Accurate segmentation of retinal vessels is a basic step in diabetic retinopathy (DR) detection. Most methods based on deep convolutional neural network (DCNN) have small receptive fields, and hence ...they are unable to capture global context information of larger regions, with difficult to identify pathological. The final segmented retina vessels contain more noise with low classification accuracy. Therefore, in this paper, we propose a DCNN structure named as D-Net. In the encoding phase, we reduced the loss of feature information by reducing the downsampling factor, which reduced the difficulty of tiny thin vessels segmentation. We use the combined dilated convolution to effectively enlarge the receptive field of the network and alleviate the "grid problem" that exists in the standard dilated convolution. In the proposed multi-scale information fusion module (MSIF), parallel convolution layers with different dilation rates are used, so that the model can obtain more dense feature information and better capture retinal vessel information of different sizes. In the decoding module, the skip layer connection is used to propagate context information to higher resolution layers, so as to prevent low-level information from passing the entire network structure. Finally, our method was verified on DRIVE, STARE, and CHASE dataset. The experimental results show that our network structure outperforms some state-of-art method, such as N 4 -fields, U-Net, and DRIU in terms of accuracy, sensitivity, specificity, and <inline-formula> <tex-math notation="LaTeX">{AUC_{ROC}} </tex-math></inline-formula>. Particularly, D-Net outperforms U-Net by 1.04 %, 1.23 %, and 2.79% in DRIVE, STARE, and CHASE dataset, respectively.
Daytime napping has been postulated as both a protective and a risk factor for depression in previous studies. In addition to these conflicting results, research gaps also exist with regard to ...controlling confounding bias between daytime napping and depression and examining the potential association within the Chinese population. To facilitate the prevention and diagnosis of depression, this study aims to provide insight into the association of daytime napping and depression in 0.5 million Chinese adults by fully controlling confounders, and further examine the modifying effects of socio-economic status (SES) and age.
Data were drawn from the baseline of a Chinese cohort study of 0.5 million adults. Depressive status was measured by the Composite International Diagnostic Inventory (CIDI). Logistic regression models were used to examine the association between depression and daytime napping adjusted for SES, sleep-related factors, lifestyle factors and related diseases. Further stratified analyses were conducted to identify the modifying effects of socio-economic status and age.
The odds ratio of depression by daytime napping was 1.15 (95% CI: 1.01-1.31) in females and 1.42 (95% CI: 1.18-1.71) in males. Factors including living in a rural area (OR = 1.31, 95% CI: 1.13-1.52), receiving less education (OR = 1.42, 95% CI: 1.22-1.66), getting married (OR = 1.24, 95% CI: 1.10-1.40) and being 45-65 years old (OR = 1.29, 95% CI: 1.12-1.49) had a modifying effect on daytime napping and depression that could strengthen the association.
A significantly positive association was found between depression and daytime napping, as well as daytime dysfunction, snoring and both shorter and longer sleep duration. Lower SES and age could possibly modify the association. Further clinical or epidemiological studies are needed to investigate the mechanism and facilitate the prevention of depression.
Glaucoma is a chronic eye disease that causes loss of vision and it is irreversible. Accurate segmentation of optic disc and optic cup is a basic step in screening glaucoma. The most existing deep ...convolutional neural network (DCNN) methods have insufficient feature information extraction, and hence they are susceptible to pathological regions and low-quality images, with have poor ability to restore context information. Finally, the accuracy of the model segmentation is low. In this paper, we propose GL-Net, a multi-label DCNN model that combines the generative adversarial networks. GL-Net consists of two network structures including a Generator and a Discriminator. In the Generator, we use skip connection to promote the fusion of low-level feature information and high-level feature information, which alleviates the difficulty of restoring detailed feature information during upsampling, and reduces the downsampling factor, effectively alleviating excessive feature information loss. In the loss function, we add the <inline-formula> <tex-math notation="LaTeX">L_{1} </tex-math></inline-formula> distance function and the cross-entropy function to prevent the mode collapse when the model is trained, which makes the segmentation result more accurate. We use transfer learning and data augmentation to alleviate the problem of insufficient data and over-fitting of the model during training. Finally, GL-Net was verified on DRISHTI-GS1 dataset. The experimental results show that GL-Net outperforms some state-of-the-art method, such as M-Net, Stack-U-Net, RACE-net, and BCRF in terms of <inline-formula> <tex-math notation="LaTeX">F1 </tex-math></inline-formula> and boundary distance localization error (BLE). Particularly, in the optic cup segmentation, GL-Net outperforms RACE-net by 3.5% and 4.16 pixels in terms of <inline-formula> <tex-math notation="LaTeX">F1 </tex-math></inline-formula> and BLE, respectively.
Reliable estimates of saturated hydraulic conductivity (Ks) are usually difficult to obtain, as Ks is regulated by a variety of soil processes acting at different spatial scales that may obscure each ...other's impacts. We hypothesized that these scale‐specific relationships could be well characterized with the aid of the noise‐assisted multivariate empirical mode decomposition (NA‐MEMD), thereby serving as a solid foundation for an accurate Ks prediction. The objective was to evaluate whether the incorporation of NA‐MEMD could improve the estimation of Ks based on the multiscale associations it unraveled. On a typical slope transect of 860 m in the black soil region of northeast China, Ks, mollic epipedon thickness, bulk density, soil organic carbon content, total (ϕ) and effective porosities (ϕe), and particle size distribution were investigated at every 20 m. Prior to NA‐MEMD, Ks was most strongly correlated with ϕe, and the linear regression models based on ϕe solely were satisfactory for Ks estimation at the scale of investigation. Adding other predictors significantly improved Ks prediction in calibration, but impaired it in validation. Upon decomposition by NA‐MEMD, Ks was found to be significantly associated with each attribute at two scales of oscillation at least. Summing up the estimates of each Ks component derived from the properties at the equivalent oscillation scales, the results outperformed the traditional multiple linear regressions made at the investigation scale, when the same sets of predictors were used. The application of NA‐MEMD, moreover, could save the tedious measurements of ϕe and ϕ. Excluding these two porosity‐related properties, Ks estimates obtained by incorporating NA‐MEMD were statistically similar or even better than those involving them before NA‐MEMD. These findings demonstrate the great potential of NA‐MEMD in untangling scale‐dependent relationships of Ks with various processes and hold important implications for future estimations of Ks and other hydraulic properties in the black soil region of northeast China and similar regions.
Core Ideas
Saturated hydraulic conductivity (Ks) was most strongly correlated with effective porosity (ϕe) in general both before and after decomposition with noise‐assisted multivariate empirical mode decomposition (NA‐MEMD).
Linear regressions based on ϕe only were adequate for Ks prediction at the scale of investigation.
Combining NA‐MEMD with multiple linear regression (MLR) predicted Ks more accurately than MLR performed at the measurement scale.
Incorporating NA‐MEMD could save tedious measurements of total (ϕ) and effective (ϕe) porosities for Ks estimation.
To effectively utilize a feature set to further improve fault diagnosis of a rolling bearing vibration signal, a method based on multi-fractal detrended fluctuation analysis (MF-DFA) and smooth ...intrinsic time-scale decomposition (SITD) was proposed. The vibration signal was decomposed into several proper rotation components by applying this new SITD method to overcome noise effects, preserve the effective signal, and improve the signal-to-noise ratio. Wavelet analysis was embedded in iteration procedures of intrinsic time-scale decomposition (ITD). For better results, an adaptive threshold function was used for signal recovery from noisy proper rotation components in the wavelet domain. Additionally, MF-DFA was used to reveal the multi-fractality present in the instantaneous amplitude of the proper rotation components. Finally, linear local tangent space alignment was applied for feature dimension reduction and to obtain fault characteristics of different types, further improving identification accuracy. The performance of the proposed method is determined to be superior to that of the ITD-MF-DFA method.
Ternary blend films were prepared with different ratios of starch/polyvinyl alcohol (PVA)/citric acid. The films were characterized by field emission scanning electron microscopy (FE-SEM), ...thermogravimetric analysis, as well as Fourier transform infrared (FTIR) analysis. The influence of different ratios of starch/polyvinyl alcohol (PVA)/citric acid and different drying times on the performance properties, transparency, tensile strength (TS), water vapor permeability (WVP), water solubility (WS), color difference (ΔE), and antimicrobial activity of the ternary blends films were investigated. The starch/polyvinyl alcohol/citric acid (S/P/C
, S/P/C
, and S/P/C
) films were all highly transparent. The S/P/C
had a 54.31 times water-holding capacity of its own weight and its mechanical tensile strength was 46.45 MPa. In addition, its surface had good uniformity and compactness. The S/P/C
and S/P/C
showed strong antimicrobial activity to
and
, which were the food-borne pathogenic bacteria used. The freshness test results of fresh figs showed that all of the blends prevented the formation of condensed water on the surface of the film, and the S/P/C
and S/P/C
prevented the deterioration of figs during storage. The films can be used as an active food packaging system due to their strong antibacterial effect.
