Progressive Two-Stage Network for Low-Light Image Enhancement.

At night, visual quality is reduced due to insufficient illumination so that it is difficult to conduct high-level visual tasks effectively. Existing image enhancement methods only focus on brightness improvement, however, improving image quality in low-light environments still remains a challenging task. In order to overcome the limitations of existing enhancement algorithms with insufficient enhancement, a progressive two-stage image enhancement network is proposed in this paper. The low-light image enhancement problem is innovatively divided into two stages. The first stage of the network extracts the multi-scale features of the image through an encoder and decoder structure. The second stage of the network refines the results after enhancement to further improve output brightness. Experimental results and data analysis show that our method can achieve state-of-the-art performance on synthetic and real data sets, with both subjective and objective capability superior to other approaches.

Blood Oxygen Level-Dependent Imaging and Intravoxel Incoherent Motion MRI of Early Contrast-Induced Acute Kidney Injury in a Rabbit Model.

BACKGROUND: To evaluate the application of blood oxygenation level-dependent (BOLD)imaging and intravoxel incoherent motion (IVIM) magnetic resonance imaging (MRI) on assessing early contrast-induced acute kidney injury (CIAKI). MATERIALS: Sixty rabbits were randomly chosen to undergo iohexol (1.0, 2.5, and 5.0 [gI/kg], respectively; n = 15 for each group) or saline injection (n = 15). In each group, 6 rabbits underwent MRI at 24 h before injection and after injection of iohexol or saline (1 h and 1, 2, 3, and 4 days); meanwhile, out of the remaining 9 rabbits, 3 were chosen for MRI acquisition, and then they were killed at specific time points (1 h, 1 day, and 3 days, respectively). RESULTS: The strong attenuation of pure molecular diffusion (D), apparent diffusion coefficient (ADC), and perfusion fraction (f) was observed at 1 day, while pseudodiffusion coefficient (D*) showed a significant decrease at 1 h after iohexol injection. A distinct elevation of apparent transverse relaxation rate (R2*) reached the maximum levels on day 1, which was consistent with the expression of hypoxia-inducible factor-1α and vascular endothelial growth factor. ADC, D, and R2* correlated well with histopathological parameters and biochemical parameters. CONCLUSION: BOLD combined with IVIM is effective to monitor renal pathophysiology associated with CIAKI.

Gadolinium Retention and Clearance in the Diabetic Brain after Administrations of Gadodiamide, Gadopentetate Dimeglumine, and Gadoterate Meglumine in a Rat Model.

PURPOSE: To evaluate gadolinium (Gd) retention and clearance in the brain of diabetic rats after administrations of gadodiamide, gadopentetate dimeglumine, and gadoterate meglumine. MATERIALS AND METHODS: Both diabetic rats (n = 52) and normal rats (n = 52) intravenously received 20 injections of 0.6 mmol Gd/kg gadodiamide, gadopentetate dimeglumine, gadoterate meglumine, or saline. Both diabetic rats and normal rats were divided into 2 subgroups of 24 and 28 rats for the 7-day and 42-day evaluations (i.e., they were sacrificed at 7 days (n = 6 per group) and 42 days (n = 7 per group)), respectively, after the last injection. For the 7-day subgroup, 6 rats were euthanized for inductively coupled plasma mass spectrometry (ICP-MS) analysis. For the 42-day subgroup, 6 rats underwent T1-weighted magnetic resonance imaging (MRI) and ICP-MS, and 1 rat was analyzed by transmission electron microscopy (TEM). RESULTS: The T1 enhancements in the deep cerebellar nuclei (DCNs) of diabetic rats were lower than those of normal rats in both linear Gd-based contrast agent (GBCA) groups (p < 0.05). The average Gd concentrations in the brains of diabetic rats were significantly lower than those of healthy rats in both the short-term groups and long-term groups (p < 0.05). The highest Gd retentions were in the olfactory bulb, DCN, and striatum with gadodiamide. Compared with the results obtained 7 days after the last injection, the residual Gd concentrations of the 42-day subgroups in the brains of diabetic rats showed no significant difference in both linear GBCA groups (p>0.05). CONCLUSIONS: Compared with normal rats, the diabetic status decreased the residual Gd concentrations in the brain after multiple administrations of gadodiamide, gadopentetate dimeglumine, and gadoterate meglumine. The clearable fraction of Gd in the brain was eliminated faster in diabetic rats than in normal rats.