UPAM-KAN: A Method for Highly Undersampled Photoacoustic Microscopy Image Reconstruction

TY - JOUR

T1 - UPAM-KAN

T2 - A Method for Highly Undersampled Photoacoustic Microscopy Image Reconstruction

AU - Li, Jingtan

AU - Zhang, Shuyan

AU - Cai, Hongxin

AU - Lee, Minjun

AU - Dong, Qunxi

AU - Sun, Naidi

AU - Hu, Bin

N1 - Publisher Copyright: © 2026 Wiley-VCH GmbH.

PY - 2026/7/22

Y1 - 2026/7/22

N2 - Photoacoustic microscopy (PAM) enables structural and functional imaging at a microscopic level, playing a vital role in biomedical imaging. However, due to the limitations imposed by its scanning mechanism, PAM faces a trade-off between spatial resolution and imaging speed. While deep learning has improved PAM imaging speed, the image reconstruction performance of existing methods under high downsampling ratios still needs improvement. To address these limitations, we propose UPAM-KAN for undersampled PAM image reconstruction. This model is based on U-Net and integrates dedicated Kolmogorov–Arnold Network layers into the tokenized intermediate representations to construct its backbone network. Furthermore, three feature fusion extraction modules are proposed to enhance details, extract multi-scale features, and fuse shallow and deep features, respectively. Compared with the leading methods in undersampled PAM image reconstruction, UPAM-KAN achieves significant SSIM improvements of 8.322% and 2.692% on the Leaf Vein and Mouse Cerebrovascular datasets with only 1.4% of fully sampled pixels, respectively. Moreover, for functional reconstruction, the model pre-trained on public datasets achieves a 3.275% SSIM improvement in oxygen saturation at the 1/4 undersampling ratio. These results demonstrate that UPAM-KAN efficiently reconstructs both structural and functional information, offering insights for high-speed PAM imaging, dynamic vascular imaging and tissue functional monitoring.

AB - Photoacoustic microscopy (PAM) enables structural and functional imaging at a microscopic level, playing a vital role in biomedical imaging. However, due to the limitations imposed by its scanning mechanism, PAM faces a trade-off between spatial resolution and imaging speed. While deep learning has improved PAM imaging speed, the image reconstruction performance of existing methods under high downsampling ratios still needs improvement. To address these limitations, we propose UPAM-KAN for undersampled PAM image reconstruction. This model is based on U-Net and integrates dedicated Kolmogorov–Arnold Network layers into the tokenized intermediate representations to construct its backbone network. Furthermore, three feature fusion extraction modules are proposed to enhance details, extract multi-scale features, and fuse shallow and deep features, respectively. Compared with the leading methods in undersampled PAM image reconstruction, UPAM-KAN achieves significant SSIM improvements of 8.322% and 2.692% on the Leaf Vein and Mouse Cerebrovascular datasets with only 1.4% of fully sampled pixels, respectively. Moreover, for functional reconstruction, the model pre-trained on public datasets achieves a 3.275% SSIM improvement in oxygen saturation at the 1/4 undersampling ratio. These results demonstrate that UPAM-KAN efficiently reconstructs both structural and functional information, offering insights for high-speed PAM imaging, dynamic vascular imaging and tissue functional monitoring.

KW - deep learning

KW - functional imaging

KW - highly undersampled images

KW - image reconstruction

KW - photoacoustic microscopy

UR - https://www.scopus.com/pages/publications/105032700125

U2 - 10.1002/lpor.202500870

DO - 10.1002/lpor.202500870

M3 - Article

AN - SCOPUS:105032700125

SN - 1863-8880

VL - 20

JO - Laser and Photonics Reviews

JF - Laser and Photonics Reviews

IS - 14

M1 - e00870

ER -