The Flow has not summarised this story yet — read it at arXiv Computer Vision.
The paper introduces a physics‑driven, cross‑domain iterative framework for self‑supervised low‑dose CT denoising. It first uses a learned sinogram prior and the LDCT noise model to separate Poisson and Gaussian noise components, then applies binomial and Gaussian data thinning to create two training pairs with independent noise realizations. These pairs train an image‑domain network whose outputs are forward‑projected to refine the prior, yielding consistent performance gains over existing self‑supervised baselines and comparable results to supervised methods.
This paper introduces a fully differentiable jitter correction technique for X‑ray phase‑contrast micro‑CT that uses a deep learning‑based image quality metric to estimate and compensate per‑projection rigid jitter directly from the acquired data, eliminating the need for a motion‑free reference scan. The method adapts a gradient‑based auto‑focus strategy to parallel‑beam geometry, benchmarks several objective functions, and validates the sensitivity of the visual information fidelity (VIF) metric to jitter artifacts. A compact 3D CNN predicts VIF scores from corrupted volumes, while a spatially selective total variation penalty suppresses spurious high‑frequency structures during optimization; experiments on biological specimens from multiple synchrotron beamlines confirm that the pipeline reliably restores fine structural detail across morphologically distinct samples.
EPC‑RawDC is a 3‑D latent diffusion model that synthesises CT images from CBCT scans by enforcing two physics‑based constraints: a geometry‑equivariant projection loss and raw data consistency using scan‑specific CBCT projections. Trained on two clinical cohorts with different scanner geometries, the method achieved SSIM of 0.97 and improved PSNR and MAE over a baseline DDPM, reducing projection and equivariant errors by 55 % and 41 % respectively.
GRIPNet is a new CT‑based pulmonary nodule detector that incorporates a Gaussian radial intensity prior, reflecting the regular pattern of intensity peaks at nodule centers and Gaussian decay outward. By replacing generic square convolutions with pinwheel, dual‑frequency, dilated masked attention, and an adaptive loss, the network aligns each module with measurable intensity properties. The method achieves state‑of‑the‑art mAP@0.5 scores of 95.3%, 91.6%, and 97.9% on KanserSet, LUNA16, and Lung‑PET‑CT‑Dx, respectively, while maintaining real‑time speed and improving high‑IoU localization.
arXiv:2607. 28759v1 Announce Type: cross Abstract: In X-ray CT, metallic objects cause beam hardening, photon starvation, and scattering, leading to projection inconsistency, streaks, dark bands, and structural distortions that compromise clinical diagnosis and quantitative analysis.
arXiv:2606. 16212v1 Announce Type: cross Abstract: Sparse-view CT reduces radiation dose and scanning time by acquiring fewer projection views, but angular undersampling makes reconstruction severely ill-posed, causing streak artifacts, structural blurring, and loss of fine details.