8.33 Hz | clockwise | tracked | ~1.54 GB |
+| `22_13_56_43` | 100 | 540 | 2048 x 2048 | ~8.33 Hz | clockwise | tracked | ~1.54 GB |
+| `22_14_29_54` | 150 | 360 | 2048 x 2048 | ~3.33 Hz | clockwise | tracked | ~1.71 GB |
+
+## Subject Metadata
+
+Animal study data — no human subjects.
+
+## Data Validation
+
+A `zea.Pipeline` reconstructs the IVUS B-mode from the raw channel data
+(RF → envelope → normalization → log compression → scan conversion). See
+[reconstruct.py](../reconstruct.py) and [pipeline.yaml](../pipeline.yaml). Linear encoder position (when applicable) and segmentation masks are overlayed on the B-modes.
+
+## Known Issues
+
+- Untracked (`15_*`) acquisitions have no `pullback_position`, so the pullback
+  trajectory panel is omitted during reconstruction.
+
+## Ethical Considerations
+
+Porcine animal study data only; contains no human subjects or PHI. Collected and
+released in compliance with applicable institutional animal care approvals and
+OpenH-RF governance and data-sharing policies.

mosaic-intelligence/pipeline.yaml

@@ -0,0 +1,37 @@
+pipeline:
+    operations:
+    -   name: keras.ops.cast
+        params:
+            dtype: float32
+    -   name: band_pass_filter
+        params:
+            passband:
+            - 20000000.0
+            - 50000000.0
+    - demodulate
+    -   name: low_pass_filter
+        params:
+            complex_channels: true
+    -   name: downsample
+        params:
+            factor: 32
+    - envelope_detect
+    -   name: normalize
+        params:
+            output_range:
+            - 0.0
+            - 1.0
+    - log_compress
+    # These ops run per-frame (mapped over the batch axis), so axes refer to the
+    # per-frame (n_theta, n_rho, 1) array, not the batched tensor.
+    -   name: keras.ops.squeeze     # (n_theta, n_rho, 1) -> (n_theta, n_rho)
+        params:
+            axis: -1
+    -   name: keras.ops.transpose   # -> (n_rho, n_theta), the layout scan_convert wants
+        params:
+            axes:
+            - 1
+            - 0
+    -   name: scan_convert          # polar -> Cartesian; coordinates passed in at run time
+        params:
+            order: 1