CN111175261A

CN111175261A - Method for detecting pulmonary tuberculosis disease based on human plasma autofluorescence spectrum

Info

Publication number: CN111175261A
Application number: CN202010036198.4A
Authority: CN
Inventors: 朱建华; 王悦铮; 陈雪融
Original assignee: Sichuan University
Current assignee: Sichuan University
Priority date: 2020-01-14
Filing date: 2020-01-14
Publication date: 2020-05-19

Abstract

The invention provides a method for detecting characteristic peak of auto-fluorescence spectrum of human plasma and diagnosing pulmonary tuberculosis disease by combining multivariate statistical analysis method and application thereof, comprising the following steps: collecting fasting venous blood of a subject, centrifugally treating to separate plasma, measuring an autofluorescence spectrum of the plasma by using a fluorescence spectrophotometer, performing second derivative treatment on the measured fluorescence spectrum, primarily screening patients with tuberculosis according to characteristic peaks of the autofluorescence spectrum, processing fluorescence spectrum data by using a multivariate statistical analysis method to obtain score data corresponding to characteristics of the tuberculosis disease, and comparing the score data with a discrimination parameter or a discrimination equation in a self-established database to diagnose whether the subject suffers from the tuberculosis disease. The detection method provided by the invention has the advantages of convenient sampling, quick detection, high diagnosis sensitivity, specificity and accuracy and low cost, can provide a novel detection method for the diagnosis of the pulmonary tuberculosis disease, and has important social and economic values and clinical application values.

Description

Method for detecting pulmonary tuberculosis disease based on human plasma autofluorescence spectrum

Technical Field

The present invention belongs to the field of medical disease diagnosis. In particular to a method for diagnosing pulmonary tuberculosis diseases by detecting characteristic peaks of a human plasma autofluorescence spectrum and combining a multivariate statistical analysis method and application thereof.

Background

The transmission route of pulmonary tuberculosis is mainly air, can be transmitted by droplets of coughing and sneezing of people, is the first killer of infectious diseases in the world, and is also one of ten causes of death in the world. According to the world health organization, about millions of new tuberculosis cases appear each year, about 1000 thousands of people infect the tuberculosis in 2017, and 130 thousands of people die from the tuberculosis. The world health organization also defines No. 3/24 as the "worldwide tuberculosis preventing and treating day". China is one of the world high-load countries of tuberculosis, and the number of patients with tuberculosis is the second world every year. At present, clinical diagnosis of tuberculosis mainly depends on mycobacterium tuberculosis sputum smear examination, chest X-ray examination, fiberbronchoscopy and the like, and the methods respectively have the defects of long time for obtaining a diagnosis result, invasive examination, X-ray radiation harm to a human body, low diagnosis sensitivity, high examination cost and the like.

Fluorescence spectroscopy can be used to detect changes in biomarkers in body fluids (i.e., blood, urine, or sweat), and can provide a new concept and method for rapid detection of abnormal body fluids and diagnosis of diseases. The value of spectroscopic diagnostics is that spectroscopic diagnostic techniques can directly detect spectroscopic characteristics of some biomolecules in a body fluid of a human, obtain characteristic spectra relating to their structure, and reveal internal information about the molecular structure in the body fluid. Therefore, in the process of malignant transformation (even in early stage of onset), the structure or content of the internal substances of the tissues and organs of the human body are changed to different degrees, and abnormal metabolism is generated, so that the structure or content of the internal substances is changed or reduced, and the change is detected, so that the aim of detecting diseases and even screening early-stage diseases can be fulfilled.

Plasma contains a large amount of biological macromolecules capable of generating fluorescence, such as endogenous fluorescent substances such as Nicotinamide Adenine Dinucleotide (NADH), Flavin Adenine Dinucleotide (FAD), riboflavin, phospholipids, hematoporphyrin and protoporphyrin, and the changes of the structure and the optical properties of the biological macromolecules can reflect the changes of the functions, the shapes and the micro-environment of tissue cells. Therefore, by detecting the autofluorescence spectrum of the plasma, the information of the tissue cytopathy and the metabolic abnormality can be provided, a certain spectroscopy basis is provided for disease diagnosis, and the related diseases can be preliminarily screened.

