Method for detecting physicochemical components of zoonosis pathogens based on multi-dimensional spectrum

Abstract

The application discloses a zoonosis pathogen physicochemical component detection method based on multi-dimensional spectrum, relates to the technical field of spectrum analysis, and guarantees systematic management of all original spectrum data of the same sample under different wave band ranges, different collection micro areas and different collection time sequences through a multi-dimensional original spectrum set Msp, provides a complete basis for subsequent background identification; accurate identification and merging of a host background fitting section and an environment background fitting section are realized through a background fitting result Fit, layering and peeling of host-derived proteins, host metabolic residues and environmental contact surface residues are realized, and interference of non-pathogen physicochemical signals is effectively reduced; after background section deduction, background drift correction and residual response reservation processing of a host background peeling spectrum set Sbg, a true abnormal response section is reserved, low-load or weak-signal zoonosis pathogen-related physicochemical components can be accurately reserved, and stability of determination is ensured.

Description

Technical Field

[0001] This invention relates to the field of spectral analysis technology, specifically to a method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy. Background Technology

[0002] In recent years, the demand for risk identification of zoonotic pathogens in scenarios such as breeding and distribution, slaughter quarantine, environmental contact surface monitoring, and human exposure screening has continued to increase, prompting various rapid detection technologies to develop towards label-free, rapid, and on-site applications. Multidimensional spectroscopy can not only obtain response information related to protein components, lipid components, nucleic acid-related components, and polysaccharide-related components from samples, but also reflect more complex physicochemical changes within samples through joint analysis of different bands, micro-regions, different acquisition sequences, or different spectral morphologies, thereby providing support for the rapid detection and risk assessment of zoonotic pathogens.

[0003] However, in current zoonotic pathogen detection scenarios, actual samples are usually not pure samples composed of a single pathogen component, but often contain complex background components such as host-derived proteins, host-derived lipids, host metabolic residues, cell debris, and organic residues from environmental contact surfaces. For example, in animal nasal swabs, throat swabs, serum trace samples, environmental swab samples, and human exposure monitoring samples, host-derived substances often occupy a stronger signal in the overall spectral response, while pathogen-related components show weak, localized, or unstable abnormal changes. Existing detection methods often directly classify or discriminate the overall spectrum without first separating and stripping the host background in the methodological workflow. This leads to the subsequent judgment process easily mistaking host background fluctuations for pathogen-related anomalies, or submerging the actual pathogen components in the host background, making it difficult to reliably obtain physicochemical component information related to zoonotic pathogens. Precisely because of this situation, it is necessary to construct a host background stripping spectral set, first separating the host background layer from the multidimensional original spectral set, and then performing physicochemical component detection on the stripped spectral set, to solve the problem that pathogen-related components are not easily displayed under complex sample conditions. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy, thus solving the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy, comprising the following steps: S1. Collect multidimensional spectral data of the sample to be tested under different detection dimensions, and perform correlation and organization to construct a multidimensional original spectral set Msp; S2. Perform hierarchical identification and fitting of host background and environmental background on the multidimensional original spectrum set Msp, and generate background fitting result Fit; S3. Based on the background fitting results Fit, perform background segment subtraction, background drift correction and residual response retention processing on the multidimensional original spectrum set Msp to construct the host background stripping spectrum set Sbg. S4. Extract physicochemical components from the host background stripped spectral set Sbg and perform cross-micro-region, cross-temporal, and cross-band correlation discrimination to generate the component discrimination set Cmp; S5. Based on the component discrimination set Cmp, determine whether there are abnormal physicochemical components related to zoonotic pathogens in the sample to be tested, and generate pathogen determination result Res.

[0006] Preferably, S1 includes S11; S11. First, determine the sample number corresponding to the same sample to be tested, and determine the band range, acquisition micro-area location and acquisition sequence of the sample to be tested according to the preset detection scheme; During the band range acquisition process, the same sample to be tested is acquired in multiple preset band ranges, so that the same sample to be tested can output multiple spectra with different band ranges. During the micro-area acquisition process, the detection area of ​​the same sample to be tested is divided into multiple micro-areas according to a preset position sequence, and a spectrum is acquired in each micro-area, so that the same sample to be tested outputs multiple spectra with different acquisition positions. During the acquisition time sequence, the same sample to be tested is acquired multiple times in sequence according to a preset time interval, so that the same sample to be tested outputs multiple spectra with different acquisition times. When each spectrum is acquired, the sample number, band range number, acquisition micro-region number, and acquisition time sequence number corresponding to the spectrum are recorded simultaneously. The spectra obtained from the same sample under all preset band ranges, all preset acquisition micro-regions, and all preset acquisition time sequences are collected and saved to form the raw spectrum group.

