Code supplementing the paper OpenQPAIData: Measurements of tissue-mimicking phantoms with reference labels for quantitative photoacoustic imaging research

Welcome to the repository accompanying the OpenQPAIData dataset, a multi-device photoacoustic imaging dataset of 30 tissue-mimicking phantoms, imaged with known optical properties. This repository contains code examples, visualization scripts, and interpolation tools to explore and utilize the dataset for algorithm development and validation in quantitative photoacoustic imaging (qPAI).

Authors and Affiliations

• Janek Gröhl
  Department of Physics, University of Cambridge, United Kingdom
  Cancer Research UK Cambridge Institute, University of Cambridge, United Kingdom
  ENI-G, a Joint Initiative of the University Medical Center Göttingen and the Max Planck Institute for Multidisciplinary Sciences, Göttingen, Germany

• Sandeep Kumar Kalva
  Institute of Pharmacology and Toxicology and Institute for Biomedical Engineering, Faculty of Medicine, University of Zurich, Switzerland
  Institute for Biomedical Engineering, Department of Information Technology and Electrical Engineering, ETH Zurich, Switzerland
  Department of Biosciences and Bioengineering, Indian Institute of Technology Bombay, Mumbai, India

• Berkan Lafci
  Institute of Pharmacology and Toxicology and Institute for Biomedical Engineering, Faculty of Medicine, University of Zurich, Switzerland
  Institute for Biomedical Engineering, Department of Information Technology and Electrical Engineering, ETH Zurich, Switzerland

• Francesca Di Cecio
  Department of Physics, University of Cambridge, United Kingdom
  Cancer Research UK Cambridge Institute, University of Cambridge, United Kingdom

• Ran Tao
  Department of Physics, University of Cambridge, United Kingdom
  Cancer Research UK Cambridge Institute, University of Cambridge, United Kingdom

• Thomas R. Else
  Department of Physics, University of Cambridge, United Kingdom
  Cancer Research UK Cambridge Institute, University of Cambridge, United Kingdom
  Now at: Department of Bioengineering, Imperial College London, United Kingdom

• Lorna Wright
  Department of Physics, University of Cambridge, United Kingdom
  Cancer Research UK Cambridge Institute, University of Cambridge, United Kingdom

• Daniel Razansky
  Institute of Pharmacology and Toxicology and Institute for Biomedical Engineering, Faculty of Medicine, University of Zurich, Switzerland
  Institute for Biomedical Engineering, Department of Information Technology and Electrical Engineering, ETH Zurich, Switzerland

• Sarah E. Bohndiek
  Department of Physics, University of Cambridge, United Kingdom
  Cancer Research UK Cambridge Institute, University of Cambridge, United Kingdom

What’s in the Dataset?

• 30 tissue-mimicking phantoms with varied optical absorption and scattering properties.
• Multi-device imaging from:
  • MSOT InVision 256-TF (iThera Medical)
  • TROPUS (Transmission–Reflection Optoacoustic and Ultrasound)
  • SVOT (Spiral Volumetric Optoacoustic Tomography)
• Ground truth labels of absorption (μa) and reduced scattering (μs′) via double-integrating-sphere measurements.
• Manual segmentations and interpolated 3D digital twins for each phantom.

Repository Structure

OpenQPAIData/
├── examples/          # Minimal working examples for loading the dataset
├── scripts/           # SVOT 2D slice interpolation to full 3D segmentations
├── visualisations/    # Scripts to reproduce all figures from the paper

examples/

Contains Python scripts demonstrating how to:

read_DIS_data.py: read the double integrating sphere measurement data.

read_PAI_data.py: read the photoacoustic imaging data, reference segmentations, and reference absorption and scattering distributions.

scripts/

interpolate_svot_segmentations.py: Reconstructs full 3D segmentation masks from sparse annotated slices for the SVOT system.

visualisations/

Scripts to reproduce all main figures from the accompanying publication, including device-specific examples and property distributions.

Dataset Access

The full dataset is available open access at:

https://doi.org/10.5281/zenodo.14044853

Dataset folders include:

• DIS/: Raw and processed data from double-integrating-sphere optical property measurements.
• invision/, tropus/, svot/: Imaging data from three devices, including reconstructions, labels, and optical property maps.
• material_mapping.json: Links segmentation labels to material-specific optical properties.

Data formats:

• Raw time series: IPASC format
• Reconstructed images and labels: .npy (NumPy arrays)

Citation

If you use this dataset or code in your research, please cite the accompanying paper.

Acknowledgements

This work was supported by the DFG (GR 5824/1 and GR 5824/2), Cancer Research UK (A29580), and others. See the full paper for the complete author list and funding information.

---

License: MIT Dataset License: CC-BY-4.0

Contact

For questions or feedback, please contact the corresponding author