- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Single-cell classification of foodborne pathogens using hyperspectral microscope imaging coupled with deep learning frameworks
摘要： A high-throughput hyperspectral microscope imaging (HMI) technology with hybrid deep learning (DL) framework defined as “Fusion-Net” was proposed for rapid classification of foodborne bacteria at single-cell level. HMI technology is useful in single-cell characterization, providing spatial, spectral and combined spatial-spectral profiles with high resolution. However, direct analysis of these high-dimensional HMI data is challenging. In this work, HMI data were decomposed into three parts as morphological features, intensity images, and spectral profiles. Multiple advanced DL frameworks including long-short term memory (LSTM) network, deep residual network (ResNet), and one-dimensional convolutional neural network (1D-CNN) were utilized, achieving classification accuracies of 92.2 %, 93.8 %, and 96.2 %, respectively. Taking advantage of fusion strategy, individual DL framework was stacked to form “Fusion-Net” that processed these features simultaneously with improved classification accuracy of up to 98.4 %. Our study demonstrated the ability of DL frameworks to assist HMI technology in single-cell classification as a diagnostic tool for rapid detection of foodborne pathogens.
关键词： Machine learning,Hyperspectral microscopy,Data fusion,Rapid detection,Foodborne pathogen
更新于2025-09-23 15:19:57
-
Classification of foodborne bacteria using hyperspectral microscope imaging technology coupled with convolutional neural networksa??
摘要： Foodborne pathogens have become ongoing threats in the food industry, whereas their rapid detection and classification at an early stage are still challenging. To address early and rapid detection, hyperspectral microscope imaging (HMI) technology combined with convolutional neural networks (CNN) was proposed to classify foodborne bacterial species at the cellular level. HMI technology can simultaneously obtain both spatial and spectral information of different live bacterial cells, while two CNN frameworks, U-Net and one-dimensional CNN (1D-CNN), were employed to accelerate the data analysis process. U-Net was used for automating cellular regions of interest (ROI) segmentation, which generated accurate cell-ROI masks in a shorter timeframe than the conventional Otsu or Watershed methods. The 1D-CNN was employed for classifying the spectral profiles extracted from cell-ROI and resulted in a higher accuracy (90%) than k-nearest neighbor (81%) and support vector machine (81%). Overall, the CNN-assisted HMI technology showed potential for foodborne bacteria detection.
关键词： Convolutional neural network,Machine learning,Hyperspectral microscopy,Food safety,Foodborne pathogen,Rapid classification
更新于2025-09-23 15:19:57
-
[IEEE 2019 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC) - Munich, Germany (2019.6.23-2019.6.27)] 2019 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC) - A Hyperspectral Microscope Based on a Birefringent Ultrastable Common-Path Interferometer
摘要： Spectral microscopy is a method to acquire the spectrum for each point in the image of a sample. The most straightforward technique uses spectral filters to collect a sequence of images at a discrete number of spectral bands. A more complete spectral characterization is hyperspectral microscopy, which acquires the whole continuous spectrum of each point of the image. A powerful approach to this aim is Fourier-transform (FT) spectrometry [1, 2], in which an optical waveform is split by an interferometer in two delayed replicas, whose interference pattern is measured by a detector as a function of their delay. The FT of the resulting interferogram yields the continuous intensity spectrum of the waveform. The FT approach is able to retrieve in parallel the spectra for all pixels in a scene and is hence suited for wide-field microscopy, but it requires controlling the delay with sub-cycle precision, which is very difficult to achieve with Michelson and Mach-Zehnder interferometers. Here we introduce a hyperspectral microscope based on the FT approach and using a compact, highly stable common-path birefringent interferometer, a version of the Translating-Wedge-based Identical pulses eNcoding System (TWINS) [3, 4]. Figure 1(a) shows the schematic setup of the microscope. Light is collected by an infinity-corrected objective, it propagates in the interferometer and it is imaged on the 2D detector (14-bits, silicon monochrome CMOS camera) by a tube lens. The component P1 polarizes the input light at 45°. A and B are (cid:302)BBO-birefringent blocks with crossed optical axes; block A is shaped in the form of two wedges, so that its total thickness can be changed by translating one of the wedges with a motorized stage. During propagation, the ordinary and extraordinary light projections accumulate a relative delay ranging from positive to negative values according to the relative thickness of A and B. P2 projects the replicas to the same polarization (45°), enabling interference. The spectral resolution of the interferometer is inversely proportional to the adjustable total phase delay. The largest position scan of our interferometer setup introduces a delay of ±250 fs at (cid:540) = 600 nm and corresponds to spectral resolution of 3 THz (~4 nm).
关键词： Fourier-transform spectrometry,TWINS,hyperspectral microscopy,birefringent interferometer
更新于2025-09-12 10:27:22