Answers are grounded in this paper and cite their sources.
Electromagnetic interference (EMI) in photoacoustic tomography (PAT) can arise from receive-chain electronics, grounding paths, laser/DAQ synchronization, and environmental coupling, and may appear as structured stripe artifacts and baseline drift in raw channel data. In this work, we propose a physics-motivated, structure-aware, learning-free Structured-Sequential Denoising (SSD) framework for ring-based PAT. SSD first suppresses quasi-synchronous cross-channel stripe-like interference using robust order-statistic processing and IQR-based time-index detection, and then performs single-channel temporal restoration to reduce baseline drift and residual random noise. Experiments on simulated and in vivo finger and zebrafish datasets show that SSD improves raw-signal and reconstructed-image quality while preserving acoustic signal fidelity under the tested ring-array setting. Compared with selected conventional, non-local, guided-filtering, and adaptive-filtering baselines, SSD achieved the best overall PSNR/SSIM on the simulation dataset and improved the mean A-line SNR by 19.21 dB and 24.05 dB over the noisy input across the finger and zebrafish datasets, respectively.
No takes yet. Share an insight, caveat, or question.
Sun et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: