Radiation dose reduction with dictionary learning based processing for head CT.

摘要: In CT, ionizing radiation exposure from the scan has attracted much concern patients and doctors. This work is aimed at improving head CT images low-dose scans by using a fast Dictionary learning (DL) based post-processing. Both Low-dose (LDCT) Standard-dose (SDCT) nonenhanced were acquired in examination multi-detector row Siemens Somatom Sensation 16 scanner. One hundred involved experiments. Two groups of LDCT with 50 % (LDCT50 %) 25 % (LDCT25 %) tube current setting SDCT. To give quantitative evaluation, Signal to noise ratio (SNR) Contrast (CNR) computed Hounsfield unit (HU) measurements GM, WM CSF tissues. A blinded qualitative analysis was also performed assess processed datasets. Fifty seventy five percent dose reductions are obtained for two (LDCT50 %, 1.15 ± 0.1 mSv; LDCT25 %, 0.58 ± 0.1 mSv; SDCT, 2.32 ± 0.1 mSv; P < 0.001). Significant SNR increase over original observed all GM–WM CNR enhancement noted DL images. Higher than reference SDCT can even be achieved LDCT50 % LDCT25 % Blinded review validates perceptual improvements brought proposed approach. Compared images, application processing associated significant improvement image quality.