Add technion README + pipeline(s) from accepted_submissions

technion/bladder/README.md

@@ -0,0 +1,147 @@
+---
+pretty_name: "OpenH-RF — Technion Bladder Pre-Beamformed Channel Data"
+license: cc-by-4.0
+task_categories:
+  - image-to-image
+tags:
+  - ultrasound
+  - iq
+  - openh-rf
+  - beamforming
+  - bladder
+  - 3d
+language:
+  - en
+size_categories:
+  - 1K<n<10K
+---
+
+# OpenH-RF — Bladder pre-beamformed RF channel data
+
+## Dataset Description
+
+Real, **in-vivo human** pre-beamformed ultrasound **channel data** for bladder
+imaging: per-element I/Q recorded before receive beamforming on a 64-element
+phased array — a sector scan of 180 transmit beams steered over ±45.13° (≈90°),
+one image line per transmit (steering angles in `scan.polar_angles`). 1,508
+frames across 14 sweeps from seven subjects. Acquired on a GE research system in
+tissue-harmonic mode; the harmonic echo is demodulated to I/Q at 3.44 MHz and
+band-pass filtered. No paired image is supplied — the B-mode is reproduced from
+the channel data by the released beamformer.
+
+## Dataset Contributor(s)
+
+Sanketh Vedula, Ortal Senouf, Dean Zadok, Alex M. Bronstein (PI) —
+Technion – Israel Institute of Technology. Primary contact: sanketh@campus.technion.ac.il.
+
+## Dataset Creation Date
+
+Source data 2018; converted to the OpenH-RF (zea) format 07/16/2026.
+
+## License / Terms of Use
+
+CC BY 4.0. The contributors confirm intent to release under CC BY 4.0 with no
+third-party IP encumbrances (proposal §8).
+
+## Intended Usage
+
+Primary: **generalized reconstruction** (§6.1) — learned receive beamforming and
+image reconstruction from raw channel data. The quasi-static bladder is also
+suited to multi-line-transmission (MLT) emulation and high-frame-rate research,
+and to anatomy/cohort interpretation (§6.5).
+
+## Dataset Characterization
+
+- **Data Collection Method:** in-vivo human (research platform) — GE Vivid S70
+  scanner with raw per-element channel access, tissue-harmonic mode.
+- **Labeling Method:** N/A — no per-frame image label; the `zea.Pipeline` in
+  `pipeline.yaml` reconstructs a B-mode from the channel data for validation.
+- **Acquisition system:** GE Vivid S70 scanner; GE 3Sc-RS 64-element phased-array
+  probe, 0.30 mm pitch; sector scan, 180 transmit beams steered over ±45.13°
+  (≈90.25° FOV), one image line per transmit.
+  Per proposal: 2.56-cycle 1.6 MHz transmit, no transmit apodization,
+  tissue-harmonic mode, harmonic echo demodulated to I/Q at 3.44 MHz and filtered,
+  ~18 fps; transversal plane with slow longitudinal probe sweep to decorrelate
+  frames.
+
+## Dataset Format
+
+zea file format, one HDF5 file per sweep (`data/<subject>.hdf5`, e.g. `a1.hdf5`,
+`ak.hdf5`, `s2.hdf5`). The source complex `double` samples were repackaged to
+`float32` I/Q with I and Q on the final channel axis (`n_ch = 2`); values are
+otherwise verbatim (band-pass filtered baseband IQ, as archived). Each file
+carries `metadata/subject/{id,type=human}`, `metadata/credit`, and
+`metadata/annotations/{anatomy=bladder, label=in vivo, view=transverse suprapubic
+pelvic ultrasound}`. Probe model (`probe.name = GE 3Sc-RS`) and scanner
+(`us_machine = GE Vivid S70`) are stored too.
+
+## Dataset Quantification
+
+- **Frames / sweeps / subjects:** 1,508 frames · 14 sweeps · 7 subjects.
+- **Train / val / test split:** N/A (contributor to define).
+- **Total size on disk:** ~90 GB.
+
+| Field | Shape | dtype | Units | Description |
+|---|---|---|---|---|
+| `data/raw_data` | `(n_frames, 180, 696, 64, 2)` | float32 | a.u. | pre-BF channel IQ: frames × tx-lines × axial × elements × {I, Q} |
+| `scan/sampling_frequency` | scalar | float32 | Hz | 3.333 MHz (IQ sample rate, from `specs`) |
+| `scan/center_frequency`, `demodulation_frequency` | scalar | float32 | Hz | 3.44 MHz (tissue-harmonic demod, from `specs`) |
+| `scan/sound_speed` | scalar | float32 | m/s | 1540 |
+| `scan/polar_angles` | `(180,)` | float32 | rad | ±45.13° steered lines (`thetaTX`) |
+| `probe/probe_geometry` | `(64, 3)` | float32 | m | element positions, 0.30 mm pitch |
+
+## Subject Metadata
+
+**Seven in-vivo human volunteers**, 14 sweeps, 1,508 frames. (The proposal's
+"six" was an undercount; verified from the acquisitions to be seven distinct
+volunteers.) No phantom is included in this collection — the calibration phantom
+is a separate submission (`../phantom/`). No PHI stored: only anonymized
+`subject.id`, `subject.type = human`, and `annotations.anatomy = bladder`.
+Age and sex were not recorded for these acquisitions.
+
+| Subject | Sweeps (files) | Frames |
+|---|---|---|
+| A | `a1`, `a2` | 215 |
+| AK | `ak` | 107 |
+| H | `h1`, `h2` | 216 |
+| O | `o1` | 108 |
+| OK | `ok1`, `ok2` | 216 |
+| P | `p1a`, `p1b`, `p2a`, `p2b` | 430 |
+| S | `s1`, `s2` | 216 |
+
+## Data Validation
+
+`reconstruct.py` reconstructs a B-mode from `raw_data` using the `zea.Pipeline`
+defined in `pipeline.yaml`: delay-and-sum beamforming on a polar scanline grid
+(one image line per transmit, receive dynamic focusing at f-number 1) → envelope
+detection → normalization → log compression → sector scan conversion. Run it on
+any file to reproduce a reference frame:
+
+```
+python reconstruct.py data/s2.hdf5 --frame 54 --out bmode_s2.png
+```
+
+Reference output: `bmode_s2.png`. The pipeline matches the acquisition's own
+receive-beamforming geometry (`code/processing/`), so the reconstruction
+reproduces the expected sector B-mode.
+
+## Known Issues
+
+- **No paired image target** (unlike the cardiac set); the B-mode is derived from
+  the channel data, not supplied.
+- **Transmit fundamental (1.6 MHz) not stored** — only the 3.44 MHz demodulation
+  frequency is in the files, so `center_frequency` equals the demodulation
+  frequency.
+
+## Ethical Considerations
+
+**Privacy safeguards (HIPAA and GDPR).** Pre-beamformed RF channel data contains
+no facial or otherwise identifying imagery. All records are de-identified to the
+HIPAA Safe Harbor standard, with direct identifiers removed and any dates
+generalized to bands. As an EU institution we additionally comply with GDPR,
+holding any pseudonymized subject identifiers separately on access-controlled
+storage and never sharing them. The released data are de-identified and contain
+only the channel signals and acquisition metadata.
+
+**Ethics.** The data were collected under ethical best practices on healthy
+volunteers.

