ABSTRACT
Because of their ultra-light, ultra-thin, and flexible design, metalenses exhibit significant potential in the development of highly integrated cameras. However, the performances of metalens-integrated camera are constrained by their fixed architectures. Here we proposed a high-quality imaging method based on deep learning to overcome this constraint. We employed a multi-scale convolutional neural network (MSCNN) to train an extensive pair of high-quality and low-quality images obtained from a convolutional imaging model. Through our method, the imaging resolution, contrast, and distortion have all been improved, resulting in a noticeable overall image quality with SSIM over 0.9 and an improvement in PSNR over 3â dB. Our approach enables cameras to combine the advantages of high integration with enhanced imaging performances, revealing tremendous potential for a future groundbreaking imaging technology.
ABSTRACT
STUDY OBJECTIVE: To assess the impact of preoperative infection with the contemporary strain of severe acute respiratory coronavirus 2 (SARS-CoV-2) on postoperative mortality, respiratory morbidity and extrapulmonary complications after elective, noncardiac surgery. DESIGN: An ambidirectional observational cohort study. SETTING: A tertiary and teaching hospital in Shanghai, China. PATIENTS: All adult patients (≥ 18 years of age) who underwent elective, noncardiac surgery under general anesthesia at Huashan Hospital of Fudan University from January until March 2023 were screened for eligibility. A total of 2907 patients were included. EXPOSURE: Preoperative coronavirus disease 2019 (COVID-19) positivity. MEASUREMENTS: The primary outcome was 30-day postoperative mortality. The secondary outcomes included postoperative pulmonary complications (PPCs), myocardial injury after noncardiac surgery (MINS), acute kidney injury (AKI), postoperative delirium (POD) and postoperative sleep quality. Multivariable logistic regression was used to assess the risk of postoperative mortality and morbidity imposed by preoperative COVID-19. MAIN RESULTS: The risk of 30-day postoperative mortality was not associated with preoperative COVID-19 [adjusted odds ratio (aOR), 95% confidence interval (CI): 0.40, 0.13-1.28, P = 0.123] or operation timing relative to diagnosis. Preoperative COVID-19 did not increase the risk of PPCs (aOR, 95% CI: 0.99, 0.71-1.38, P = 0.944), MINS (aOR, 95% CI: 0.54, 0.22-1.30; P = 0.168), or AKI (aOR, 95% CI: 0.34, 0.10-1.09; P = 0.070) or affect postoperative sleep quality. Patients who underwent surgery within 7 weeks after COVID-19 had increased odds of developing delirium (aOR, 95% CI: 2.26, 1.05-4.86, P = 0.036). CONCLUSIONS: Preoperative COVID-19 or timing of surgery relative to diagnosis did not confer any added risk of 30-day postoperative mortality, PPCs, MINS or AKI. However, recent COVID-19 increased the risk of POD. Perioperative brain health should be considered during preoperative risk assessment for COVID-19 survivors.