Combined Mutual Learning Net for Raman Spectral Microbial Strain Identification.

Infectious diseases pose a significant threat to global health, yet traditional microbiological identification methods suffer from drawbacks, such as high costs and long processing times. Raman spectroscopy, a label-free and noninvasive technique, provides rich chemical information and has tremendous potential in fast microbial diagnoses. Here, we propose a novel Combined Mutual Learning Net that precisely identifies microbial subspecies. It demonstrated an average identification accuracy of 87.96% in an open-access data set with thirty microbial strains, representing a 5.76% improvement. 50% of the microbial subspecies accuracies were elevated by 1% to 46%, especially for E. coli 2 improved from 31% to 77%. Furthermore, it achieved a remarkable subspecies accuracy of 92.4% in the custom-built fiber-optical tweezers Raman spectroscopy system, which collects Raman spectra at a single-cell level. This advancement demonstrates the effectiveness of this method in microbial subspecies identification, offering a promising solution for microbiology diagnosis.

Spatially Modulated Fiber Speckle for High-Sensitivity Refractive Index Sensing.

A fiber speckle sensor (FSS) based on a tapered multimode fiber (TMMF) has been developed to measure liquid analyte refractive index (RI) in this work. By the lateral and axial offset of input light into TMMF, several high-order modes are excited in TMMF, and the speckle pattern is spatially modulated, which affects an asymmetrical speckle pattern with a random intensity distribution at the output of TMMF. When the TMMF is immersed in the liquid analyte with RI variation, it influences the guided modes, as well as the mode interference, in TMMF. A digital image correlations method with zero-mean normalized cross-correlation coefficient is explored to digitize the speckle image differences, analyzing the RI variation. It is found that the lateral- and axial-offsets-induced speckle sensor can enhance the RI sensitivity from 6.41 to 19.52 RIU-1 compared to the one without offset. The developed TMMF speckle sensor shows an RI resolution of 5.84 × 10-5 over a linear response range of 1.3164 to 1.3588 at 1550 nm. The experimental results indicate the FSS provides a simple, efficient, and economic approach to RI sensing, which exhibits an enormous potential in the image-based ocean-sensing application.

Magnetic microspheres enhanced peanut structure cascaded lasso shaped fiber laser biosensor for cancer marker-CEACAM5 detection in serum.

The detection of trace cancer markers in body fluids such as blood/serum is crucial for cancer diseases screening and treatment, which requires high sensitivity and specificity of biosensors. In this study, a peanut structure cascaded lasso (PSCL) shaped fiber sensing probe based on fiber laser demodulation method was proposed to specifically detect the carcinoembryonic antigen related cell adhesion molecules 5 (CEACAM5) protein in serum. Thanks for the narrow linewidth and high signal-to-noise ratio (SNR) of the laser spectrum, it is easier to distinguish small spectral changes than interference spectrum. Adding the antibody modified magnetic microspheres (MMS) to form the sandwich structure of "antibody-antigen-antibody-MMS", and amplified the response caused by biomolecular binding. The limit of detection (LOD) for CEACAM5 in buffer could reach 0.11 ng/mL. Considering the common threshold of 5 ng/mL for CEA during medical screening and the cut off limit of 2.5 ng/mL for some kits, the LOD of proposed biosensor meets the actual needs. Human serum samples from a hospital were used to validate the real sensing capability of proposed biosensor. The deviation between the measured value in various serum samples and the clinical value ranged from 1.9 to 9.8 %. This sensing scheme holds great potential to serve as a point of care testing (POCT) device and extend to more biosensing applications.

RSPSSL: A novel high-fidelity Raman spectral preprocessing scheme to enhance biomedical applications and chemical resolution visualization.

Raman spectroscopy has tremendous potential for material analysis with its molecular fingerprinting capability in many branches of science and technology. It is also an emerging omics technique for metabolic profiling to shape precision medicine. However, precisely attributing vibration peaks coupled with specific environmental, instrumental, and specimen noise is problematic. Intelligent Raman spectral preprocessing to remove statistical bias noise and sample-related errors should provide a powerful tool for valuable information extraction. Here, we propose a novel Raman spectral preprocessing scheme based on self-supervised learning (RSPSSL) with high capacity and spectral fidelity. It can preprocess arbitrary Raman spectra without further training at a speed of ~1 900 spectra per second without human interference. The experimental data preprocessing trial demonstrated its excellent capacity and signal fidelity with an 88% reduction in root mean square error and a 60% reduction in infinite norm ([Formula: see text]) compared to established techniques. With this advantage, it remarkably enhanced various biomedical applications with a 400% accuracy elevation (ΔAUC) in cancer diagnosis, an average 38% (few-shot) and 242% accuracy improvement in paraquat concentration prediction, and unsealed the chemical resolution of biomedical hyperspectral images, especially in the spectral fingerprint region. It precisely preprocessed various Raman spectra from different spectroscopy devices, laboratories, and diverse applications. This scheme will enable biomedical mechanism screening with the label-free volumetric molecular imaging tool on organism and disease metabolomics profiling with a scenario of high throughput, cross-device, various analyte complexity, and diverse applications.