[0007] Preferably, S1 further includes S12; S12. Read each spectrum in the raw spectrum group and generate a unique spectrum number for each spectrum according to the combination of sample number, band range number, acquisition micro-area number and acquisition time sequence number. After generating a unique spectrum number, the sample type, sampling location, sampling time, band range, acquisition micro-area location, and acquisition sequence of each spectrum are written into the record information of that spectrum; After the recording information is written, all spectra belonging to the same test sample are grouped into the same sample subset according to the same sample number; Then, within the same sample subset, all spectra are arranged in the order of their band range numbers, in the order of their acquisition micro-region numbers, and in the order of their acquisition time sequence numbers, so that any spectra can be clearly identified as belonging to the same sample, the same band range, the same acquisition micro-region, or the same acquisition time sequence. After completing the generation of unique spectrum numbers, writing of record information, and grouping and sorting, all the spectra are constructed into a multidimensional original spectrum set Msp.

[0008] Preferably, S2 includes S21; S21. First, read each spectrum in the multidimensional original spectrum set Msp, and determine the host background reference range that the spectrum should be matched first according to the sample type and sampling location corresponding to the spectrum. Based on the host background reference range, each spectral segment in the spectrum is compared sequentially to identify the spectral segments that correspond to the response locations of host-derived proteins, host-derived lipids, host metabolic residues, and host cell debris, and these segments are marked as candidate segments for host background. After marking the candidate segments of the host background, the remaining segments in the spectrum are compared according to the sampling scenario and sampling carrier corresponding to the spectrum. The spectrum segments that are consistent with the response positions of environmental contact surface residues, sampling carrier residues and non-pathogen organic residues are identified and marked as candidate segments of the environmental background. For each candidate host background and candidate environmental background segment identified in the spectrum, record the start position, end position, band range, and intensity change of the corresponding segment. The host background candidate segments and environmental background candidate segments corresponding to all spectra in the multidimensional original spectrum set Msp are collected and saved to form the background recognition set Rec.

[0009] Preferably, S2 further includes S22; S22. Read the host background candidate segment and environment background candidate segment corresponding to each spectrum in the background recognition set Rec, and arrange them according to the position order of the spectrum segments within the same spectrum; Adjacent and continuously changing candidate segments of the host background are merged to form a fitted segment of the host background. Adjacent environmental background candidate segments with consistent and continuous changing trends are merged to form a fitted environmental background segment. After merging the fitted segments, calculate the coverage length, average spectral intensity, and spectral intensity fluctuation of each host background fitted segment in the corresponding spectral map, and calculate the coverage length, average spectral intensity, and spectral intensity fluctuation of each environmental background fitted segment in the corresponding spectral map. Then, the host background fitting segment and the environment background fitting segment in each spectrum are written into the corresponding record, so that each spectrum forms a background record containing the background segment position, background segment length, background segment type and background response degree; All background records corresponding to the spectra are collected and saved, and a background fitting result Fit is generated.

[0010] Preferably, S3 includes S31; S31. Read the host background fitting segment and environment background fitting segment corresponding to each spectrum in the background fitting result Fit, and locate the spectrum segment to be processed in the corresponding spectrum according to the background segment position recorded in each background fitting segment. For each background fitting segment obtained by positioning, according to the average spectral intensity of the background fitting segment, the corresponding background value is subtracted point by point from the spectral intensity value of each sampling point within the coverage area of ​​the background fitting segment to complete the background segment subtraction. After the background segment is removed, the starting and ending spectral intensity values ​​of each retained spectral segment are read based on the retained spectral segments that are not marked as background fitting segments in the current spectrum. The baseline flattening process is then performed on the entire spectrum according to the spectral intensity change trend between adjacent retained spectral segments to eliminate the overall upward or downward drift caused by background residue and complete the background drift correction. Each spectrum after background subtraction and background drift correction is saved together to form a corrected spectrum group Cor.

