PARTICULATE SCIENCE AND TECHNOLOGY
·
2023
·
IF 2.5
Weijie Mao,Biao Ma,Qianqi Zhang,Xu Hua
Phase change material (PCM) mixed in soil can effectively adjust the subgrade temperature. In order to analyze the temperature field change of soil containing PCM more accurately from the microscopic point of view, it is a key problem to construct a refined geometric model. This paper scans the fracture plane of soil containing PCM through the stereoscopic microscope and prepares the label image corresponding to the sample image. Then a U-Net semantic segmentation model is established, which can accurately recognize PCM area and soil area. Finally, a geometric model of soil containing PCM that is consistent with reality is established, and its heat transfer characteristics are analyzed by finite element method. The accuracy evaluation of the segmentation model and the visualization of the recognition results show that the evaluation indexes of U-Net are the best. P, R, F1, and MIoU of U-Net are 0.889, 0.918, 0.903, and 0.901, respectively. The refined model established in this paper can not only observe the overall temperature regulation effect of the material but also accurately describe the internal temperature distribution details of soil containing PCM during heat transfer. Through numerical simulation, it is found that the maximum temperature of mixture decreases by 1.6 °C and the minimum temperature of mixture increases by 2.3 °C.
研究领域:
土木工程
·
计算机视觉
·
热力学
✓ 已验证
产品关联
DOI
高影响因子
FOOD AND CHEMICAL TOXICOLOGY
·
2025
·
IF 3.5
Zhiyuan Ning,Yingming Zhang,Shikun Zhang,Xianfeng Lin,Lixin Kang,Nuo Duan,Zhouping Wang,Shijia Wu
Acrylamide (AA), a food hazard generated during thermal processing, poses significant safety risks due to its toxicity. Conventional methods for AA toxicology are time-consuming and inadequate for analyzing cellular morphology. This study developed a novel approach combining deep learning models (U-Net and ResNet34) with cell fluorescence imaging. U-Net was used for cell segmentation, generating a single-cell dataset, while ResNet34 trained the dataset over 200 epochs, achieving an 80% validation accuracy. This method predicts AA concentration ranges by matching cell fluorescence features with the dataset and analyzes cellular phenotypic changes under AA exposure using k-means clustering and CellProfiler. The approach overcomes the limitations of traditional toxicological methods, offering a direct link between cell phenotypes and hazard toxicology. It provides a high-throughput, accurate solution to evaluate AA toxicology and refines the understanding of its cellular impacts.
研究领域:
生物信息学
·
毒理学
✓ 已验证
产品关联
DOI
PubMed
高影响因子
MICROCHEMICAL JOURNAL
·
2025
·
IF 5.1
Shunshun Bao,Shaoze Zhi,Xueying Yang,Yinfeng Wang,Zhengchun Yang,Jun Liu,Peng Pan,Yangyang Qi,Jie He
This study presents an electrochemiluminescence (ECL) sensing platform based on graphene quantum dots/silver nanoparticles (GQDs/AgNPs) composites, combined with a lightweight deep learning model within a portable detection system to achieve highly sensitive point-of-care monitoring of methimazole (MMI). By utilizing the synergistic effect of GQDs interface modification and AgNPs localized surface plasmon resonance (LSPR), the sensor significantly enhances the ECL signal intensity of tris(2,2′-bipyridyl)ruthenium(II) (Ru(bpy) 3 2+ ) on paper-based electrodes, while exhibiting excellent stability. In addition, for the problem of uneven luminescent spot distribution and edge blurring in ECL images, we use an improved U-Net neural network model to achieve accurate extraction of the center bright luminescent spot through sub-pixel level segmentation and background noise suppression. The detection system utilizes a Raspberry Pi core with a 3D printing module to significantly reduce the time and cost per test, and achieves a detection limit of 0.327 nM (linear range: 10 −9 – 10 −3 M). Through the design strategy of “material-algorithm-system”, this study provides a highly sensitive, stable and practical solution for monitoring drug concentrations in the urine of hyperthyroid patients, which promotes the implementation of precision medicine in primary care.
研究领域:
分析化学
·
纳米技术
·
生物医学工程
·
电化学
·
机器学习
·
药物分析
✓ 已验证
产品关联
DOI
高影响因子
ADVANCED FUNCTIONAL MATERIALS
·
2025
·
IF 19
Xiaobin Ren,Zhirong Liu,Song Wang,Zelin Cao,Kaikai Gao,Guangdong Zhou,Lu Ye,Bai Sun
Despite the progress in machine vision, existing image recognition systems still operate in isolation from the physical world, lacking dynamic temporal encoding, in-memory processing, and intrinsic interpretability. Here, a memristive vision architecture that integrates multimodal sensing, physical memory, and medical inference within a unified device-level framework is developed. Specifically, spatial gradients from retinal fundus images are encoded into Ag/Indigo/FTO memristor array via pulse modulation, generating conductance maps (“G maps”) that serve both as nonvolatile memory and as neuromorphic inputs for lesion segmentation via lightweight U-Net models. Beyond static vision, the same memristor exhibits strong temporal sensitivity, enabling biologically realistic nociceptive encoding and Morse code recognition. These results demonstrate a cross-task, cross-scale system that merges physical encoding with machine intelligence, bridging the gap between material dynamics and high-level perception. Therefore, this paradigm offers a scalable, interpretable hardware foundation for next-generation bio-perceptive computing and intelligent diagnostics.
研究领域:
神经形态工程
·
生物医学成像
·
人工智能
·
非易失性存储器
·
可穿戴传感器
·
医学诊断
✓ 已验证
产品关联
DOI
高影响因子
危险品化学品经营许可证（带存储）