technion/bladder/pipeline.yaml

@@ -0,0 +1,33 @@
+# B-mode beamforming config for the SLT phased-array sector dataset.
+#
+# Single-line-transmit: 180 steered transmits over a +/-45.13 deg sector,
+# per-element IQ (n_ch=2, already demodulated to baseband at 3.44 MHz).
+# Reconstructed scanline-by-scanline on a polar grid (one image line per
+# transmit, receive dynamic focusing), then scan-converted for display.
+# This mirrors the dataset's own beamformer (sltBFTRYIQ.m).
+
+parameters:
+  f_number: 0
+  selected_transmits: all
+  n_ch: 2                       # IQ (baseband) data
+  enable_scanline: true         # one image line per transmit
+  grid_type: polar              # steered rays from a common apex
+  polar_limits: [-0.7876, 0.7876]   # thetaTX min/max, radians (+/-45.13 deg)
+  zlims: [0.0, 0.1608]          # metres: n_ax * c / (2 * fs)
+  grid_size_z: 696              # depth samples per line (== n_ax)
+
+pipeline:
+  operations:
+    - name: keras.ops.cast
+      params:
+        dtype: float32
+    # No demodulate: data is already baseband IQ (n_ch=2, demodulated at 3.44 MHz).
+    - name: beamform
+      params:
+        beamformer: delay_and_sum
+        num_patches: 90
+        enable_aligned_apodization: true   # scanline one-hot transmit mask
+    - name: envelope_detect
+    - name: normalize
+    - name: log_compress
+    - name: scan_convert       # polar -> cartesian for display

