Untrained CNN and coefficient-space regularization for joint reconstruction in photoacoustic tomography
Abstract
The joint reconstruction of the initial pressure and spatially varying sound speed in photoacoustic tomography from boundary measurements is considered. A regularized data-driven framework is proposed in which the unknown fields are represented in finite-dimensional wavelet spaces and fixed nonlinear features generated by an untrained convolutional neural network (CNN) are used to reconstruct the initial-pressure coefficients through Tikhonov-regularized least squares. A learned coefficient-space prior is subsequently used to infer the squared sound-speed coefficients. Numerical experiments demonstrate accurate reconstruction on held-out data and strong generalization to previously unseen phantom geometries. The frozen reconstruction map also remains stable under measurement noise not represented during training, with only minor changes in reconstruction accuracy for noise levels up to 10%. In comparison with the supervised trained-CNN approach considered in this work, the untrained representation yields substantially smaller reconstruction errors while requiring approximately one-sixth of the computational time. These results demonstrate that fixed random convolutional features combined with regularized coefficient-space learning provide an efficient and robust approach to joint reconstruction in photoacoustic tomography.
Related
- Initial condition recovery in nonlinear damped viscous photoacoustic tomography using a convolutional neural network-guided gradient-free optimization framework
- 3D Uncertainty Quantification for Photoacoustic Tomography
- CLEAR: Complex Learned Explicit Analytical Regularization for Ultra-Accelerated 4D Flow CMR Reconstruction
- Reconstruction and well-posedness of an inverse coefficient problem for a diffusion equation
- Simultaneous Drift Calibration and Reconstruction for Scanning-Probe X-Ray Tomography
BibTeX
@misc{bozi2026untrained,
title = {{Untrained CNN and coefficient-space regularization for joint reconstruction in photoacoustic tomography}},
author = {Alfio Bozi and Anwesa Dey and Souvik Roy},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.23315},
eprint = {2610.23315},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.23315},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Untrained CNN and coefficient-space regularization for joint reconstruction in photoacoustic tomography
AU - Alfio Bozi
AU - Anwesa Dey
AU - Souvik Roy
PY - 2026
DA - 2026/10/16
PB - arXiv
DO - 10.48550/arxiv.2610.23315
UR - https://arxiv.org/abs/2610.23315
AB - The joint reconstruction of the initial pressure and spatially varying sound speed in photoacoustic tomography from boundary measurements is considered. A regularized data-driven framework is proposed in which the unknown fields are represented in finite-dimensional wavelet spaces and fixed nonlinear features generated by an untrained convolutional neural network (CNN) are used to reconstruct the initial-pressure coefficients through Tikhonov-regularized least squares. A learned coefficient-space prior is subsequently used to infer the squared sound-speed coefficients. Numerical experiments demonstrate accurate reconstruction on held-out data and strong generalization to previously unseen phantom geometries. The frozen reconstruction map also remains stable under measurement noise not represented during training, with only minor changes in reconstruction accuracy for noise levels up to 10%. In comparison with the supervised trained-CNN approach considered in this work, the untrained representation yields substantially smaller reconstruction errors while requiring approximately one-sixth of the computational time. These results demonstrate that fixed random convolutional features combined with regularized coefficient-space learning provide an efficient and robust approach to joint reconstruction in photoacoustic tomography.
ER -