Multi-Scale Shape Adaptive Network for Raindrop Detection and Removal from a Single Image.

Removing raindrops from a single image is a challenging problem due to the complex changes in shape, scale, and transparency among raindrops. Previous explorations have mainly been limited in two ways. First, publicly available raindrop image datasets have limited capacity in terms of modeling raindrop characteristics (e.g., raindrop collision and fusion) in real-world scenes. Second, recent deraining methods tend to apply shape-invariant filters to cope with diverse rainy images and fail to remove raindrops that are especially varied in shape and scale. In this paper, we address these raindrop removal problems from two perspectives. First, we establish a large-scale dataset named RaindropCityscapes, which includes 11,583 pairs of raindrop and raindrop-free images, covering a wide variety of raindrops and background scenarios. Second, a two-branch Multi-scale Shape Adaptive Network (MSANet) is proposed to detect and remove diverse raindrops, effectively filtering the occluded raindrop regions and keeping the clean background well-preserved. Extensive experiments on synthetic and real-world datasets demonstrate that the proposed method achieves significant improvements over the recent state-of-the-art raindrop removal methods. Moreover, the extension of our method towards the rainy image segmentation and detection tasks validates the practicality of the proposed method in outdoor applications.

Efficacy and safety of blood purification in the treatment of deep burns: A systematic review and meta-analysis.

INTRODUCTION: This meta-analysis aimed to systematically review and evaluate randomized controlled trials (RCTs) and cohort studies examining the efficacy and safety of blood purification in the treatment of patients with deep burns. METHODS: The PubMed, Cochrane Library, and Embase databases and relevant references were systematically searched for RCTs and cohort studies published until the end of September 2020 to investigate the potential of blood purification in improving the prognosis of severely burned patients. The primary outcome of this systematic review was overall patient mortality; secondary outcomes included the incidence of sepsis and infection prevention (vital signs and routine blood tests). RESULTS: A total of 6 RCTs and 1 cohort study were included, with a total of 538 burn patients (274 patients who received blood purification and 264 control patients). Compared with patients who received conventional treatment, those treated with blood purification displayed significant 2-day reduction in mortality and sepsis with relative risks of 0.62 and 0.41, respectively (95% confidence intervals [CIs], 0.74-0.82 and 0.25-0.67, respectively; P < .05). In terms of vital signs and blood biochemistry, the respiratory rates and blood urea nitrogen levels of patients in the blood purification group 3 days post-treatment were significantly higher than those in the control group (randomized standard deviations (SMDs), 0.78 and 0.77, respectively; 95% CIs, 0.33-1.23 and 1.22-0.31, respectively; P < .05). However, there were no significant differences between groups on day 3 with regard to temperature (P = .32), heart rate (P = .26), white blood cell count (P = .54), or neutrophil count (P = .74), potentially owing to the small sample size or the relatively short intervention time. Heterogeneous differences existed between the groups with respect to blood urea nitrogen (SMD = -1.22; 95% CI, -2.16 to -0.40; P < .00001) and Cr (SMD = -3.13; 95% CI, -4.92 to -1.33; P < .00001) on day 7. No systematic adverse events occurred. CONCLUSIONS: Blood purification treatment for deep burn patients can significantly reduce the mortality rate and the incidence of complications.