technion/cardiac/README.md

@@ -0,0 +1,136 @@
+---
+pretty_name: "OpenH-RF — Technion Cardiac Pre-Beamformed Channel Data"
+license: cc-by-4.0
+task_categories:
+  - image-to-image
+tags:
+  - ultrasound
+  - iq
+  - openh-rf
+  - beamforming
+  - cardiac
+  - 3d
+language:
+  - en
+size_categories:
+  - n<1K
+---
+
+# OpenH-RF — Cardiac pre-beamformed RF channel data (paired with DAS targets)
+
+## Dataset Description
+
+Real, **in-vivo human** pre-beamformed ultrasound **channel data** for cardiac
+imaging: per-element I/Q recorded before receive beamforming on a 64-element
+phased array — a sector scan of 140 transmit beams steered over ±37.5°, one image
+line per transmit (steering angles in `scan.polar_angles`). Each frame is **paired with
+its conventional delay-and-sum reconstruction** (stored as `beamformed_data`),
+making this a ready-made input→target set for learned reconstruction /
+beamforming. 777 frames across 25 cine loops from six subjects (a–f).
+
+## Dataset Contributor(s)
+
+Sanketh Vedula, Ortal Senouf, Dean Zadok, Alex M. Bronstein (PI) —
+Technion – Israel Institute of Technology. Primary contact: sanketh@campus.technion.ac.il.
+
+## Dataset Creation Date
+
+Acquired 2018; converted to the OpenH-RF (zea) format 07/16/2026.
+
+## License / Terms of Use
+
+CC BY 4.0. The data is the contributors' own research acquisition, cleared for
+CC BY 4.0 with no third-party IP encumbrances.
+
+## Intended Usage
+
+Primary: **generalized reconstruction** (§6.1) — learning to map raw per-element
+channel data to a focused image (learned receive/transmit beamforming,
+super-resolution, clutter suppression), trained and evaluated against the paired
+delay-and-sum target. Secondary: motion estimation across the cardiac cine loops
+(§6.4) and anatomy/cohort interpretation (§6.5).
+
+## Dataset Characterization
+
+- **Data Collection Method:** in-vivo human (research platform) — GE Vivid S70
+  scanner with raw per-element channel access.
+- **Labeling Method:** derived ground truth — the paired `beamformed_data` is the
+  conventional delay-and-sum reconstruction of each frame.
+- **Acquisition system:** GE Vivid S70 scanner; GE 3Sc-RS 64-element phased-array
+  probe, 0.30 mm pitch; sector scan, 140 acquisition lines over a ~75° sector
+  (±37.5°); 2.5 MHz transmit; apical four-chamber view (A4C).
+
+## Dataset Format
+
+zea file format, one HDF5 file per cine loop (`data/<subject><clip>.hdf5`, e.g.
+`a1.hdf5` = subject a, clip 1; `f2.hdf5` = patient-set subject f). The source
+complex `int16` samples were repackaged to `float32` I/Q with I and Q on the
+final channel axis (`n_ch = 2`); values are otherwise verbatim. Each file carries
+`metadata/subject/{id,type=human}`, `metadata/credit`, and
+`metadata/annotations/{anatomy=cardiac, label=in vivo, view=apical four-chamber (A4C)}`. Probe
+model (`probe.name = GE 3Sc-RS`) and scanner (`us_machine = GE Vivid S70`) are
+stored too.
+
+## Dataset Quantification
+
+- **Frames / cines / subjects:** 777 frames · 25 cine loops · 6 subjects (a–f).
+- **Train / val / test split:** N/A (contributor to define; the `f2` patient set
+  is a natural held-out cine).
+- **Total size on disk:** ~19 GB.
+
+| Field | Shape | dtype | Units | Description |
+|---|---|---|---|---|
+| `data/raw_data` | `(n_frames, 140, 680, 64, 2)` | float32 | a.u. | pre-BF channel IQ: frames × tx-lines × axial × elements × {I, Q} |
+| `data/beamformed_data.values` | `(n_frames, 652, 140, 2)` | float32 | a.u. | paired delay-and-sum target (complex IQ): frames × depth × line × {I, Q} |
+| `data/beamformed_data.coordinates` | `(652, 140, 3)` | float32 | m | per-pixel polar coordinates (⚠️ depth scale approximate) |
+| `scan/sampling_frequency` | scalar | float32 | Hz | 6.0 MHz — **best estimate**, axial rate not stored (see Known Issues) |
+| `scan/center_frequency`, `demodulation_frequency` | scalar | float32 | Hz | 2.5 MHz (cardiac fundamental) |
+| `scan/sound_speed` | scalar | float32 | m/s | 1540 |
+| `scan/polar_angles` | `(140,)` | float32 | rad | ±37.5° steered lines |
+| `probe/probe_geometry` | `(64, 3)` | float32 | m | element positions, 0.30 mm pitch |
+
+## Subject Metadata
+
+Six subjects (a–e main set, f patient set), 777 frames across 25 cine loops.
+In-vivo human; no PHI stored (only `subject.id` a1…f2, `subject.type = human`,
+`anatomy = cardiac`, `view = apical four-chamber (A4C)`). Age and sex were not recorded for these
+acquisitions.
+
+## Data Validation
+
+`reconstruct.py` reconstructs a B-mode from `raw_data` using the `zea.Pipeline`
+defined in `pipeline.yaml`: delay-and-sum on a polar scanline grid (one image line
+per acquisition line, receive dynamic focusing) → envelope detection →
+normalization → log compression → sector scan conversion. Run:
+
+```
+python reconstruct.py data/a1.hdf5 --frame 15 --out bmode_a1.png
+```
+
+Reference output: `bmode_a1.png`. Each frame is also paired with its conventional
+delay-and-sum reconstruction in `beamformed_data` (the target for the raw→image
+learning task) — note its depth scale is approximate because the acquisition axial
+rate is not stored (see Known Issues).
+
+## Known Issues
+
+- **Axial sample rate not stored.** The consolidated source `.mat` files do not
+  carry the acquisition header, so `sampling_frequency` (6.0 MHz) is a best
+  estimate and the reconstructed depth scale is approximate. This applies both to
+  the raw→image reconstruction and to the paired `beamformed_data` target (exact
+  in value, approximate in depth axis).
+- **Sector-angle convention.** Lines are stored as ±37.5° centred about
+  boresight, the physically correct convention for a phased array.
+
+## Ethical Considerations
+
+**Privacy safeguards (HIPAA and GDPR).** Pre-beamformed RF channel data contains
+no facial or otherwise identifying imagery. All records are de-identified to the
+HIPAA Safe Harbor standard, with direct identifiers removed and any dates
+generalized to bands. As an EU institution we additionally comply with GDPR,
+holding any pseudonymized subject identifiers separately on access-controlled
+storage and never sharing them. The released data are de-identified and contain
+only the channel signals and acquisition metadata.
+
+**Ethics.** The data were collected under ethical best practices on healthy
+volunteers.