At present, the fluorescence spectrum technology has been reported to be applied to diagnosis of lung cancer, breast cancer, gastric cancer and leukemia, but the fluorescence spectrum technology is combined with the autofluorescence spectrum of human plasma to identify and diagnose tuberculosis diseases, and no relevant report is found at home and abroad.

Disclosure of Invention

The invention aims to provide a method for detecting a pulmonary tuberculosis disease by using a plasma autofluorescence spectrum and application thereof, aiming at the problems of low diagnosis sensitivity, high examination cost, long time for obtaining a diagnosis result and invasive examination in the current clinical pulmonary tuberculosis diagnosis method. The method comprises the steps of firstly carrying out plasma centrifugation treatment on human blood under an aseptic condition to obtain a plasma solution in a physiological state, measuring by using a fluorescence spectrophotometer to obtain an autofluorescence spectrum of the plasma, detecting characteristic peaks of the autofluorescence spectrum of the human plasma, and diagnosing the tuberculosis by combining multivariate statistical analysis methods (such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA)).

Because the fluorescence spectrum signal of the sample obtained by adopting the fluorescence spectrum technology is higher than the Raman spectrum signal by several orders of magnitude, the problems of low Raman scattering cross section, weak spectrum signal and low detection signal-to-noise ratio are avoided. The method comprises the steps of collecting plasma autofluorescence spectra of a certain number of healthy people and tuberculosis patients, establishing a plasma autofluorescence spectrum and scatter distribution database, preliminarily screening the tuberculosis patients according to second-order derivative characteristic peaks of the plasma autofluorescence spectra of the subjects, calculating by using a multivariate statistical analysis method (such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA)) to obtain score data of the PCA of the subjects, and finally comparing the PCA score data of the subjects with the scatter distribution maps of the healthy people and the tuberculosis patients in the self-established database and a Fisher linear discriminant equation to judge whether the subjects are the tuberculosis patients or the healthy people.

The invention has the advantages of easy sample acquisition, quick and convenient detection, non-invasion, low cost, high diagnosis sensitivity and accuracy and the like, and provides a novel detection method for clinical diagnosis of tuberculosis diseases.

The specific technical scheme of the invention is as follows:

(1) collecting fasting venous blood of a subject by using a vacuum blood collection tube added with an EDTA (ethylene diamine tetraacetic acid) anticoagulant; (2) separating a physiological plasma sample by centrifugation under aseptic conditions; (3) putting a certain amount of plasma into a quartz cuvette and placing the cuvette into a sample cell of a fluorescence spectrophotometer, and operating the fluorescence spectrophotometer to measure and obtain an autofluorescence spectrum of the plasma; (4) carrying out second derivative treatment on the measured autofluorescence spectrum, and carrying out primary screening on the subject according to characteristic peaks of the autofluorescence spectrum; (5) processing autofluorescence spectrum of the subject by multivariate statistical analysis method (such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA)) to obtain score data of PCA; (6) and comparing the PCA score data of the autofluorescence spectrum of the subject with the scattered point distribution diagram and the Fisher linear discriminant equation of the healthy person and the tuberculosis patient in the self-established database, and diagnosing whether the subject is the tuberculosis patient according to different areas of the scattered point distribution diagram and the Fisher linear discriminant line.

The technical scheme of the invention has the following beneficial effects:

1. the invention provides a method for auxiliary screening of tuberculosis diseases, which comprises the steps of obtaining an autofluorescence spectrum of a plasma sample through the measurement of a fluorescence spectrophotometer, carrying out primary screening by using an autofluorescence spectrum characteristic peak obtained by second derivative processing, and distinguishing a tuberculosis patient from a healthy person by combining a principal component analysis method and a linear discrimination method, and has the advantage of high detection speed.

2. The invention can realize the screening and diagnosis of diseases by utilizing blood samples which are very easy to obtain clinically, and solves the problems of complicated sample pretreatment and long time consumption in pathological diagnosis.

3. The blood sample adopted by the invention is a conventional sample for clinical examination, is very easy to obtain, has much less pain and discomfort to the examination object compared with invasive examination, has no special requirements on the physiological signs of the examination object, and is also much simpler to operate.

4. The invention measures the autofluorescence spectrum of blood plasma by using light wave as light source, and compared with clinical X-ray film examination and CT examination, the detection method can avoid the radiation hazard of X-ray to human body, and has high detection safety.

5. The steps of plasma sample preparation, fluorescence spectrum measurement, spectrum principal component analysis and calculation, linear discrimination and the like can be completed within 30 minutes and the result can be obtained. Solves the problems of low diagnosis sensitivity, high examination cost and long time for obtaining diagnosis results in the existing clinical diagnosis method.

Drawings

FIG. 1 is an example of an autofluorescence spectrum of plasma of a healthy person measured according to the present invention (dotted line) and an autofluorescence spectrum of plasma of a person with pulmonary tuberculosis measured according to the present invention (solid line).

FIG. 2 is a second derivative plot (dotted line) of the autofluorescence spectrum of healthy human plasma measured by the present invention and a second derivative plot (solid line) of the autofluorescence spectrum of human plasma measured by the present invention.

FIG. 3 is the PCA score scatter diagram distribution of the autofluorescence spectra of healthy human plasma and human plasma of pulmonary tuberculosis drawn by the characteristic values of PC1 and PC2 and the discrimination result of Fisher linear discrimination equation, in the figure, the hollow dots are the statistical analysis result of the autofluorescence spectra of healthy human plasma, the solid dots are the statistical analysis result of the autofluorescence spectra of human plasma of pulmonary tuberculosis, and the dotted lines are Fisher discrimination lines.

Detailed Description

The present invention will be further illustrated by the following detailed description.

1. Preparation of human plasma samples

The fasting venous blood of the test subject is extracted in the period of 7-8 am in the middle of the morning under aseptic condition, collected into a vacuum blood collection tube containing EDTA anticoagulant, and then placed into a small centrifuge for centrifugal treatment for 10-15 minutes at the rotating speed of 2500 rpm. Approximately 1-2ml of plasma was removed from the centrifuge tube using a pipette gun and placed into a sterile quartz cuvette.

2. Fluorescence spectroscopy detection process

Placing the quartz cuvette containing the plasma sample into a sample cell of a fluorescence spectrophotometer, and operating the fluorescence spectrophotometer to measure and obtain an autofluorescence spectrum (as shown in the attached figure 1) of the plasma, wherein the measurement parameters are as follows: the excitation wavelength is 400nm, the fluorescence scanning wavelength range is 425-675 nm, the scanning wavelength interval is 0.1nm, the widths of the excitation slit and the emission slit are 10 nm, and the measurement speed is medium speed.

3. Second derivative analysis of plasma autofluorescence spectra

Firstly, smoothing the autofluorescence spectrum to reduce noise, then performing second derivative processing on the autofluorescence spectrum of plasma (as shown in figure 2), selecting the characteristic peak with most significant difference, and primarily screening out tuberculosis patients.

