A generative adversarial neural network with multi-attention feature extraction for fundus lesion segmentation.

PURPOSE: Fundus lesion segmentation determines the location and size of diabetes retinopathy in fundus image, which assists doctors in developing the best eye treatment plan. However, owing to the scattered distribution and the similarity of lesions, it is extremely difficult to extract representative lesions feature and accurately segment lesions area. METHODS: To solve the thorny problem, a generative adversarial network with multi-attention feature extraction is developed to segment diabetic retinopathy region. The main contributions are as follows: (1) An improved residual U-Net network combining with self-attention mechanism is designed as generative network to fully extract local and global feature of lesions while reducing the loss of key feature information. Considering the correlation between the same lesions feature of different samples, external attention mechanism is introduced in the residual U-Net network to focus on the relevant features of the same lesions in different samples throughout the entire dataset. (2) A discriminative network based on the PatchGAN structure is designed to further enhance the segmentation ability of generation network by discriminating between true and false samples. RESULTS: The proposed network is evaluated on the public dataset IDRiD, which achieved the Dice correlation coefficients of 75.7%, 76.53%, 50.06%, and 45.89% for EX, SE, MA, and HE, respectively. CONCLUSION: The experimental results show the generative adversarial neural network qualified for accurate segmentation of diabetic retinopathy from fundus image well.

Potential impact of bacteria on the transport of PFAS in porous media.

The transport and fate of per- and poly-fluoroalkyl substances (PFAS) in soil and groundwater is a topic of critical concern. A number of factors and processes may influence the transport and fate of PFAS in porous media. One factor that has received minimal attention to date is the impact of bacteria on the retention and transport of PFAS, which is the focus of this current study. The first part of this work comprised a critical review of prior studies to delineate observed PFAS-bacteria interactions and to summarize the mechanisms of PFAS sorption and retention by bacteria. Retention of PFAS by bacteria can occur through sorption onto cell surfaces and/or by incorporation into the cell interior. Factors such as the molecular structure of PFAS, solution chemistry, and bacterial species can affect the magnitude of PFAS sorption. The influence of bacteria on the retention and transport of PFAS was investigated in the second part of the study with a series of batch and miscible-displacement experiments. Batch experiments were conducted using Gram-negative Pseudomonas aeruginosa and Gram-positive Bacillus subtilis to quantify the sorption of perfluorooctane sulfonic acid (PFOS). The results indicated that both bacteria showed strong adsorption of PFOS, with no significant difference in adsorption capacity. Miscible-displacement experiments were then conducted to examine the retention and transport of PFOS in both untreated sand and sand inoculated with Pseudomonas aeruginosa or Bacillus subtilis for 1 and 3 days. The transport of PFOS exhibited greater retardation for the experiments with inoculated sand. Furthermore, the enhanced sorption was greater for the 3-day inoculation compared to the 1-day, indicating that biomass is an important factor affecting PFOS transport. A mathematical model representing transport with nonlinear and rate-limited sorption successfully simulated the observed PFOS transport. This study highlights the need for future studies to evaluate the effect of bacteria on the transport of PFAS in soil and groundwater.

Multi-Modal Feature Fusion-Based Multi-Branch Classification Network for Pulmonary Nodule Malignancy Suspiciousness Diagnosis.

Detecting and identifying malignant nodules on chest computed tomography (CT) plays an important role in the early diagnosis and timely treatment of lung cancer, which can greatly reduce the number of deaths worldwide. In view of the existing methods in pulmonary nodule diagnosis, the importance of clinical radiological structured data (laboratory examination, radiological data) is ignored for the accuracy judgment of patients' condition. Hence, a multi-modal fusion multi-branch classification network is constructed to detect and classify pulmonary nodules in this work: (1) Radiological data of pulmonary nodules are used to construct structured features of length 9. (2) A multi-branch fusion-based effective attention mechanism network is designed for 3D CT Patch unstructured data, which uses 3D ECA-ResNet to dynamically adjust the extracted features. In addition, feature maps with different receptive fields from multi-layer are fully fused to obtain representative multi-scale unstructured features. (3) Multi-modal feature fusion of structured data and unstructured data is performed to distinguish benign and malignant nodules. Numerous experimental results show that this advanced network can effectively classify the benign and malignant pulmonary nodules for clinical diagnosis, which achieves the highest accuracy (94.89%), sensitivity (94.91%), and F1-score (94.65%) and lowest false positive rate (5.55%).

Impact of a Hydrocarbon Surfactant on the Retention and Transport of Perfluorooctanoic Acid in Saturated and Unsaturated Porous Media.

The transport and retention behavior of perfluorooctanoic acid (PFOA) in the presence of a hydrocarbon surfactant under saturated and unsaturated conditions was investigated. Miscible-displacement transport experiments were conducted at different PFOA and sodium dodecyl sulfate (SDS) input ratios to determine the impact of SDS on PFOA adsorption at solid-water and air-water interfaces. A numerical flow and transport model was employed to simulate the experiments. The PFOA breakthrough curves for unsaturated conditions exhibited greater retardation compared to those for saturated conditions in all cases, owing to air-water interfacial adsorption. The retardation factor for PFOA with a low concentration of SDS (PFOA-SDS ratio of 10:1) was similar to that for PFOA without SDS under unsaturated conditions. Conversely, retardation was greater in the presence of higher levels of SDS (1:1 and 1:10) with retardation factors increasing from 2.4 to 2.9 and 3.6 under unsaturated conditions due to enhanced adsorption at the solid-water and air-water interfaces. The low concentration of SDS had no measurable impact on PFOA air-water interfacial adsorption coefficients (Kia) determined from the transport experiments. The presence of SDS at the higher PFOA-SDS concentration ratios increased the surface activity of PFOA, with transport-determined Kia values increased by 27 and 139%, respectively. The model provided very good independently predicted simulations of the measured breakthrough curves and showed that PFOA and SDS experienced various degrees of differential transport during the experiments. These results have implications for the characterization and modeling of poly-fluoroalkyl substances (PFAS) migration potential at sites wherein PFAS and hydrocarbon surfactants co-occur.