technion/cardiac/pipeline.yaml

@@ -0,0 +1,37 @@
+# B-mode beamforming config for the in-vivo cardiac sector dataset.
+#
+# Single-line-acquisition: 140 lines over a ~75 deg sector, per-element IQ
+# (n_ch=2, baseband). Reconstructed scanline-by-scanline on a polar grid
+# (one image line per acquisition, receive dynamic focusing), then scan
+# converted for display.
+#
+# NOTE: the axial sample rate and exact focus geometry are NOT stored with
+# this dataset (see convert.py / README). The depth scale is therefore
+# approximate. The authoritative reconstruction target is the paired
+# beamformed_data ("labels") shipped in the file.
+
+parameters:
+  f_number: 0
+  selected_transmits: all
+  n_ch: 2                       # IQ (baseband) data
+  enable_scanline: true         # one image line per acquisition line
+  grid_type: polar
+  polar_limits: [-0.6545, 0.6545]   # +/- 37.5 deg, radians
+  zlims: [0.0, 0.0872]          # metres: n_ax * c / (2 * fs_estimate) — APPROXIMATE
+  grid_size_z: 680
+
+pipeline:
+  operations:
+    - name: keras.ops.cast
+      params:
+        dtype: float32
+    # No demodulate: data is already baseband IQ (n_ch=2).
+    - name: beamform
+      params:
+        beamformer: delay_and_sum
+        num_patches: 70
+        enable_aligned_apodization: true
+    - name: envelope_detect
+    - name: normalize
+    - name: log_compress
+    - name: scan_convert

