Hybrid deep learning and optimal graph search method for optical coherence tomography layer segmentation in diseases affecting the optic nerve.

Accurate segmentation of retinal layers in optical coherence tomography (OCT) images is critical for assessing diseases that affect the optic nerve, but existing automated algorithms often fail when pathology causes irregular layer topology, such as extreme thinning of the ganglion cell-inner plexiform layer (GCIPL). Deep LOGISMOS, a hybrid approach that combines the strengths of deep learning and 3D graph search to overcome their limitations, was developed to improve the accuracy, robustness and generalizability of retinal layer segmentation. The method was trained on 124 OCT volumes from both eyes of 31 non-arteritic anterior ischemic optic neuropathy (NAION) patients and tested on three cross-sectional datasets with available reference tracings: Test-NAION (40 volumes from both eyes of 20 NAION subjects), Test-G (29 volumes from 29 glaucoma subjects/eyes), and Test-JHU (35 volumes from 21 multiple sclerosis and 14 control subjects/eyes) and one longitudinal dataset without reference tracings: Test-G-L (155 volumes from 15 glaucoma patients/eyes). In the three test datasets with reference tracings (Test-NAION, Test-G, and Test-JHU), Deep LOGISMOS achieved very high Dice similarity coefficients (%) on GCIPL: 89.97±3.59, 90.63±2.56, and 94.06±1.76, respectively. In the same context, Deep LOGISMOS outperformed the Iowa reference algorithms by improving the Dice score by 17.5, 5.4, and 7.5, and also surpassed the deep learning framework nnU-Net with improvements of 4.4, 3.7, and 1.0. For the 15 severe glaucoma eyes with marked GCIPL thinning (Test-G-L), it demonstrated reliable regional GCIPL thickness measurement over five years. The proposed Deep LOGISMOS approach has potential to enhance precise quantification of retinal structures, aiding diagnosis and treatment management of optic nerve diseases.

Examining spatiotemporal crowdsensing and caching for population-dynamic OTT content delivery.

This study proposes a novel spatiotemporal crowdsensing and caching (SCAC) framework to address the surging demands of urban wireless network traffic. In the context of rampant urbanization and ubiquitous digitization in cities, effective data traffic management is crucial for maintaining a dynamic urban ecosystem. Leveraging user mobility patterns and content preferences, this study formulates an offloading policy to alleviate congestion across urban areas. Our approach uses an AI-based method at the cell level, providing a practical and scalable solution that can be readily adapted to bustling metropolitan areas. The implementation of our model demonstrated its effectiveness in reflecting real-world urban dynamics, resulting in significant reductions in peak-hour traffic and robust performance across diverse urban settings. The deployment strategy initiates from densely populated transportation hubs, gradually expanding to broader urban areas. This systematic expansion adheres to a policy framework that emphasizes data privacy and sustainable urban development, ensuring alignment with societal needs and regulatory frameworks. By addressing technological efficacy and societal impact, this study enhances the understanding of urban wireless traffic management. It offers mobile network operators, policymakers, and urban planners a comprehensive strategy to harness the potential of spatiotemporal technology, thereby ensuring that cities remain dynamic, efficient, and well-prepared for the future of digital connectivity.

OspF directly attenuates the activity of extracellular signal-regulated kinase during invasion by Shigella flexneri in human dendritic cells.

Shigella spp., Gram-negative pathogenic bacteria, deliver various effector molecules into the host cell cytoplasm through their type III secretion system to facilitate their invasive process and control the host innate immune responses. Although the function of these effectors is well characterized in epithelial cells during Shigella infection, it has not been elucidated in the dendritic cell (DC), a major antigen presenting cell playing an important role in the initiation of immune responses. In this study, we showed that an invasive Shigella strain (M90T), but not its non-invasive counterpart strain (BS176) induced apoptotic cell death in the human monocyte-derived DCs. Confocal microscopy using a lysosome-associated membrane protein 2 specific antibody demonstrated that the M90T escaped from phagosomes 2h post-DC invasion while BS176 remained in the phagosome. Furthermore, Shigella expressed outer Shigella protein F (OspF), one of the effector proteins that are released through type III secretion system during the invasion, at non-secretion state and further up-regulated OspF expression in the cytoplasm of DC during the invasion. Interestingly, in the host cell, OspF could directly bind to the extracellular signal-regulated kinase (Erk) 1/2 and dephosphorylate phospho-Erk. These results suggest that induction of OspF is enhanced during Shigella invasion of DCs and decreases the phosphorylation level of Erk1/2, which could be at least partially involved in the apoptotic death of DC, eventually resulting in the down-regulation of the host immune response.