The Research Of Real-Time In Vivo Detection Of Biological Parameters By The Optical Fiber Raman Spectroscopy

Optical fiber Raman(The optical fiber Raman)is based on Raman Effects,which can realize material composition real-time detection.As it can realize non-contact detection of material composition and optical fiber probe is flexible,lightweight and miniaturized,the optical fiber Raman is gradually applied in biological parameters detection in recent years,especially in the fields of blood testing and tumor tissue detection.However,due to the accuracy requirements of medical testing,there are many factors affecting the measurement results in real-time detection of biological parameters in vivo,among which fiber Raman probe designing and the Raman spectrum data processing of biological parameters have a particularly obvious impact on biomedical testing.This paper built a set of optical fiber Raman system,and compared it with the existing commercial portable Raman system.At the same time in order to improve the real-time detection accuracy in clinical environment,improved the traditional Raman spectra identification algorithm.The nude mouse model experiments and clinical experiments is carried out to verify the accuracy and feasibility of the system.The specific work is as follows:(1)This paper introduces the basic concepts and development trends of fiber Raman detection technology,analyzes the baseline removal algorithm and smoothing algorithm in Raman spectrum preprocessing algorithm.Savitzky-Golay smoothing algorithm and unweighted penalty least square algorithm is selected as the preprocessing algorithm combination.(2)The composition of the fiber Raman system is studied and the difficulties were analyzed.The fiber Raman system was built through the design of system parameters.The accuracy and effectiveness of the fiber Raman system were verified through different alcohol concentration detection experiments and commercial portable Raman spectrometer comparison.(3)The self-built fiber Raman system was used to carry out cancer tissue detection experiments.The feasibility of Raman spectroscopy for detecting oral squamous cell carcinoma was verified by tissue sectioning and micro-Raman device.the feasibility of the fiber Raman system in the detection of oral squamous cell carcinoma was demonstrated compared the spectrum detected by the fiber Raman system with the microRaman device.(4)The bio-parametric Raman spectroscopy data of the fiber Raman system was accumulated by animal model experiments and clinical tissue testing experiments.By comparing the traditional machine learning recognition algorithm with the deep learning recognition algorithm,shown that the depth learning algorithm is able to accurately realize the identification and classification of squamous cell carcinoma tissue.