RESUMO
Spatial-frequency shift (SFS) imaging microscopy can break the diffraction limit of fluorescently labeled and label-free samples by transferring the high spatial-frequency information into the passband of microscope. However, the resolution improvement is at the cost of decreasing temporal resolution since dozens of raw SFS images are needed to expand the frequency spectrum. Although some deep learning methods have been proposed to solve this problem, no neural network that is compatible to both labeled and label-free SFS imaging has been proposed. Here, we propose the joint spatial-Fourier channel attention network (JSFCAN), which learns the general connection between the spatial domain and Fourier frequency domain from complex samples. We demonstrate that JSFCAN can achieve a resolution similar to the traditional algorithm using nearly 1/4 raw images and increase the reconstruction speed by two orders of magnitude. Subsequently, we prove that JSFCAN can be applied to both fluorescently labeled and label-free samples without architecture changes. We also demonstrate that compared with the typical spatial domain optimization network U-net, JSFCAN is more robust to deal with deep-SFS images and noisy images. The proposed JSFCAN provides an alternative route for fast SFS imaging reconstruction, enabling future applications for real-time living cell research.
RESUMO
BACKGROUND: Bacterial meningitis has long been a severe infectious disease in neonates, as well as a leading cause of adverse outcomes. We designed this study to know the factors for poor prognosis in neonatal bacterial meningitis. METHODS: We enrolled children aged less than 1 month who were admitted to Mackay Memorial Hospital from 1984 to 2008 and had culture-proven bacterial meningitis. The laboratory data and children's clinical features were recorded. The patients' outcomes were divided into four groups: death, having sequelae, complete recovery, and loss to follow-up. Patients with the outcomes of death and having sequelae were regarded as having a poor prognosis. Those who were lost to follow-up were excluded from the analysis of outcome. Multivariate analyses were performed to find the risk factors for poor prognosis. RESULTS: One hundred fifty-six neonates fulfilled the inclusion criteria. Among these, 96 were boys (61.5%) and 102 (65.4%) had concomitant bacteremia. Group B streptococci (39.1%) and Escherichia coli (20.1%) were the two leading pathogens. Excluding those who were lost to follow-up (4.5%), 22 of 149 patients (14.8%) died, 36 (24.2%) had sequelae, and 91 (61.1%) recovered completely. Cerebrospinal fluid (CSF) protein more than 500 mg/dL at admission {odds ratio (OR): 171.18 [95% confidence interval (CI): 25.6-1000]}, predisposition to congenital heart disease [OR: 48.96 (95% CI: 6.06-395.64)], hearing impairment found during hospitalization [OR: 23.40 (95% CI: 3.62-151.25)], and seizure at admission or during hospitalization [OR: 10.10 (95% CI: 2.11-48.32)] were the factors predicting poor prognosis. CONCLUSION: In this 25-year study of newborns with bacterial meningitis, approximately one-seventh of the patients died, while two-fifths had sequelae. Nearly two-thirds of these had concomitant bacteremia. Group B streptococci and E. coli remained the two leading pathogens throughout the study period. Several factors for poor prognosis in newborns with culture-proven bacterial meningitis were found: high CSF protein concentration, congenital heart disease, hearing impairment, and seizure.