神经网络推进了发光寿命的极限 纳米传感
Liyan Ming1,2, Irene Zabala-Gutierrez3, Paloma Rodríguez-Sevilla1
1Nanomaterials for Bioimaging Group (nanoBIG), Departamento de Física de Materiales, Facultad de Ciencias, Autonomous University of Madrid, Madrid, 28049, Spain.
Advanced materials (Deerfield Beach, Fla.)
|October 3, 2023
概括
一个新的U形卷积神经网络 (U-NET) 增强了发光寿命传感,在低信号噪声比 (SNR) 条件下提高了准确性. 这种机器学习方法可以为生物学和微流体学中的应用进行精确的热量测量.
科学领域:
- 光电学是指光电子产品.
- 生物医学传感传感器
- 机器学习应用 机器学习应用
背景情况:
- 基于发光寿命的传感提供了非侵入性监测,但受到低信号噪声比 (SNR) 的限制.
- 短暂的曝光时间和不透明生物组织的光散射带来了挑战.
- 现有的方法,如衰变曲线拟合,与糟糕的SNR数据作斗争.
研究的目的:
- 通过U型卷积神经网络 (U-NET) 克服SNR在发光寿命感应方面的局限性.
- 提高发光寿命估计的精度和一致性,特别是对于温度计.
- 在具有挑战性的实验条件下证明U-NET的有效性.
主要方法:
- 应用U型卷积神经网络 (U-NET) 进行信号处理.
- 开发Ag2S纳米温度计用于发光寿命温度计.
- 在极低的SNR条件下进行实验验证,包括自由落下和悬浮滴.
主要成果:
- 与传统方法相比,U-NET显著提高了发光寿命估计的准确性和一致性.
- 使用Ag2S纳米温度计,即使在极低的SNR,也可以实现精确的热读数.
- 在自由落下的水滴和透过不透明介质中证明了成功的热监测.
结论:
- U-NET有效地解决了SNR在发光寿命传感方面的局限性.
- 这种基于机器学习的方法扩大了在体内和微流体系统中发光传感的适用性.
- 该研究鼓励进一步将机器学习纳入光学传感技术.
更多相关视频
08:31Luminescence Resonance Energy Transfer to Study Conformational Changes in Membrane Proteins Expressed in Mammalian Cells
Published on: September 16, 2014
12.1K
10:41Visualizing Protein Kinase A Activity In Head-fixed Behaving Mice Using In Vivo Two-photon Fluorescence Lifetime Imaging Microscopy
Published on: June 7, 2019
8.5K
相关概念视频
01:23Photoluminescence: Fluorescence and Phosphorescence
2.1K
Photoluminescence is a process where a molecule absorbs light energy and re-emits it in the form of light. This phenomenon occurs when a substance absorbs photons, promoting its electrons to higher energy level excited states, followed by a relaxation process in which the electrons return to their original ground state energy levels and emit light. Photoluminescence is widely observed in various materials, including semiconductors, and organic and inorganic compounds.
A pair of electrons in a...
A pair of electrons in a...
2.1K
01:14Photoluminescence: Applications
427
Photoluminescence offers a wide range of applications due to its inherent sensitivity and selectivity. This technique allows for both direct and indirect analyses of the analyte. Direct quantitative analysis is possible when the analyte exhibits a favorable quantum yield for fluorescence or phosphorescence. However, an indirect analysis may be feasible if the analyte is not fluorescent or phosphorescent, or if the quantum yield is unfavorable. Indirect methods include reacting the analyte with...
427