Войти
аналитика - Medical imaging reconstruction
Non-Parametric Bayesian Deep Learning Approach for Uncertainty-Aware MRI Reconstruction from Undersampled Data
A non-parametric Bayesian deep learning framework is proposed to reconstruct high-quality MRI images from undersampled k-space data while providing quantitative measures of uncertainty in the reconstructed images.
Reconstructing High-Quality Standard-Dose PET Images from Multi-Dose-Level Low-Dose PET Images with Dose Level Awareness
A novel two-phase framework with dose level awareness is proposed to effectively reconstruct high-quality standard-dose PET images from multi-dose-level low-dose PET images.
Robust and Structure-Preserving Reconstruction of Limited Angle Cone-Beam Computed Tomography Using a Prior Frequency-Guided Diffusion Model
A prior frequency-guided diffusion model (PFGDM) framework is developed to robustly and accurately reconstruct limited angle cone-beam computed tomography (LA-CBCT) images by leveraging high-frequency information from patient-specific prior CT scans as anatomical priors.
Продукты | Ресурсы
© 2024 by Linnk AI