technion/phantom/README.md

@@ -0,0 +1,104 @@
+---
+pretty_name: "OpenH-RF — Technion Phantom Pre-Beamformed Channel Data"
+license: cc-by-4.0
+task_categories:
+  - image-to-image
+tags:
+  - ultrasound
+  - iq
+  - openh-rf
+  - beamforming
+  - phantom
+  - 3d
+language:
+  - en
+size_categories:
+  - n<1K
+---
+
+# OpenH-RF — Tissue-mimicking phantom pre-beamformed RF channel data
+
+## Dataset Description
+
+Pre-beamformed ultrasound **channel data** from a tissue-mimicking phantom,
+acquired on the same 64-element phased-array sector scheme as the in-vivo
+collection (180 transmit beams steered over ±45.13°, one image line per
+transmit), for **calibration and verification**. Contains
+resolvable point targets and an anechoic cyst — a clean reference for validating
+beamforming and reconstruction. 12 frames, one acquisition.
+
+## Dataset Contributor(s)
+
+Sanketh Vedula, Ortal Senouf, Dean Zadok, Alex M. Bronstein (PI) —
+Technion – Israel Institute of Technology. Primary contact: sanketh@campus.technion.ac.il.
+
+## Dataset Creation Date
+
+Source data 2018; converted to the OpenH-RF (zea) format 07/16/2026.
+
+## License / Terms of Use
+
+CC BY 4.0 (proposal §8).
+
+## Intended Usage
+
+Calibration and end-to-end verification of the beamforming/reconstruction
+pipeline (point-target resolution, cyst contrast). Phantom tier (×1).
+
+## Dataset Characterization
+
+- **Data Collection Method:** phantom — tissue-mimicking phantom (Gammex 403GS LE,
+  Gammex Inc., Middleton, WI, USA), acquired on the same scanner/probe as the
+  in-vivo collection for calibration.
+- **Labeling Method:** N/A (calibration target; known phantom geometry).
+- **Acquisition system:** GE Vivid S70 scanner; GE 3Sc-RS 64-element phased-array
+  probe, 0.30 mm pitch; sector scan, 180 transmit beams steered over ±45.13°
+  (≈90.25° FOV), one image line per transmit; IQ demodulated at 3.44 MHz.
+
+## Dataset Format
+
+zea file format, a single HDF5 file `data/ph.hdf5`. Source complex samples
+repackaged to `float32` I/Q (`n_ch = 2`), values verbatim. Carries
+`metadata/subject/{id=ph, type=phantom}`, `metadata/credit`, probe model
+(`probe.name = GE 3Sc-RS`) and scanner (`us_machine = GE Vivid S70`). ("phantom"
+is recorded only as `subject.type`, not as an anatomy or label.)
+
+## Dataset Quantification
+
+- **Frames / acquisitions:** 12 frames · 1 acquisition.
+- **Total size on disk:** ~0.8 GB.
+
+| Field | Shape | dtype | Units | Description |
+|---|---|---|---|---|
+| `data/raw_data` | `(12, 180, 696, 64, 2)` | float32 | a.u. | pre-BF channel IQ: frames × tx-lines × axial × elements × {I, Q} |
+| `scan/sampling_frequency` | scalar | float32 | Hz | 3.333 MHz |
+| `scan/center_frequency`, `demodulation_frequency` | scalar | float32 | Hz | 3.44 MHz |
+| `scan/sound_speed` | scalar | float32 | m/s | 1540 |
+| `scan/polar_angles` | `(180,)` | float32 | rad | ±45.13° steered lines |
+| `probe/probe_geometry` | `(64, 3)` | float32 | m | element positions, 0.30 mm pitch |
+
+## Subject Metadata
+
+N/A — inanimate phantom (GAMMEX 403GS LE); `subject.type = phantom`.
+
+## Data Validation
+
+`reconstruct.py` reconstructs a B-mode from `raw_data` using the `zea.Pipeline`
+in `pipeline.yaml` (delay-and-sum on a polar scanline grid → envelope →
+normalization → log compression → sector scan conversion). Run:
+
+```
+python reconstruct.py data/ph.hdf5 --frame 6 --out bmode_ph.png
+```
+
+Reference output: `bmode_ph.png` — resolvable point targets and a well-defined
+anechoic cyst at ~65 mm.
+
+## Known Issues
+
+- Same scan scheme and probe as the in-vivo bladder collection (GE
+  tissue-harmonic); acquired as its calibration reference. GAMMEX 403GS LE.
+
+## Ethical Considerations
+
+None — inanimate phantom, no human or animal subjects.

technion/phantom/pipeline.yaml

@@ -0,0 +1,33 @@
+# B-mode beamforming config for the SLT phased-array sector dataset.
+#
+# Single-line-transmit: 180 steered transmits over a +/-45.13 deg sector,
+# per-element IQ (n_ch=2, already demodulated to baseband at 3.44 MHz).
+# Reconstructed scanline-by-scanline on a polar grid (one image line per
+# transmit, receive dynamic focusing), then scan-converted for display.
+# This mirrors the dataset's own beamformer (sltBFTRYIQ.m).
+
+parameters:
+  f_number: 0
+  selected_transmits: all
+  n_ch: 2                       # IQ (baseband) data
+  enable_scanline: true         # one image line per transmit
+  grid_type: polar              # steered rays from a common apex
+  polar_limits: [-0.7876, 0.7876]   # thetaTX min/max, radians (+/-45.13 deg)
+  zlims: [0.0, 0.1608]          # metres: n_ax * c / (2 * fs)
+  grid_size_z: 696              # depth samples per line (== n_ax)
+
+pipeline:
+  operations:
+    - name: keras.ops.cast
+      params:
+        dtype: float32
+    # No demodulate: data is already baseband IQ (n_ch=2, demodulated at 3.44 MHz).
+    - name: beamform
+      params:
+        beamformer: delay_and_sum
+        num_patches: 90
+        enable_aligned_apodization: true   # scanline one-hot transmit mask
+    - name: envelope_detect
+    - name: normalize
+    - name: log_compress
+    - name: scan_convert       # polar -> cartesian for display