[0011] Preferably, S3 further includes S32; S32. Perform residual response retention processing and reorganization on the calibration spectrum group Cor. Read each spectrum in the calibration spectrum group Cor, and check whether there are still remaining spectrum segments after the background segment subtraction and background drift correction are completed in each spectrum according to the original background fitting segment position recorded by the background fitting result Fit. Then, the original background fitting segment position in each spectrum is screened after background segment subtraction and background drift correction. The remaining spectrum segments that meet the conditions for continuous retention and spectral intensity retention are retained as residual response segments, and the remaining spectrum segments that do not meet the retention conditions are deleted. After the residual response of each spectrum is preserved, the corresponding spectra are grouped into the same sample subset according to the same sample number, and then sorted and organized in the same sample subset according to the band range number, the acquisition micro-area number, and the acquisition time sequence number. All spectra after residual response retention processing and sorting are constructed into a host background stripped spectrum set Sbg.

[0012] Preferably, S4 includes S41; S41. Extract the component segments from each spectrum in the host background stripping spectrum set Sbg. Read each spectrum in the host background stripping spectrum set Sbg and find the local spectrum segments that continuously rise and then continuously fall, or continuously fall and then continuously rise, according to the change of spectral intensity value along the band position. For each local spectral band found, record the starting sampling point position, ending sampling point position, peak sampling point position, peak spectral intensity value, and local spectral band coverage length; Based on the band position of each local spectral segment, the corresponding local spectral segments are then classified into candidate segments for lipid components, candidate segments for protein components, candidate segments for nucleic acid-related components, or candidate segments for polysaccharide-related components, respectively. For local spectral segments in the same spectrum where the starting and ending sampling points overlap, retain the local spectral segments with higher peak intensity values ​​and delete the local spectral segments with lower peak intensity values. After the local spectral segments in each spectrum have been classified and filtered, they are saved together to form a candidate group set Seg.

[0013] Preferably, S4 further includes S42; S42. Perform cross-micro-region, cross-time series and cross-band association discrimination on each candidate component segment in the candidate component set Seg. First, classify all candidate component segments in the candidate component set Seg into the same sample subset according to the same sample number. In the same sample subset, candidate component segments that belong to the same component category and whose peak sampling point positions differ by no more than a preset number of points are merged and compared. For each candidate component segment, the number of times it appears in different acquisition micro-regions, the number of times it appears repeatedly in different acquisition time sequences, and the number of times it is repeatedly detected in different band ranges are counted. When any candidate component segment simultaneously meets the preset number of micro-region occurrences, preset number of time-series repetitions, and preset number of band detections, the candidate component segment is determined to be a stable candidate component; If any candidate component segment does not meet any of the preset micro-area occurrence count, preset time sequence repetition count, and preset band detection count, the candidate component segment is determined to be an occasional interference component and deleted. Candidate component segments that are all identified as stable candidate components in the same sample subset are saved together to form a component discrimination set Cmp.

[0014] Preferably, S5 includes S51; S51. Determine whether there are abnormal physicochemical components related to zoonotic pathogens in the sample to be tested based on the component discrimination set Cmp. Among them, each stable candidate component segment in the component discrimination set Cmp is read, and the number of micro-regions, the number of acquisition time series, and the number of band ranges appearing in the same sample are counted respectively. If any stable candidate component segment appears twice or more in each of the three dimensions of micro-region, acquisition time sequence and band range, it is determined that the sample to be tested has abnormal physicochemical components related to zoonotic pathogens. Otherwise, the sample is determined to have no abnormal physicochemical components related to zoonotic pathogens; The pathogen determination result Res is generated based on the determination result, representing the final detection status.

[0015] This invention provides a method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy, which has the following beneficial effects: (1) The multidimensional original spectrum set Msp ensures that all original spectral data of the same sample under different band ranges, different acquisition micro-regions and different acquisition time sequences are systematically managed, providing a complete foundation for subsequent background identification; the background fitting result Fit achieves the layered stripping of host-origin proteins, host metabolic residues and environmental contact surface residues by accurately identifying and merging the host background fitting segment and the environmental background fitting segment, effectively reducing the interference of non-pathogen physicochemical signals; the host background stripping spectrum set Sbg retains the real abnormal response segment after completing the background segment subtraction, background drift correction and residual response retention processing, so that the physicochemical components related to zoonotic pathogens with low load or weak signal can be accurately retained; the component discrimination set Cmp removes occasional interfering components by repeated verification across micro-regions, time sequences and bands, ensuring the stability of the judgment; the finally generated pathogen judgment result Res can accurately determine whether there are abnormal physicochemical components related to zoonotic pathogens in the sample when the number of micro-regions, acquisition time sequence and band range repetitions all meet the requirements.

