98%
921
2 minutes
20
Addressing the issue of insufficient key feature extraction leading to low recognition rates in existing deep learning-based flow pattern identification methods, this paper proposes a novel flow pattern image recognition model, Enhanced DenseNet with transfer learning (ED-DenseNet). The model enhances the deep feature extraction capability by introducing a multi-branch structure, incorporating an ECA attention mechanism into Dense Blocks and dilated convolutions into Transition Layers to achieve multi-scale feature extraction and refined channel information processing. Considering the limited scale of the experimental dataset, pretrained DenseNet121 weights on ImageNet were transferred to ED-DenseNet using transfer learning. On a gas-liquid two-phase flow image dataset containing Annular, Bubbly, Churn, Dispersed, and Slug flow patterns, ED-DenseNet achieved an overall recognition accuracy of 97.82%, outperforming state-of-the-art models such as Flow-Hilbert-CNN, especially in complex and transitional flow scenarios. Additionally, the model's generalization and robustness were further validated on a nitrogen condensation two-phase flow dataset, demonstrating superior adaptability compared to other methods.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12186962 | PMC |
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0325784 | PLOS |
Commun Biol
September 2025
Division of Neurobiology, Faculty of Biology, Ludwig-Maximilians-Universität München, Planegg - Martinsried, Germany.
The internal resistance of axons to ionic current flow determines action potential conduction velocity. Although mitochondria support axonal function, axons have been modeled as organelle-free cables, and mitochondrial impact on conduction velocity, specifically by increasing internal resistance, remains understudied. We combine computational modeling and electron microscopy of forebrain premotor axons controlling birdsong production.
View Article and Find Full Text PDFClin Breast Cancer
August 2025
Breast Disease Center, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China. Electronic address:
Background: Triple-negative breast cancer (TNBC) carries a substantial risk of recurrence and metastasis, posing significant threats to patients' health and quality of life. Centrosomal protein 55 (CEP55) has been demonstrated to exhibit elevated expression levels in TNBC. However, its molecular regulatory mechanism in TNBC remains unclear.
View Article and Find Full Text PDFMed Eng Phys
October 2025
Ansys Inc., Houston, TX 77094, USA.
Introduction: Benchtop and animal models have traditionally been used to study the propagation of Onyx Liquid Embolic Systems (Onyx) used in the treatment of brain arteriovenous malformations (AVM). However, such models are costly, do not provide sufficient detail to elucidate how variations in Onyx viscosity alter flow dynamics, and rely on some trial-and-error, resulting in elongated timelines for product development.
Objectives: The goal of this study was to leverage Computational Fluid Dynamics (CFD) simulations to predict the behavior of different Onyx formulations.
Biochim Biophys Acta Mol Basis Dis
September 2025
Department of Cardiology, the Second Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang Province, China; State Key Laboratory of Frigid Zone Cardiovascular Disease, Harbin, 150086, Heilongjiang, China; The Key Laboratory of Myocardial Ischemia, Harbin Medical University, Ministry
Background And Aims: Viral myocarditis is an inflammatory pathology of the myocardium that involves innate immune responses, especially those involving neutrophils. However, strategies targeting neutrophils to alleviate inflammation have not achieved complete success. Alpha lipoic acid (ALA), a natural organosulfur compound, has the capacity to modulate immune cell behavior.
View Article and Find Full Text PDFEur Heart J Cardiovasc Imaging
September 2025
Department of Diagnostic and Interventional Radiology, University Hospital Bonn, Venusberg-Campus 1, 53127, Bonn, Germany.
Aims: Fetal circulation undergoes complex changes in congenital heart disease (CHD) that are challenging to assess with fetal echocardiography. This study aimed to assess clinical feasibility and diagnostic value of 4D flow cardiac magnetic resonance (CMR) in fetal CHD.
Methods And Results: Pregnant women in advanced third trimester pregnancy with fetal CHD were prospectively recruited for fetal CMR between 08/2021 and 11/2024.