PulseAugur
EN
LIVE 08:51:27

New self-supervised method enhances CT scan reconstruction without ground-truth data

By PulseAugur Editorial · [3 sources] ·

Researchers have developed a self-supervised method called Noise2Inverse Learned Primal-Dual (N2I-LPD) to improve X-ray computed tomography reconstruction. This new approach allows for the training of reconstruction operators without requiring ground-truth data, which is often unavailable in practical scenarios like low-dose or sparse-angle imaging. By leveraging the statistical independence of noise in CT scans, N2I-LPD demonstrates improved reconstruction quality compared to classical methods and other neural network approaches like U-Net. AI

IMPACT Enables more accurate medical imaging in scenarios where ground-truth data is unavailable, potentially improving diagnostic capabilities.

RANK_REASON The cluster contains a research paper detailing a new self-supervised learning method for image reconstruction.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

New self-supervised method enhances CT scan reconstruction without ground-truth data

COVERAGE [3]

  1. arXiv cs.LG TIER_1 English(EN) · Antti S\"allinen, Siiri Rautio, Santeri Kaupinm\"aki, Andreas Hauptmann ·

    Enabling self-supervised learned primal dual with Noise2Inverse

    arXiv:2606.26991v1 Announce Type: cross Abstract: X-ray computed tomography reconstruction is an ill-posed inverse problem, particularly in low-dose and sparse-angle settings where measurements are noisy and incomplete. While learned reconstruction methods such as the Learned Pri…

  2. arXiv cs.LG TIER_1 English(EN) · Andreas Hauptmann ·

    Enabling self-supervised learned primal dual with Noise2Inverse

    X-ray computed tomography reconstruction is an ill-posed inverse problem, particularly in low-dose and sparse-angle settings where measurements are noisy and incomplete. While learned reconstruction methods such as the Learned Primal-Dual algorithm achieve strong performance, the…

  3. Hugging Face Daily Papers TIER_1 English(EN) ·

    Enabling self-supervised learned primal dual with Noise2Inverse

    X-ray computed tomography reconstruction is an ill-posed inverse problem, particularly in low-dose and sparse-angle settings where measurements are noisy and incomplete. While learned reconstruction methods such as the Learned Primal-Dual algorithm achieve strong performance, the…

RELATED ENTITIES

RELATED TOPICS