[0016] (2) By precisely processing the host background fitting segment and environmental background fitting segment recorded in each spectrum in the background fitting result Fit, background segment deduction, baseline drift correction and retention of abnormal responses are achieved, thus forming a complete host background stripping spectrum set Sbg. When performing residual response retention processing on the correction spectrum group Cor, if the original background fitting segment position still retains more than three consecutive points after processing and the average spectral intensity is higher than the average value of the left and right adjacent retained spectrum segments, then the segment is retained as a residual response segment; isolated points or those whose average spectral intensity does not exceed that of the adjacent segment are set to zero. After screening and sorting by sample number, band number, acquisition micro-area number and acquisition time sequence number, the finally constructed host background stripping spectrum set Sbg can retain candidate abnormal responses while removing host and environmental background interference.

[0017] (3) By accurately extracting candidate physicochemical component segments from the host background stripping spectrum set Sbg after background stripping and performing repeated verification across micro-regions, time series, and wavelength bands, a component discrimination set Cmp is formed and the final pathogen determination result Res is generated, thus realizing the reliable detection of physicochemical components of zoonotic pathogens with low load or weak signal. Subsequently, the candidate component segments are classified and grouped into the candidate component set Seg according to lipids, proteins, nucleic acids, and polysaccharides. In the cross-micro-region, cross-time series, and cross-wave band association discrimination, if the candidate component segments of the same component category and whose peak sampling point positions differ by no more than a preset number of points appear in more than two different micro-regions, more than two different acquisition time series, and more than two different wavelength bands, they are determined to be stable candidate components; otherwise, they are determined to be occasional interfering components and deleted. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the steps of the method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy according to the present invention; Figure 2 This is a schematic diagram of background recognition and fitting. Figure 3 This is a schematic diagram of component extraction, cross-dimensional discrimination, and pathogen identification. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0020] Example 1 This invention provides a method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy. Please refer to [link / reference]. Figure 1 This includes the following steps: S1. Collect multidimensional spectral data of the sample to be tested under different detection dimensions, and perform correlation and organization to construct a multidimensional original spectral set Msp; S2. Perform hierarchical identification and fitting of host background and environmental background on the multidimensional original spectrum set Msp, and generate background fitting result Fit; S3. Based on the background fitting results Fit, perform background segment subtraction, background drift correction and residual response retention processing on the multidimensional original spectrum set Msp to construct the host background stripping spectrum set Sbg. S4. Extract physicochemical components from the host background stripped spectral set Sbg and perform cross-micro-region, cross-temporal, and cross-band correlation discrimination to generate the component discrimination set Cmp; S5. Based on the component discrimination set Cmp, determine whether there are abnormal physicochemical components related to zoonotic pathogens in the sample to be tested, and generate pathogen determination result Res.

[0021] In this embodiment, the multidimensional original spectral set Msp ensures that all original spectral data of the same sample under different waveband ranges, different acquisition micro-regions, and different acquisition times are systematically managed, providing a complete foundation for subsequent background identification. The background fitting result Fit achieves layered stripping of host-origin proteins, host metabolic residues, and environmental contact surface residues by accurately identifying and merging host background fitting segments and environmental background fitting segments, effectively reducing interference from non-pathogen physicochemical signals. After completing background segment subtraction, background drift correction, and residual response retention processing, the host background stripping spectral set Sbg retains the true abnormal response segments, enabling the accurate retention of low-load or weak-signal zoonotic pathogen-related physicochemical components. The component discrimination set Cmp eliminates occasional interfering components through repeated verification across micro-regions, time series, and wavebands, ensuring the stability of the determination. The final pathogen determination result Res can accurately determine whether the sample has abnormal zoonotic pathogen-related physicochemical components when the number of micro-regions, acquisition times, and waveband ranges all meet the requirements. For example, for an animal nasal swab sample, if a stable candidate component is detected in at least two micro-regions, two time series, and two band ranges across multiple samplings at the nasal cavity sampling site, the conclusion that the sample is abnormal can be directly output through the pathogen determination result Res. This solves the problem of easily missing low-load pathogen signals in traditional methods, effectively improving the accuracy, stability, and traceability of detection, while ensuring that real pathogen signals can still be reliably identified in samples from complex environments.

[0022] Example 2 Specifically: S1 includes S11; S11. First, determine the sample number corresponding to the same sample to be tested, and determine the band range, acquisition micro-area location and acquisition sequence of the sample to be tested according to the preset detection scheme; During the band range acquisition process, the same sample to be tested is acquired in multiple preset band ranges, so that the same sample to be tested can output multiple spectra with different band ranges. During the micro-area acquisition process, the detection area of ​​the same sample to be tested is divided into multiple micro-areas according to a preset position sequence, and a spectrum is acquired in each micro-area, so that the same sample to be tested outputs multiple spectra with different acquisition positions. During the acquisition time sequence, the same sample to be tested is acquired multiple times in sequence according to a preset time interval, so that the same sample to be tested outputs multiple spectra with different acquisition times. When each spectrum is acquired, the sample number, band range number, acquisition micro-region number, and acquisition time sequence number corresponding to the spectrum are recorded simultaneously. The spectra obtained from the same sample under all preset band ranges, all preset acquisition micro-regions and all preset acquisition time sequences are collected and saved to form the original spectrum group Raw. The Raw spectral group is used to represent all the original spectral acquisition results of the same sample before background processing.

[0023] S1 further includes S12; S12. Read each spectrum in the raw spectrum group and generate a unique spectrum number for each spectrum according to the combination of sample number, band range number, acquisition micro-area number and acquisition time sequence number. After generating a unique spectrum number, the sample type, sampling location, sampling time, band range, acquisition micro-area location, and acquisition sequence of each spectrum are written into the record information of that spectrum; After the recording information is written, all spectra belonging to the same test sample are grouped into the same sample subset according to the same sample number; Then, within the same sample subset, all spectra are arranged in the order of their band range numbers, in the order of their acquisition micro-region numbers, and in the order of their acquisition time sequence numbers, so that any spectra can be clearly identified as belonging to the same sample, the same band range, the same acquisition micro-region, or the same acquisition time sequence. After completing the generation of unique spectrum numbers, writing of record information, and grouping and sorting, all spectra are constructed into a multidimensional original spectrum set Msp; The multidimensional original spectrum set Msp is used to provide the input basis for subsequent background fitting results Fit, which can be retrieved and compared by sample, by band range, by acquisition micro-region, and by acquisition time sequence.

[0024] In this embodiment, by implementing step S1 of the present invention, the spectral data of the sample under test under different band ranges, acquisition micro-regions, and acquisition times can be systematically collected and rigorously organized to form a structured multidimensional original spectrum set Msp. In the specific operation, each spectrum is assigned a unique spectral number based on the sample number, band range number, acquisition micro-region number, and acquisition time sequence number. The corresponding sample type, sampling location, sampling time, and acquisition parameters are also fully recorded, ensuring that the position and attributes of each spectrum within the entire spectrum set are completely traceable. For example, for an animal nasal swab sample, a total of 18 spectra are collected across three band ranges, three acquisition micro-regions, and two acquisition times. For each spectrum, other spectra of the same sample and their corresponding band and micro-region information can be directly found. By grouping and sorting all spectra of the same sample according to band, micro-region, and time sequence, not only is the speed and accuracy of data retrieval guaranteed, but the subsequent background fitting and spectral band comparison operations can also accurately locate the target spectral band, avoiding data chaos or comparison errors. This improves the organization and traceability of multidimensional spectral analysis, and provides a complete, structured, and searchable input basis for the subsequent generation of background fitting results (Fit), which is conducive to achieving accurate background stripping and identification of abnormal physicochemical components.

[0025] Example 3 Please see Figure 2 Specifically: S2 includes S21; S21. First, read each spectrum in the multidimensional original spectrum set Msp, and determine the host background reference range that the spectrum should be matched first according to the sample type and sampling location corresponding to the spectrum. It should be noted that the method for determining the host background reference range is defined as follows: First, read the sample type and sampling location corresponding to the current spectrum. Then, retrieve the background segment range that matches the sample type and sampling location from the pre-established host background reference table as the host background reference range that is preferentially matched for this spectrum.

[0026] The host background reference table is established as follows: Benchmark spectra of healthy hosts were collected under different sample types and sampling sites, and the healthy hosts did not contain the target zoonotic pathogens. Multiple benchmark spectra of the same sample type and the same sampling location are compared and overlapped, and spectral segments that repeat a preset number of times and whose spectral intensity fluctuations are within a preset range are retained. The retained spectral bands are recorded according to the starting and ending band positions and written under the corresponding sample type and sampling location name to form a host background reference table.

[0027] During implementation, if the information corresponding to the current spectrum is an animal nasal swab sample and a nasal cavity sampling site, then the entire background segment range corresponding to the animal nasal swab sample and nasal cavity sampling site in the host background reference table is directly called. If the information corresponding to the current spectrum is personnel exposure sample and pharyngeal sampling site, then directly call all background segment ranges corresponding to personnel exposure sample - pharyngeal sampling site in the host background reference table; Subsequently, host background matching is performed on the current spectrum only within these called background segments, thereby avoiding indiscriminate searching across the entire spectrum and improving the accuracy and processing speed of host background identification. Based on the host background reference range, each spectral segment in the spectrum is compared sequentially to identify the spectral segments that correspond to the response locations of host-derived proteins, host-derived lipids, host metabolic residues, and host cell debris, and these segments are marked as candidate segments for host background. After marking the candidate segments of the host background, the remaining segments in the spectrum are compared according to the sampling scenario and sampling carrier corresponding to the spectrum. The spectrum segments that are consistent with the response positions of environmental contact surface residues, sampling carrier residues and non-pathogen organic residues are identified and marked as candidate segments of the environmental background. For each candidate host background and candidate environmental background segment identified in the spectrum, record the start position, end position, band range, and intensity change of the corresponding segment. The host background candidate segments and environmental background candidate segments corresponding to all spectra in the multidimensional original spectrum set Msp are collected and saved to form the background recognition set Rec; The background recognition set Rec is used to represent the host background location and environmental background location that have been identified in each spectrum.

[0028] S2 further includes S22; S22. Read the host background candidate segment and environment background candidate segment corresponding to each spectrum in the background recognition set Rec, and arrange them according to the position order of the spectrum segments within the same spectrum; Adjacent and continuously changing candidate segments of the host background are merged to form a fitted segment of the host background. Adjacent environmental background candidate segments with consistent and continuous changing trends are merged to form a fitted environmental background segment. Among them, a continuous increase or decrease in spectral intensity change is predefined as ≥3 consecutive sampling points, and a consistent trend is considered when the change direction of more than 3 consecutive points is consistent. If adjacent segments have the same trend of change (≥3 consecutive points), they are merged into one fitting segment; After merging the fitted segments, calculate the coverage length, average spectral intensity, and spectral intensity fluctuation of each host background fitted segment in the corresponding spectral map, and calculate the coverage length, average spectral intensity, and spectral intensity fluctuation of each environmental background fitted segment in the corresponding spectral map. Then, the host background fitting segment and the environment background fitting segment in each spectrum are written into the corresponding record, so that each spectrum forms a background record containing the background segment position, background segment length, background segment type and background response degree; Save the background records corresponding to all spectra in a centralized manner and generate the background fitting result Fit; The background fitting result Fit is used to provide the basis for processing the background segment position and background response degree for subsequent background segment subtraction, background drift correction and residual response preservation processing on the multidimensional original spectrum set Msp; It should be noted that: For each host background fitting segment or each environment background fitting segment, the coverage length is defined as: The number of consecutive sampling points covered by the fitted segment in the corresponding spectrum, or the band span between the starting and ending band positions of the fitted segment.

[0029] In practice, if the spectral data is stored in the form of discrete sampling points, then starting from the first sampling point of the fitting segment and ending at the last sampling point of the fitting segment, the total number of sampling points between the two points is counted, and the number of sampling points is defined as the coverage length of the fitting segment. If the spectral data is stored in the form of continuous band coordinates, the end band position of the fitted segment is subtracted from the start band position of the fitted segment, and the difference is defined as the coverage length of the fitted segment.

[0030] For example, if a fitted segment extends from the 120th sampling point to the 168th sampling point, then the coverage length of the fitted segment is defined as 49 sampling points; if a fitted segment extends from band position 950 to band position 1010, then the coverage length of the fitted segment is defined as 60 band units.

[0031] The average spectral intensity is defined as: The arithmetic mean of the spectral intensity values ​​of all sampling points within the fitted segment.

[0032] During implementation, first read the spectral intensity value corresponding to each sampling point in the fitting segment, then add all the spectral intensity values ​​together, and finally divide the sum by the total number of sampling points in the fitting segment. The result is defined as the average spectral intensity of the fitting segment.

