Revitalizing Convolutional Network for Image Restoration.

Image restoration aims to reconstruct a high-quality image from its corrupted version, playing essential roles in many scenarios. Recent years have witnessed a paradigm shift in image restoration from convolutional neural networks (CNNs) to Transformerbased models due to their powerful ability to model long-range pixel interactions. In this paper, we explore the potential of CNNs for image restoration and show that the proposed simple convolutional network architecture, termed ConvIR, can perform on par with or better than the Transformer counterparts. By re-examing the characteristics of advanced image restoration algorithms, we discover several key factors leading to the performance improvement of restoration models. This motivates us to develop a novel network for image restoration based on cheap convolution operators. Comprehensive experiments demonstrate that our ConvIR delivers state-ofthe- art performance with low computation complexity among 20 benchmark datasets on five representative image restoration tasks, including image dehazing, image motion/defocus deblurring, image deraining, and image desnowing.

The causality of gut microbiota on onset and progression of sepsis: a bi-directional Mendelian randomization analysis.

Background: Several observational studies have proposed a potential link between gut microbiota and the onset and progression of sepsis. Nevertheless, the causality of gut microbiota and sepsis remains debatable and warrants more comprehensive exploration. Methods: We conducted a two-sample Mendelian randomization (MR) analysis to test the causality between gut microbiota and the onset and progression of sepsis. The genome-wide association study (GWAS) summary statistics for 196 bacterial traits were extracted from the MiBioGen consortium, whereas the GWAS summary statistics for sepsis and sepsis-related outcomes came from the UK Biobank. The inverse-variance weighted (IVW) approach was the primary method used to examine the causal association. To complement the IVW method, we utilized four additional MR methods. We performed a series of sensitivity analyses to examine the robustness of the causal estimates. Results: We assessed the causality of 196 bacterial traits on sepsis and sepsis-related outcomes. Genus Coprococcus2 [odds ratio (OR) 0.81, 95% confidence interval (CI) (0.69-0.94), p = 0.007] and genus Dialister (OR 0.85, 95% CI 0.74-0.97, p = 0.016) had a protective effect on sepsis, whereas genus Ruminococcaceae UCG011 (OR 1.10, 95% CI 1.01-1.20, p = 0.024) increased the risk of sepsis. When it came to sepsis requiring critical care, genus Anaerostipes (OR 0.49, 95% CI 0.31-0.76, p = 0.002), genus Coprococcus1 (OR 0.65, 95% CI 0.43-1.00, p = 0.049), and genus Lachnospiraceae UCG004 (OR 0.51, 95% CI 0.34-0.77, p = 0.001) emerged as protective factors. Concerning 28-day mortality of sepsis, genus Coprococcus1 (OR 0.67, 95% CI 0.48-0.94, p = 0.020), genus Coprococcus2 (OR 0.48, 95% CI 0.27-0.86, p = 0.013), genus Lachnospiraceae FCS020 (OR 0.70, 95% CI 0.52-0.95, p = 0.023), and genus Victivallis (OR 0.82, 95% CI 0.68-0.99, p = 0.042) presented a protective effect, whereas genus Ruminococcus torques group (OR 1.53, 95% CI 1.00-2.35, p = 0.049), genus Sellimonas (OR 1.25, 95% CI 1.04-1.50, p = 0.019), and genus Terrisporobacter (OR 1.43, 95% CI 1.02-2.02, p = 0.040) presented a harmful effect. Furthermore, genus Coprococcus1 (OR 0.42, 95% CI 0.19-0.92, p = 0.031), genus Coprococcus2 (OR 0.34, 95% CI 0.14-0.83, p = 0.018), and genus Ruminiclostridium6 (OR 0.43, 95% CI 0.22-0.83, p = 0.012) were associated with a lower 28-day mortality of sepsis requiring critical care. Conclusion: This MR analysis unveiled a causality between the 21 bacterial traits and sepsis and sepsis-related outcomes. Our findings may help the development of novel microbiota-based therapeutics to decrease the morbidity and mortality of sepsis.

Image Restoration via Frequency Selection.

Image restoration aims to reconstruct the latent sharp image from its corrupted counterpart. Besides dealing with this long-standing task in the spatial domain, a few approaches seek solutions in the frequency domain by considering the large discrepancy between spectra of sharp/degraded image pairs. However, these algorithms commonly utilize transformation tools, e.g., wavelet transform, to split features into several frequency parts, which is not flexible enough to select the most informative frequency component to recover. In this paper, we exploit a multi-branch and content-aware module to decompose features into separate frequency subbands dynamically and locally, and then accentuate the useful ones via channel-wise attention weights. In addition, to handle large-scale degradation blurs, we propose an extremely simple decoupling and modulation module to enlarge the receptive field via global and window-based average pooling. Furthermore, we merge the paradigm of multi-stage networks into a single U-shaped network to pursue multi-scale receptive fields and improve efficiency. Finally, integrating the above designs into a convolutional backbone, the proposed Frequency Selection Network (FSNet) performs favorably against state-of-the-art algorithms on 20 different benchmark datasets for 6 representative image restoration tasks, including single-image defocus deblurring, image dehazing, image motion deblurring, image desnowing, image deraining, and image denoising.

