Physics-Guided Diffusion Models for 3-D Reconstruction Framework in Lung Electrical Impedance Tomography

Authors:
Publisher website
Google Scholar
Abstract
Electrical impedance tomography (EIT) is a non-invasive, radiation-free and real-time functional imaging technique with significant potential for pulmonary monitoring. While three-dimensional (3D) reconstruction offers comprehensive impedance information of the lungs over two-dimensional (2D) reconstruction, it faces challenges from increased unknown conductivity parameters, domain expansion and severe ill-posedness of the inverse problem. To address these, we propose Physics-Guided Diffusion Reconstruction (PGDR), a novel framework integrating generative diffusion models with physical constraint. PGDR leverages diffusion models to learn 3D conductivity priors and incorporates voltage measurements into the sampling process for consistency with both voltage measurements and predictions. We employ score-based diffusion models (SDM) and denoising diffusion implicit models (DDIM) with PGDR, comparing against Graz consensus (GREIT), Tikhonov and U²-Net algorithms via simulations and phantom experiments. In simulation experiments, the Physically Guided DDIM Reconstruction (PGDDIR) method demonstrates superior performance, achieving peak signal-to-noise ratio (PSNR) gains of 18.949 dB, 29.978 dB, and 34.397 dB over U2-Net, Tikhonov, and GREIT, respectively, and structural similarity index (SSIM) improvements of 0.138, 0.059, and 0.463 compared to these methods. Its pixel accuracy (PA) and intersection over union (IoU) rank second. In tissue-mimicking phantom experiments, PGDDIR attains the highest PA and IoU values, with PSNR and SSIM ranking second. These findings underscore diffusion-based frameworks’ promise for accurate, robust 3D EIT lung imaging.
Funding Information
  • National Key Research and Development Program of China (2024YFC3016603)
  • National Natural Science Foundation of China (62171316)

This publication has 29 references indexed in Scilit: