Esophageal cancer early screening system and screening method using fluorescently-labeled exosome protein

Abstract

The invention relates to the technical field of esophageal cancer early screening, in particular to an esophageal cancer early screening system and screening method using fluorescence-labeled exosome protein. Comprising the following steps: acquiring and pre-treating a patient body fluid sample to obtain a pre-treated body fluid sample; performing exosome separation on the pretreated body fluid sample to obtain an exosome suspension; carrying out specific labeling on the exosome suspension by using a fluorescent nanoprobe to obtain a fluorescently labeled exosome solution; performing fluorescence signal detection on the exosome solution subjected to fluorescence labeling, and obtaining fluorescence intensity data and multispectral fluorescence image data based on a fluorescence signal; based on the fluorescence intensity data and the multispectral fluorescence image data, in combination with clinical data of a patient, intelligent comprehensive analysis is carried out, and a screening result is obtained. The technical problems that in early screening of esophageal cancer, the screening sensitivity and accuracy are low, and data analysis is inaccurate are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of early screening for esophageal cancer, and particularly to an early screening system and method for esophageal cancer that utilize fluorescently labeled exosomal proteins. Background Art

[0002] Esophageal cancer is one of the malignant tumors with a relatively high fatality rate globally. Especially in developing countries, the incidence and mortality of esophageal cancer are very high. Due to the lack of obvious early symptoms, many patients are already in the advanced stage at the time of diagnosis. Therefore, early screening and early diagnosis of esophageal cancer are of crucial importance. Currently, the conventional diagnostic methods for esophageal cancer mainly rely on endoscopic examination, imaging detection, and tissue biopsy. However, these methods not only have invasiveness but also are costly, making it difficult to be widely applied in population screening. In addition, endoscopic examination has high technical requirements for operating doctors and there are certain misdiagnosis and missed diagnosis rates.

[0003] As small vesicles secreted by cells, exosomes have gradually become an important object of liquid biopsy in recent years due to their rich biomarkers, especially their characteristics in tumor-related proteins. Exosomes can be extracted from body fluid samples such as blood, saliva, and urine, providing a non-invasive method for cancer detection. By detecting specific tumor markers on the surface of exosomes, important biomarkers can be provided for the early screening of esophageal cancer. However, there have always been technical challenges in the efficiency and specificity of exosome isolation and purification. Existing exosome isolation methods often have problems such as insufficient isolation purity and complex sample processing.

[0004] At the same time, the above existing technologies also have technical problems in the early screening of esophageal cancer, such as low sensitivity and accuracy of screening, and inaccurate data analysis. Summary of the Invention

[0005] The present invention provides an early screening system and method for esophageal cancer that utilize fluorescently labeled exosomal proteins to solve the technical problems of low sensitivity and accuracy of screening and inaccurate data analysis in the early screening of esophageal cancer.

[0006] An early screening system and method for esophageal cancer that utilize fluorescently labeled exosomal proteins of the present invention specifically include the following technical solutions:

[0007] An early screening method for esophageal cancer that utilizes fluorescently labeled exosomal proteins includes the following steps:

[0008] S1. Obtain and preprocess a patient's body fluid sample to obtain a preprocessed body fluid sample; isolate exosomes from the preprocessed body fluid sample to obtain an exosome suspension; specifically label the exosome suspension with a fluorescent nanoprobe to obtain a fluorescently labeled exosome solution;

[0009] S2. Detect the fluorescence signal of the exosome solution after fluorescence labeling. Based on the fluorescence signal, obtain the fluorescence intensity data and multi-spectral fluorescence image data. Based on the fluorescence intensity data and multi-spectral fluorescence image data, combined with the patient's clinical data, perform intelligent comprehensive analysis to obtain the screening result.

[0010] Preferably, the S1 specifically includes:

[0011] Introduce the preprocessed body fluid sample into the microfluidic chip. The microfluidic chip is designed with channels. Use acoustic sorting technology to separate according to particles of different sizes and densities. After separation, guide the exosomes to specific channels, separate the target exosomes smaller than extracellular vesicles, remove the extracellular vesicles of larger particles, and isolate the exosomes to obtain an exosome suspension.

[0012] Preferably, the S1 specifically includes:

[0013] After exosome separation, enter the labeling process of fluorescent nanoprobes to efficiently label specific tumor-related proteins on the exosome surface. At the end of the fluorescent probe labeling process, use centrifugation and washing steps to remove the unbound fluorescent labeling probes, and retain the fluorescent labeling bound to the exosome surface to obtain a fluorescent-labeled exosome solution.

[0014] Preferably, the S2 specifically includes:

[0015] Based on the fluorescence intensity data and multi-spectral fluorescence image data, combined with the patient's clinical data, use the intelligent comprehensive esophageal cancer screening analysis algorithm for intelligent comprehensive analysis. The intelligent comprehensive esophageal cancer screening analysis algorithm is based on the comprehensive analysis of the fluorescence intensity data, multi-spectral fluorescence image data and patient's clinical data. Through dynamic deformation, multi-scale filtering, high-dimensional expansion, and high-order non-linear interaction processing, obtain the analysis result, calculate the esophageal cancer risk score based on the analysis result, and generate the screening result.

[0016] Preferably, the S2 specifically includes:

[0017] During the implementation of the intelligent comprehensive esophageal cancer screening analysis algorithm, normalize the fluorescence intensity data, multi-spectral fluorescence image data and patient's clinical data to obtain the normalized fluorescence intensity data, normalized multi-spectral fluorescence image data and normalized patient's clinical data.

[0018] Preferably, the S2 specifically includes:

[0019] In the implementation process of the intelligent comprehensive esophageal cancer screening analysis algorithm, dynamic deformation processing is performed on the normalized fluorescence intensity data to obtain the characteristics of the fluorescence intensity data after dynamic deformation processing; for the normalized multispectral fluorescence image data, a multi-scale filtering method is used to extract features from different scales to obtain the multispectral fluorescence image data after multi-scale filtering; for the normalized patient clinical data, high-dimensional expansion is performed to increase the complexity of the features to obtain the characteristics of the patient clinical data after high-dimensional expansion.

[0020] Preferably, the S2 specifically includes:

[0021] In the implementation process of the intelligent comprehensive esophageal cancer screening analysis algorithm, high-order non-linear interaction analysis is performed on the non-linear relationship between the characteristics of the fluorescence intensity data after dynamic deformation processing, the characteristics of the multispectral fluorescence image data after multi-scale filtering, and the characteristics of the patient clinical data after high-dimensional expansion to obtain interaction features. The specific formula is as follows:

[0022]

[0023] Wherein, is the th characteristic of the fluorescence intensity data after dynamic deformation processing, the th characteristic of the patient clinical data after high-dimensional expansion, and the th characteristic of the multispectral fluorescence image data after multi-scale filtering; is the th characteristic of the fluorescence intensity data after dynamic deformation processing; is the th multispectral fluorescence image data after multi-scale filtering; is the th characteristic of the patient clinical data after high-dimensional expansion; is the first weight factor; is the first adjustment factor; is the second weight factor; is the second adjustment factor; is the third weight factor; is the third adjustment factor.

[0024] Preferably, the S2 specifically includes:

[0025] Based on the interaction features, the esophageal cancer risk score is calculated. The specific risk score formula is as follows:

[0026]

[0027] Wherein, R esophageal is the esophageal cancer risk score; is the weight of each interaction feature; is the scaling factor of the interaction feature; is the bias term; Num1, Num2, and Num3 are the dimensions of the fluorescence enhancement data feature after dynamic deformation processing, the patient clinical data feature after high-dimensional expansion, and the multispectral fluorescence image data feature after multi-scale filtering obtained after the above processing, respectively.

[0028] An early esophageal cancer screening system using fluorescently labeled exosome proteins, comprising the following parts:

[0029] A sample collection and pretreatment module, an exosome isolation module, a fluorescent labeling module, a multispectral fluorescence detection module, and an intelligent screening and analysis module;

[0030] The sample collection and pretreatment module acquires and preprocesses the patient's body fluid sample to obtain the preprocessed body fluid sample, and sends the preprocessed body fluid sample to the exosome isolation module;

[0031] The exosome isolation module separates exosomes from the preprocessed body fluid sample through a microfluidic chip and acoustic sorting technology to obtain an exosome suspension, and sends the exosome suspension to the fluorescent labeling module;

[0032] The fluorescent labeling module specifically labels the target protein in the exosome suspension using fluorescent nanoprobes to obtain a fluorescently labeled exosome solution, and sends the fluorescently labeled exosome solution to the multispectral fluorescence detection module;

[0033] The multispectral fluorescence detection module detects the fluorescence signal of the fluorescently labeled exosome solution to obtain fluorescence intensity data and multispectral fluorescence image data, and sends the fluorescence intensity data and multispectral fluorescence image data to the intelligent screening and analysis module;

[0034] The intelligent screening and analysis module performs intelligent comprehensive analysis based on the fluorescence intensity data and multispectral fluorescence image data, combined with the patient's clinical data, to obtain an analysis result, calculates the esophageal cancer risk score based on the analysis result, and generates a screening result.

[0035] The beneficial effects of the technical solution of the present invention are:

[0036] 1. Through the combination of quantum dot fluorescent probes and DNA nanoprobes, combined with the fluorescence resonance energy transfer (FRET) technology, the detection sensitivity of exosome markers can be significantly enhanced. The high brightness and high stability of quantum dots enable the fluorescence signal to remain stable during the detection process, ensuring that low-concentration tumor markers can also be accurately detected, thereby improving the sensitivity and accuracy of early screening.

[0037] 2. The microfluidic chip and acoustic sorting technology are adopted to precisely focus and separate exosomes by using the high-frequency acoustic wave effect. This separation method combining microfluidic technology and acoustic technology improves the separation efficiency and purity of exosomes, can effectively exclude other impurities such as extracellular vesicles and non-target proteins, ensure the high purity of exosomes, and is beneficial for subsequent labeling and detection.

[0038] 3. Through the intelligent comprehensive analysis of fluorescence intensity data, multi-spectral fluorescence image data and patient clinical data, the risk of esophageal cancer can be evaluated more accurately. Especially, the introduction of non-linear interaction modeling effectively captures the complex dependence relationships between various data sources, making the final esophageal cancer risk score more accurate and reliable. Brief Description of the Drawings

[0039] Figure 1 Structural diagram of an early esophageal cancer screening system using fluorescence-labeled exosome proteins according to the present invention;

[0040] Figure 2 Flowchart of an early esophageal cancer screening method using fluorescence-labeled exosome proteins according to the present invention. Detailed Embodiments

[0041] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0043] The specific solutions of an early esophageal cancer screening system and screening method using fluorescence-labeled exosome proteins provided by the present invention will be specifically described below in conjunction with the drawings.

[0044] Referring to the appendix Figure 1 , which shows the structural diagram of an early esophageal cancer screening system using fluorescence-labeled exosome proteins provided by an embodiment of the present invention. The system includes the following parts:

[0045] Sample collection and pretreatment module, exosome separation module, fluorescence labeling module, multi-spectral fluorescence detection module, intelligent screening and analysis module;

[0046] Sample collection and pretreatment module, which acquires and preprocesses a patient's body fluid sample to remove cell debris and obtains a pretreated body fluid sample for subsequent exosome isolation, and sends the pretreated body fluid sample to the exosome isolation module;

[0047] Exosome isolation module, which isolates exosomes from the pretreated body fluid sample by microfluidic chip and acoustic sorting technology to obtain an exosome suspension, and sends the exosome suspension to the fluorescence labeling module;

[0048] Fluorescence labeling module, which specifically labels target proteins such as CD63, EGFR, EpCAM, HER2, etc. in the exosome suspension using fluorescent nanoprobes to obtain a fluorescently labeled exosome solution, and sends the fluorescently labeled exosome solution into the multi-spectral fluorescence detection module;

[0049] Multi-spectral fluorescence detection module, which detects the fluorescence signal of the fluorescently labeled exosome solution to obtain fluorescence intensity data and multi-spectral fluorescence image data, and sends the fluorescence intensity data and multi-spectral fluorescence image data to the intelligent screening and analysis module;

[0050] Intelligent screening and analysis module, which performs intelligent comprehensive analysis based on the fluorescence intensity data and multi-spectral fluorescence image data, combined with the patient's clinical data, to obtain an analysis result, calculates the esophageal cancer risk score based on the analysis result, and generates a screening result.

[0051] Refer to Appendix Figure 2 , which shows a flowchart of a method for early screening of esophageal cancer by fluorescently labeling exosome proteins provided by an embodiment of the present invention. The method includes the following steps:

[0052] S1. Acquire and preprocess a patient's body fluid sample to obtain a pretreated body fluid sample; isolate exosomes from the pretreated body fluid sample to obtain an exosome suspension; specifically label the exosome suspension using fluorescent nanoprobes to obtain a fluorescently labeled exosome solution;

[0053] Obtain a body fluid sample from a patient, such as blood, saliva or urine, and pre-treat the body fluid sample of the patient to obtain a pre-treated body fluid sample suitable for exosome isolation. Taking blood as an example, the collected body fluid sample of the patient is first centrifuged according to the centrifugation parameters of a centrifuge set by the expert experience method to remove large particle impurities such as red blood cells, white blood cells and platelets in the blood, leaving only smaller extracellular vesicles and exosomes, and obtaining a centrifuged body fluid sample. The centrifuged liquid sample is the supernatant, which contains exosomes and their proteins. Then, it is filtered through a filter membrane with a size of 0.22 μm selected according to the expert experience method to obtain a filtered body fluid sample, removing smaller impurities therein to ensure that the substances in the patient's body fluid sample are roughly uniform and suitable for subsequent processing; Subsequently, according to the specific application scenario, the filtered body fluid sample is diluted with a buffer solution (phosphate buffer solution or Tris-HCl buffer solution) to adjust the pH and ion balance, obtaining a pre-treated body fluid sample to ensure the stability of exosomes for subsequent separation and labeling. The technical means adopted in the pre-treatment process are all well-known to those skilled in the art and will not be elaborated here.

[0054] Further, introduce the pre-treated body fluid sample into a microfluidic chip. The microfluidic chip is designed with tiny channels. Using acoustic sorting technology, separation is carried out according to particles of different sizes and densities. After separation, exosomes will be guided to specific channels to separate the target exosomes smaller than extracellular vesicles, removing extracellular vesicles with larger particles and separating out exosomes to obtain an exosome suspension;

[0055] Using acoustic sorting technology (Acoustofluidics), an exosome suspension can be separated more efficiently. The acoustic sorting technology uses the acoustic field effect generated by high-frequency sound waves to focus exosomes in a specific direction according to size, quickly separating out exosomes and further precisely screening exosomes with sizes between 50 - 150 nm. At the same time, in order to improve the purity of exosomes, CD9 / CD63 / CD81 immunomagnetic beads are used to selectively enrich the exosomes after preliminary separation. The immunomagnetic beads can bind to specific proteins on the surface of exosomes to further purify exosomes and exclude other irrelevant extracellular vesicles or proteins, that is, the immunomagnetic beads and exosomes are aggregated by an external magnetic field, and then unbound impurities are removed by washing with a buffer solution.

[0056] After exosome isolation, it enters the labeling stage of fluorescent nanoprobes, aiming to efficiently label specific tumor-related proteins on the surface of exosomes to ensure high specificity and high sensitivity of the screening results. Quantum dot fluorescent probes are used. Quantum dots are nanomaterials with excellent optical properties, having a narrow emission spectrum, a wide excitation spectrum, and a high quantum yield, capable of providing stable signals with high brightness. By binding quantum dots to specific antibodies of tumor-related proteins (such as EGFR, EpCAM, HER2) on the target exosomes, the surface proteins of the target exosomes are labeled.

[0057] To further enhance the labeling effect, DNA nanoprobes (DNA Aptamer) are used. Through electrostatic adsorption and covalent bonding, DNA nanoprobes can form complexes on the surface of quantum dots and exosome proteins. The complexes can be effectively detected by fluorescence microscopy and flow cytometry. At the same time, to improve the fluorescence intensity and accuracy of labeling, the existing fluorescence resonance energy transfer (FRET) technology is adopted. In this process, when quantum dots approach DNA nanoprobes, the FRET phenomenon occurs, that is, energy is transferred from one fluorescent probe to another, thereby enhancing the intensity of the fluorescence signal and providing more accurate detection results. At this time, after the surface of exosomes is labeled with fluorescent probes, strong fluorescent signals are obtained, enabling subsequent analysis to more clearly identify the tumor-related proteins carried in exosomes.

[0058] At the end of the process of using fluorescent probes for labeling, centrifugation and washing steps well-known to those skilled in the art are adopted to remove unbound fluorescently labeled probes, ensuring that only the fluorescent labels bound to the surface of exosomes are left, obtaining a solution of exosomes labeled with fluorescence.

[0059] S2. Detect the fluorescence signal of the solution of exosomes labeled with fluorescence. Based on the fluorescence signal, obtain fluorescence intensity data and multi-spectral fluorescence image data; based on the fluorescence intensity data and multi-spectral fluorescence image data, combined with the clinical data of patients, conduct intelligent comprehensive analysis to obtain the screening results.

[0060] When detecting the fluorescence signal of the solution of exosomes labeled with fluorescence, first, the fluorescence signal of the solution of exosomes labeled with fluorescence is excited. According to the specific application scenario, the excitation source can be a femtosecond laser, and the femtosecond laser can provide precise pulse width and high energy output to ensure efficient excitation of the fluorescence signal.

[0061] After exciting the fluorescently labeled exosome solution with an excitation light source, the fluorescently labeled exosome solution will emit fluorescent signals at specific wavelengths. A multispectral filter and a multi-channel photodetector are used to capture the fluorescent signals emitted from the fluorescently labeled exosome solution. The fluorescent signals are spectrally processed by the multispectral filter and enter different fluorescence detection channels respectively, and each fluorescence detection channel corresponds to a specific fluorescent wavelength. Further, each fluorescence detection channel converts the captured fluorescent signal into a digital signal to obtain fluorescence intensity data; at the same time, in order to detect fluorescent signals at multiple different wavelengths, the existing multispectral imaging technology is adopted to collect and process the fluorescent signals at different wavelengths in the same multispectral fluorescence image. By the parallel operation of multiple fluorescence detection channels, the fluorescent signals of multiple markers are synchronously detected to obtain multispectral fluorescence image data.

[0062] Further, based on the fluorescence intensity data and the multispectral fluorescence image data, combined with the patient clinical data obtained from the existing database, an intelligent comprehensive esophageal cancer screening analysis algorithm is used for intelligent comprehensive analysis to obtain an analysis result. Based on the analysis result, an esophageal cancer risk score is calculated and a screening result is generated. The intelligent comprehensive esophageal cancer screening analysis algorithm is based on the comprehensive analysis of the fluorescence intensity data, the multispectral fluorescence image data and the patient clinical data, and through processing such as dynamic deformation, multi-scale filtering, high-dimensional expansion, and high-order non-linear interaction, so as to more accurately predict the risk of esophageal cancer. The specific implementation process is as follows:

[0063] The fluorescence intensity data, the multispectral fluorescence image data and the patient clinical data are normalized to ensure the comparability of each data source and eliminate the differences between different data scales, and the normalized fluorescence intensity data, the normalized multispectral fluorescence image data and the normalized patient clinical data are obtained;

[0064] For the normalized fluorescence intensity data, dynamic deformation processing is performed to extract features with more information.

[0065] Based on the fluorescence intensity data S i (t) of the i-th channel after normalization, a recursive deformation formula is introduced to capture the change characteristics of the fluorescent signal over time. The following recursive deformation formula is used:

[0066]

[0067] where is the feature of the fluorescence intensity data after dynamic deformation at the current time t; α1 is the first dynamic weight factor, which is used to control the fluorescence intensity data S of the i-th channel after normalization at the current time t i(t)'s contribution to the final conversion result is determined by the expert experience method; β1 is the second dynamic weight factor, which is used to control the fluorescence intensity data S of the i-th channel normalized at the past moment i (τ)'s contribution is determined by the expert experience method; S i (τ) is the fluorescence intensity data of the i-th channel normalized at time τ; t0 is the starting time point, indicating the start time of the dynamic change of the normalized fluorescence intensity data, which can be the time at the start of the experiment; γ1 is the attenuation factor, which is used to control the attenuation rate of the normalized fluorescence intensity data over time and is inferred based on the biological behavior of exosomes and the physical characteristics of the fluorescence signal; τ is the integration variable, indicating the past time point;

[0068] For the normalized multi-spectral fluorescence image data, a multi-scale filtering method is adopted to extract features from different scales. At each scale, the normalized multi-spectral fluorescence image data is processed through different filters to obtain the multi-spectral fluorescence image data after multi-scale filtering, and it is weighted according to the following formula:

[0069]

[0070] where, is the filtering result of the normalized multi-spectral fluorescence image data at the pixel position (x, y), that is, the feature of the multi-spectral fluorescence image data after multi-scale filtering, contains various information at different scales in the multi-spectral fluorescence image, reflecting the local feature of the normalized multi-spectral fluorescence image data at this position; I(x, y) is the brightness value of the normalized multi-spectral fluorescence image data at the pixel point (x, y); n is the scale index in the multi-scale filtering; N is the total number of scales of the multi-scale filtering, which is determined according to the specific scenario; w n is the weight of the feature of the multi-spectral fluorescence image data after multi-scale filtering at the n-th scale, which is determined by the expert experience method; is the result of filtering the normalized multi-spectral fluorescence image data at scale n, is the filter function, which is a convolution operation or other forms of local feature extraction, such as Gaussian filtering, Laplace filtering, etc., and is determined by the expert experience method; is the variance of the Gaussian function, which determines the width or scale size of the filter response and is calculated through the preset scale n.

[0071] For the normalized patient clinical data, high-dimensional expansion is used to increase the complexity of the features. The high-dimensional expansion formula is as follows:

[0072]

[0073] Among them, is the j-th patient's clinical data feature after high-dimensional expansion; C j is the normalized j-th patient's clinical data, representing a certain clinical information of the patient, such as age, gender, family medical history, past medical history, etc.; P is the number of times of high-dimensional expansion of the normalized patient's clinical data, determined by the expert experience method; k is the index of the number of times of high-dimensional expansion of the normalized patient's clinical data; δ1 is the exponential factor in the high-dimensional expansion process, used to control the intensity of the patient's clinical data feature after high-dimensional expansion, determined by the expert experience method; M is the number of times of cumulative summation, used to control the complexity of the ratio calculation between the normalized patient's clinical data and the integer m; is the normalized j-th patient's clinical data divided by the increasing integer m, indicating C j performs ratio operations with values at different scales. By performing a division operation on C j it reduces the influence of its large-range data on the patient's clinical data feature after high-dimensional expansion and enhances the adaptability to different data scales.

[0074] Furthermore, in order to capture the complex interactions among the fluorescence intensity data, multi-spectral fluorescence image data, and patient clinical data, a high-order non-linear interaction analysis is performed on the non-linear relationship among the fluorescence intensity data features after dynamic deformation processing, multi-scale filtered multi-spectral fluorescence image data features, and patient clinical data features after high-dimensional expansion obtained above to obtain interaction features. The specific formula is as follows:

[0075]

[0076] Among them, is the -th fluorescence intensity data feature after dynamic deformation processing, the -th patient's clinical data feature after high-dimensional expansion, and the -th multi-scale filtered multi-spectral fluorescence image data feature. As any element in the interaction feature data, it captures the non-linear interaction effect between different data; is the first weight factor, used to control the contribution degree of the -th fluorescence intensity data feature after dynamic deformation processing in the interaction feature, determined by the expert experience method; is the first adjustment factor, used to control the influence degree of the -th patient's clinical data feature after high-dimensional expansion on the -th fluorescence intensity data feature after dynamic deformation processing, determined by the expert experience method; is the second weight factor, used to control the Characteristics of multi - spectral fluorescence image data after multi - scale filtering The contribution degree in the interaction features is determined according to the expert experience method; is the second adjustment factor, used to control the influence degree of the clinical data features of the patient after the th high - dimensional expansion on the multi - spectral fluorescence image data features after the th multi - scale filtering, which is determined according to the expert experience method; is the third weight factor, used to control the contribution degree of the clinical data features of the patient after the th high - dimensional expansion in the interaction features, which is determined according to the expert experience method; is the third adjustment factor, used to control the influence degree of the fluorescence intensity data features after dynamic deformation processing on the clinical data features of the patient after the

[0077] Finally, based on the interaction features, the esophageal cancer risk score R esophageal is calculated. The specific risk score formula is as follows:

[0078]

[0079] Among them, is the weight of each interaction feature, which is determined according to the expert experience method; log is the logarithmic function; is the scaling factor of the interaction features, which is determined according to the expert experience method; is the bias term, used to adjust the range and offset of the esophageal cancer risk score, which is determined by the experimental method; Num1, Num2, and Num3 are the dimensions of the fluorescence intensity data features after dynamic deformation processing, the clinical data features of the patient after high - dimensional expansion, and the multi - spectral fluorescence image data features after multi - scale filtering obtained after the above processing respectively.

[0080] Based on the esophageal cancer risk score R esophageal , early screening of esophageal cancer is carried out to obtain the screening result, that is, when the value of R esophageal is between [0, 1], approaching 1 indicates high risk, and approaching 0 indicates low risk.

[0081] In summary, an early screening system and method for esophageal cancer using fluorescently labeled exosome proteins are completed.

[0082] The sequence of the invention embodiments is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0083] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the key point of each embodiment is to illustrate the differences from other embodiments.

[0084] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for early screening of esophageal cancer using fluorescently labeled exosome proteins, characterized in that: The following steps are involved: S1. Obtain and pre-treat a patient's body fluid sample to obtain a pre-treated body fluid sample; Exosomes are separated from the pretreated body fluid sample to obtain an exosome suspension; Use fluorescent nanoprobes to specifically label the exosome suspension to obtain a fluorescently labeled exosome solution; S2. Detecting the fluorescence signal of the fluorescently labeled exosome solution, and obtaining fluorescence intensity data and multi-spectral fluorescence image data based on the fluorescence signal; Based on fluorescence intensity data and multispectral fluorescence image data, combined with patient clinical data, intelligent comprehensive analysis is performed to obtain screening results.

2. The method for early screening of esophageal cancer using fluorescently labeled exosome proteins according to claim 1, characterized in that: The S1 specifically includes: The pretreated body fluid sample is introduced into a microfluidic chip, which is designed with channels. Acoustic wave sorting technology is used to separate particles of different sizes and densities. After separation, the exosomes are guided to specific channels to separate target exosomes that are smaller than extracellular vesicles, and the extracellular vesicles with larger particles are removed. The exosomes are separated to obtain an exosome suspension.

3. The method for early screening of esophageal cancer using fluorescently labeled exosome proteins according to claim 2, characterized in that: The S1 specifically includes: After the exosomes are separated, the fluorescent nanoprobe labeling stage is entered to efficiently label specific tumor-related proteins on the surface of the exosomes. At the end of the fluorescent probe labeling process, centrifugation and washing steps are used to remove unbound fluorescent labeled probes, retaining the fluorescent labels bound to the surface of the exosomes to obtain a fluorescently labeled exosome solution.

4. The method for early screening of esophageal cancer using fluorescently labeled exosome proteins according to claim 1, characterized in that: The S2 specifically includes: Based on fluorescence intensity data and multi-spectral fluorescence image data, combined with patient clinical data, an intelligent comprehensive analysis is performed using an intelligent comprehensive esophageal cancer screening analysis algorithm. The intelligent comprehensive esophageal cancer screening analysis algorithm is based on a comprehensive analysis of fluorescence intensity data, multi-spectral fluorescence image data and patient clinical data, and obtains analysis results through dynamic deformation, multi-scale filtering, high-dimensional expansion, and high-order nonlinear interactive processing. The esophageal cancer risk score is calculated based on the analysis results, and a screening result is generated.

5. The method for early screening of esophageal cancer using fluorescently labeled exosome proteins according to claim 4, characterized in that: The S2 specifically includes: In the process of implementing the intelligent comprehensive esophageal cancer screening and analysis algorithm, the fluorescence intensity data, the multispectral fluorescence image data and the patient's clinical data are normalized to obtain the normalized fluorescence intensity data, the normalized multispectral fluorescence image data and the normalized patient's clinical data.

6. The method for early screening of esophageal cancer using fluorescently labeled exosome proteins according to claim 5, characterized in that: The S2 specifically includes: In the process of implementing the intelligent comprehensive esophageal cancer screening and analysis algorithm, the normalized fluorescence intensity data is dynamically deformed to obtain the fluorescence intensity data characteristics after dynamic deformation processing; the normalized multispectral fluorescence image data is subjected to a multiscale filtering method to extract features from different scales to obtain multispectral fluorescence image data after multiscale filtering; for the normalized patient clinical data, the complexity of the features is increased through high-dimensional expansion to obtain the patient clinical data characteristics after high-dimensional expansion.

7. The method for early screening of esophageal cancer using fluorescently labeled exosome proteins according to claim 6, characterized in that: The S2 specifically includes: In the process of implementing the intelligent comprehensive esophageal cancer screening analysis algorithm, a high-order nonlinear interaction analysis is performed on the nonlinear relationship between the fluorescence intensity data features after dynamic deformation processing, the multi-spectral fluorescence image data features after multi-scale filtering, and the patient clinical data features after high-dimensional expansion to obtain the interaction features. The specific formula is as follows: in, It is The fluorescence data features after dynamic deformation processing, The clinical data characteristics of patients after high-dimensional expansion and The interactive features between the multi-spectral fluorescence image data features after multi-scale filtering; It is The fluorescence intensity data characteristics after dynamic deformation processing; It is Multi-spectral fluorescence image data after multi-scale filtering; It is The clinical data characteristics of patients after high-dimensional expansion; is the first weight factor; It is the first regulatory factor; is the second weighting factor; It is the second regulatory factor; is the third weighting factor; It is the third regulatory factor.

8. The method for early screening of esophageal cancer using fluorescently labeled exosome proteins according to claim 7, characterized in that: The S2 specifically includes: Based on the interactive features, the esophageal cancer risk score is calculated. The specific risk score formula is as follows: Among them, R esophageal is the esophageal cancer risk score; is the weight of each interaction feature; is the scaling factor of the interaction feature; is the bias term; Num1, Num2, and Num3 are the dimensions of the fluorescence emphasized data features after dynamic deformation processing, the patient clinical data features after high-dimensional expansion, and the multi-spectral fluorescence image data features after multi-scale filtering obtained after the above processing.

9. An early screening system for esophageal cancer using fluorescently labeled exosome proteins, applied to an early screening method for esophageal cancer using fluorescently labeled exosome proteins as claimed in claim 1, characterized in that: Includes the following sections: Sample collection and pretreatment module, exosome separation module, fluorescence labeling module, multi-spectral fluorescence detection module, intelligent screening and analysis module; The sample collection and pretreatment module obtains and pretreatments the patient's body fluid samples to obtain the pretreated body fluid samples, and sends the pretreated body fluid samples to the exosome separation module; The exosome separation module separates exosomes from the pre-treated body fluid sample through microfluidic chips and acoustic wave sorting technology to obtain an exosome suspension, and then sends the exosome suspension to the fluorescent labeling module; A fluorescent labeling module uses fluorescent nanoprobes to specifically label the target protein in the exosome suspension to obtain a fluorescently labeled exosome solution, and sends the fluorescently labeled exosome solution to a multi-spectral fluorescence detection module; The multi-spectral fluorescence detection module detects the fluorescence signal of the fluorescently labeled exosome solution, obtains the fluorescence intensity data and the multi-spectral fluorescence image data, and sends the fluorescence intensity data and the multi-spectral fluorescence image data to the intelligent screening and analysis module; The intelligent screening and analysis module performs intelligent comprehensive analysis based on fluorescence intensity data and multispectral fluorescence image data, combined with patient clinical data, to obtain analysis results, calculate the esophageal cancer risk score based on the analysis results, and generate screening results.

Images