4. Principal component analysis and Linear discriminant analysis (PCA-LDA) of autofluorescence spectra of plasma

And 3, establishing a spectral data matrix for the plasma autofluorescence spectral data after smoothing treatment in the step 3, and carrying out standardization treatment on the spectral data matrix. And (3) carrying out PCA analysis on the standardized spectral data matrix to obtain scores (PC values) corresponding to each principal component, then selecting two PCA score values with the most significant difference as a diagnosis basis, namely comparing the PCA score data of the autofluorescence spectrum of the subject with the scatter point distribution diagram of the healthy person and the tuberculosis patient in the self-established database and a Fisher linear discriminant equation (shown in figure 3), and diagnosing whether the subject is the tuberculosis patient according to different areas of the scatter point distribution diagram and the Fisher linear discriminant line, thereby distinguishing the healthy person from the tuberculosis patient.

All data and values of the invention are measured by actual measurement, and the identification conclusion of the blood plasma is objectively given by fluorescence spectrum data and does not depend on subjective judgment of an observer. Plasma sample preparation, measurement and statistical data analysis can be completed within a half hour. The actual measurement shows that the diagnosis sensitivity, specificity and accuracy of the method are high by combining the auto-fluorescence spectrum of the human plasma with the principal component analysis and linear discriminant analysis methods (the figure 3 shows a typical scattered point distribution diagram and a Fisher linear discriminant line obtained by measuring the actual plasma sample, and the result shows that the diagnosis sensitivity, specificity and accuracy of the diagnosis method are 91.7%, 100% and 95%), and the method has wide application prospect in the field of clinical diagnosis of tuberculosis diseases.

The invention discloses a method for diagnosing pulmonary tuberculosis diseases based on detecting characteristic peaks of a human plasma autofluorescence spectrum and combining a multivariate statistical analysis method and application thereof. It should be understood that the method of detecting tuberculosis disease of the present invention is not dependent on the detection equipment, and any fluorescence spectrometer or fluorescence spectrophotometer can be used for detection. It will be appreciated by those skilled in the art that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made to the fluorescence detection parameters, data processing, and statistical analysis methods without departing from the spirit and scope of the invention. Therefore, the detailed description and examples of the invention should not be construed as limiting the scope of the invention. The invention is limited only by the appended claims.

Claims

1. A detection method for detecting the human plasma autofluorescence spectrum and diagnosing the pulmonary tuberculosis by combining the characteristic peak discrimination and multivariate statistical analysis method comprises the following steps:

(1) collecting fasting venous blood of a subject by using a vacuum blood collection tube added with anticoagulant;

(2) separating plasma in physiological state by centrifugation under aseptic conditions;

(3) measuring an autofluorescence spectrum of the plasma sample;

(4) carrying out second derivative treatment on the measured fluorescence spectrum, and primarily screening tuberculosis patients according to characteristic peaks of the fluorescence spectrum;

(5) processing autofluorescence spectrum data of a subject by using a multivariate statistical analysis method to obtain score data corresponding to the characteristics of the pulmonary tuberculosis disease;

(6) and comparing the score data of the autofluorescence spectrum of the subject with the discrimination parameters or discrimination equations of healthy people and patients with tuberculosis in the self-built database, and diagnosing whether the subject is the patient with tuberculosis according to the position of the parameters or different areas of the discrimination equations.

2. The detection method according to claim 1, characterized in that: and (3) carrying out autofluorescence spectroscopy on the plasma in the step (3), wherein the key detection range is 425-675 nm.

3. The detection method according to claim 1, characterized in that: and (4) performing second derivative processing on the fluorescence spectrum, so that a spectral characteristic peak corresponding to the fluorescent substance can be distinguished, and the accuracy of primary screening is improved.

4. The detection method according to claim 1, characterized in that: and (5) processing the autofluorescence spectrum data of the testee by using a multivariate statistical analysis method, so as to specify score data corresponding to the characteristics of the tuberculosis disease and eliminate the influence of irrelevant or secondary factors on diagnosis.

5. The detection method according to claim 1, characterized in that: and (6) measuring a certain amount of plasma autofluorescence spectra of tuberculosis patients and normal healthy people, and analyzing and processing the spectral data by combining a multivariate statistical analysis method to obtain the self-established database.