Dual-domain strip attention for image restoration.

Image restoration aims to reconstruct a latent high-quality image from a degraded observation. Recently, the usage of Transformer has significantly advanced the state-of-the-art performance of various image restoration tasks due to its powerful ability to model long-range dependencies. However, the quadratic complexity of self-attention hinders practical applications. Moreover, sufficiently leveraging the huge spectral disparity between clean and degraded image pairs can also be conducive to image restoration. In this paper, we develop a dual-domain strip attention mechanism for image restoration by enhancing representation learning, which consists of spatial and frequency strip attention units. Specifically, the spatial strip attention unit harvests the contextual information for each pixel from its adjacent locations in the same row or column under the guidance of the learned weights via a simple convolutional branch. In addition, the frequency strip attention unit refines features in the spectral domain via frequency separation and modulation, which is implemented by simple pooling techniques. Furthermore, we apply different strip sizes for enhancing multi-scale learning, which is beneficial for handling degradations of various sizes. By employing the dual-domain attention units in different directions, each pixel can implicitly perceive information from an expanded region. Taken together, the proposed dual-domain strip attention network (DSANet) achieves state-of-the-art performance on 12 different datasets for four image restoration tasks, including image dehazing, image desnowing, image denoising, and image defocus deblurring. The code and models are available at https://github.com/c-yn/DSANet.

Bayesian inference for survival prediction of childhood Leukemia.

BACKGROUND: Childhood Leukemia is the most common type of cancer among children. Nearly 39% of cancer-induced childhood deaths are attributable to Leukemia. Nevertheless, early intervention has long been underdeveloped. Moreover, there are still a group of children succumbing to their cancer due to the cancer care resource disparity. Therefore, it calls for an accurate predictive approach to improve childhood Leukemia survival and mitigate these disparities. Existing survival predictions rely on a single best model, which fails to consider model uncertainties in predictions. Prediction from a single model is brittle, with model uncertainty neglected, and inaccurate prediction could lead to serious ethical and economic consequences. METHODS: To address these challenges, we develop a Bayesian survival model to predict patient-specific survivals by taking model uncertainty into account. Specifically, we first develop a survival model predict time-varying survival probabilities. Second, we place different prior distributions over various model parameters and estimate their posterior distribution with full Bayesian inference. Third, we predict the patient-specific survival probabilities changing with respect to time by considering model uncertainty induced by posterior distribution. RESULTS: Concordance index of the proposed model is 0.93. Moreover, the standardized survival probability of the censored group is higher than that of the deceased group. CONCLUSIONS: Experimental results indicate that the proposed model is robust and accurate in predicting patient-specific survivals. It can also help clinicians track the contribution of multiple clinical attributes, thereby enabling well-informed intervention and timely medical care for childhood Leukemia.

Postoperative analgesia efficacy of erector spinae plane block in adult abdominal surgery: A systematic review and meta-analysis of randomized trials.

Objectives: Erector spinae plane block (ESPB) has been used for many thoracic and abdominal surgeries. However, evidence of its analgesic efficacy following abdominal surgery, compared with that of thoracic analgesia, is insufficient. Our study explored the analgesic effect of ESPB after abdominal surgery. Methods: We searched PubMed, Embase, Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov. Primary outcomes were pain scores at 6, 12 and 24 h and 24-h opioid consumption. Secondary outcomes included time to first rescue analgesia, length of hospital stay, and incidence of postoperative nausea and vomiting (PONV). We calculated standardized mean differences (SMDs) with 95% confidence intervals (CIs) for primary outcomes and mean differences (MDs) and risk ratios (RRs) with 95% CIs for secondary outcomes. Results: We systematically included 1,502 cases in 24 trials. Compared with placebo, ESPB significantly reduced pain scores at 6 h (SMD -1.25; 95% CI -1.79 to -0.71), 12 h (SMD -0.85; 95% CI -1.33 to -0.37) and 24 h (SMD -0.84; 95% CI -1.30 to -0.37) and 24-h opioid consumption (SMD -0.62; 95% CI -1.19 to -0.06) post-surgery. ESPB prolonged the time to first rescue analgesia and decreased the incidence of PONV. Compared with transversus abdominal plane block (TAPB), ESPB significantly reduced pain scores at 6, 12, and 24 h and 24-h opioid consumption and prolonged the time to first rescue analgesia postsurgically. Furthermore, subgroup analysis showed that ESPB significantly reduced pain scores at various time points and opioid consumption within 24 h after laparoscopic cholecystectomy, percutaneous nephrolithotomy and bariatric surgery. Conclusion: Compared with placebo, ESPB improves the postoperative analgesic efficacy after abdominal surgery. Furthermore, our meta-analysis confirmed that ESPB provides more beneficial analgesic efficacy than TAPB. Systematic review registration: [https://www.crd.york.ac.uk/PROSPEROFILES/301491_STRATEGY_20220104.pdf], identifier [CRD42022301491].