Purpose
To observe the clinical effect of implanting intraocular lens (IOL) without viscoelastic agent in phacoemulsification cataract surgery.
Methods
A retrospective analysis of 87 patients (87 ...eyes) with senile cataracts was performed. According to whether viscoelastic agents were injected before IOL implantation, they were divided into 2 groups. The endothelial cell density (ECD), intraocular pressure (IOP), uncorrected distance visual acuities (UDVA), and operation complications were evaluated preoperatively and postoperatively, and operation time was recorded.
Results
The experimental group (no viscoelastic group) contained 42 cases (42 eyes), and the control group (viscoelastic group) contained 45 cases (45 eyes). The operation time of the experimental group (7.261 ± 0.75 s) was significantly shorter than that of the control group (7.802 ± 1.14 s) (
t
= −2.629,
P
= 0.01). The IOP of the experimental group (16.71 ± 2.04 mmHg) was significantly lower than that of the control group (19.56 ± 8.98 mmHg) at 4 h after surgery (
t
= −2.065,
P
= 0.04). There was no statistically significant difference in UCDV, ECD between the two groups of patients (
P
> 0.05). There were no complications of intraoperative PCR in both groups.
Conclusions
Implantation of IOL without viscoelastic agent can shorten the operation time and reduce the occurrence of high IOP in the early postoperative period.
Idiopathic pulmonary fibrosis (IPF) refers to chronic progressive fibrotic interstitial pneumonia. It is called a "tumor-like disease" and cannot be cured using existing clinical drugs. Therefore, ...new treatment options are urgently needed. Studies have proven that ferroptosis is closely related to the development of IPF, and ferroptosis inhibitors can slow down the occurrence of IPF by chelating iron or reducing lipid peroxidation. For example, the ferroptosis inhibitor deferoxamine (DFO) was used to treat a mouse model of pulmonary fibrosis, and DFO successfully reversed the IPF phenotype and increased the survival rate of mice from 50% to 90%. Given this, we perceive that the treatment of IPF by delivering ferroptosis inhibitors is a promising option. However, the delivery of ferroptosis inhibitors faces two bottlenecks: low solubility and targeting. For one thing, we consider preparing ferroptosis inhibitors into nanomedicines to improve solubility. For another thing, we propose to deliver nanomedicines through pulmonary drug-delivery system (PDDS) to improve targeting. Compared with oral or injection administration, PDDS can achieve better delivery and accumulation in the lung, while reducing the systemic exposure of the drug, and is an efficient and safe drug-delivery method. In this paper, three possible nanomedicines for PDDS and the preparation methods thereof are proposed to deliver ferroptosis inhibitors for the treatment of IPF. Proper administration devices and challenges in future applications are also discussed. In general, this perspective proposes a promising strategy for the treatment of IPF based on inhalable nanomedicines carrying ferroptosis inhibitors, which can inspire new ideas in the field of drug development and therapy of IPF.
Reading speed in intermittent exotropia (IXT) children has been minimally examined. This study assessed reading speed in school-age children with IXT and determined clinical characteristics of IXT ...that impacted their reading ability. We compared the reading speed of 63 school-age (10-14 years) children with IXT to 44 age-matched normal counterparts. In addition, the correlation between reading speed and clinical characteristics of IXT were evaluated. The reading speed in children with IXT was 231 ± 51 CPM, while reading speed in normal counterparts was 257 ± 33 CPM. Age, gender were found to be factors associated with reading speed in children with IXT. After adjusting for the age and gender, we found a significant correlation between the LogTNO and reading speed in IXT group based on a generalized linear model (p = 0.014). These data show that reading speed was slower in school-age children with IXT assessed with the International Reading Speed Texts. When age and gender were adjusted, poor stereo function at near was found to be related with a slower reading speed.