# Moving beyond simulation: data-driven quantitative photoacoustic imaging using tissue-mimicking phantoms

arXiv: https://arxiv.org/abs/2306.06748 · HTML: https://mlsanity.com/paper/2306.06748/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 52 references that are in the map.
- First posted on arXiv 11 Jun 2023.

## Builds on (most cited first, 6 of 6)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Model based learning for accelerated, limited-view 3D photoacoustic tomography](https://mlsanity.com/paper/1708.09832/) (2017)
- [Deep learning for biomedical photoacoustic imaging: A review](https://mlsanity.com/paper/2011.02744/) (2020)
- [Deep Learning for Photoacoustic Tomography from Sparse Data](https://mlsanity.com/paper/1704.04587/) (2017)
- [Unsupervised Domain Transfer with Conditional Invertible Neural Networks](https://mlsanity.com/paper/2303.10191/) (2023)
- [Photoacoustic image synthesis with generative adversarial networks](https://mlsanity.com/paper/2103.15510/) (2021)

## Built on by (most cited first, 10 of 10)

- [A study of why we need to reassess full reference image quality assessment with medical images](https://mlsanity.com/paper/2405.19097/) (2024)
- [A study on the adequacy of common IQA measures for medical images](https://mlsanity.com/paper/2405.19224/) (2024)
- [Digital twins enable full-reference quality assessment of photoacoustic image reconstructions](https://mlsanity.com/paper/2505.24514/) (2025)
- [Parameter choices in HaarPSI for IQA with medical images](https://mlsanity.com/paper/2410.24098/) (2024)
- [Distribution-informed and wavelength-flexible data-driven photoacoustic oximetry](https://mlsanity.com/paper/2403.14863/) (2024)
- [Joint Segmentation and Image Reconstruction with Error Prediction in Photoacoustic Imaging using Deep Learning](https://mlsanity.com/paper/2407.02653/) (2024)
- [PhotIQA: A photoacoustic image data set with image quality ratings](https://mlsanity.com/paper/2507.03478/) (2025)
- [Application of a Virtual Imaging Framework for Investigating a Deep Learning-Based Reconstruction Method for 3D Quantitative Photoacoustic Computed Tomography](https://mlsanity.com/paper/2510.03431/) (2025)
- [Optical Inversion and Spectral Unmixing of Spectroscopic Photoacoustic Images with Physics-Informed Neural Networks](https://mlsanity.com/paper/2602.16357/) (2026)
- [Deep Image Prior for photoacoustic tomography can mitigate limited-view artifacts](https://mlsanity.com/paper/2604.19176/) (2026)

## Abstract (arXiv)

Accurate measurement of optical absorption coefficients from photoacoustic imaging (PAI) data would enable direct mapping of molecular concentrations, providing vital clinical insight. The ill-posed nature of the problem of absorption coefficient recovery has prohibited PAI from achieving this goal in living systems due to the domain gap between simulation and experiment. To bridge this gap, we introduce a collection of experimentally well-characterised imaging phantoms and their digital twins. This first-of-a-kind phantom data set enables supervised training of a U-Net on experimental data for pixel-wise estimation of absorption coefficients. We show that training on simulated data results in artefacts and biases in the estimates, reinforcing the existence of a domain gap between simulation and experiment. Training on experimentally acquired data, however, yielded more accurate and robust estimates of optical absorption coefficients. We compare the results to fluence correction with a Monte Carlo model from reference optical properties of the materials, which yields a quantification error of approximately 20%. Application of the trained U-Nets to a blood flow phantom demonstrated spectral biases when training on simulated data, while application to a mouse model highlighted the ability of both learning-based approaches to recover the depth-dependent loss of signal intensity. We demonstrate that training on experimental phantoms can restore the correlation of signal amplitudes measured in depth. While the absolute quantification error remains high and further improvements are needed, our results highlight the promise of deep learning to advance quantitative PAI.