[0033] If there are n sampling points in the fitted segment, and the spectral intensity values ​​corresponding to each sampling point are the 1st spectral intensity value, the 2nd spectral intensity value to the nth spectral intensity value, then the average spectral intensity of the fitted segment is defined as the sum of the spectral intensity values ​​of all sampling points divided by n; therefore, the average spectral intensity represents the overall spectral intensity level of the fitted segment in the corresponding spectrum.

[0034] The amplitude of the spectral intensity fluctuation is defined as: The difference between the maximum and minimum spectral intensity values ​​within the fitted segment.

[0035] During implementation, the spectral intensity values ​​of all sampling points within the fitting segment are read point by point. The maximum and minimum spectral intensity values ​​are then identified. The minimum spectral intensity value is subtracted from the maximum spectral intensity value, and the result is defined as the spectral intensity fluctuation amplitude of the fitting segment.

[0036] For example, if the maximum spectral intensity of all sampling points within a certain fitting segment is 86 and the minimum spectral intensity is 31, then the spectral intensity fluctuation range of that fitting segment is defined as 55; therefore, the spectral intensity fluctuation range represents the degree of fluctuation in spectral intensity changes within that fitting segment.

[0037] For example: coverage length, average spectral intensity, spectral intensity fluctuation amplitude: Discrete form of sampling points: Points 120 to 168 → Coverage length = 49 sampling points; Continuous band format: Band 950~1010 → Coverage length = 60 band units; Average spectral intensity = arithmetic mean of spectral intensity values ​​at all sampling points; Spectral intensity fluctuation amplitude = maximum spectral intensity value - minimum spectral intensity value; In this embodiment, by implementing step S2 of the present invention, accurate hierarchical identification and fitting of the host background and environmental background for each spectrum in the multidimensional original spectrum set Msp can be achieved, thereby significantly improving the accuracy and processing efficiency of background stripping. In specific operation, each spectrum will call the corresponding background segment range from the pre-established host background reference table according to the sample type and sampling site, avoiding indiscriminate searching on the entire spectrum, reducing erroneous matching, and improving computational efficiency. For example, for the nasal cavity sampling site of animal nasal swab samples, host background matching is performed only within the background segment range corresponding to the nasal cavity sampling site of the animal nasal swab sample. This can quickly and accurately identify the spectral segments that are consistent with the response positions of host-derived proteins, host-derived lipids, host metabolic residues, and host cell debris, and mark them as candidate segments of host background. Subsequently, according to the sampling scene and sampling carrier, environmental residue spectral segments are identified in the remaining spectral segments and marked as candidate segments of environmental background. By merging adjacent candidate segments with consistent and continuous changing trends to form host background fitting segments and environmental background fitting segments, and calculating the coverage length, average spectral intensity, and spectral intensity fluctuation amplitude of each segment, the spatial and intensity characteristics of each background segment can be quantified. For example, for sampling points 120-168 of the spectral band, the coverage length is 49 sampling points, the average spectral intensity is calculated to be 42, and the spectral intensity fluctuation amplitude is 55; the continuous spectral band coverage length of bands 950-1010 is 60 band units. By forming a complete background identification set Rec and generating the background fitting result Fit, accurate background location and intensity basis can be provided for subsequent background segment subtraction, background drift correction, and residual response retention processing. This ensures the reliable retention of abnormal physicochemical signals after background stripping, improves the feasibility and accuracy of detecting low-load or weak-signal pathogens, and reduces the risk of misjudgment due to background interference.

[0038] Example 4 Specifically: S3 includes S31; S31. Read the host background fitting segment and environment background fitting segment corresponding to each spectrum in the background fitting result Fit, and locate the spectrum segment to be processed in the corresponding spectrum according to the background segment position recorded in each background fitting segment. For each background fitting segment obtained by positioning, according to the average spectral intensity of the background fitting segment, the corresponding background value is subtracted point by point from the spectral intensity value of each sampling point within the coverage area of ​​the background fitting segment to complete the background segment subtraction. After the background segment is removed, the starting and ending spectral intensity values ​​of each retained spectral segment are read based on the retained spectral segments that are not marked as background fitting segments in the current spectrum. The baseline flattening process is then performed on the entire spectrum according to the spectral intensity change trend between adjacent retained spectral segments to eliminate the overall upward or downward drift caused by background residue and complete the background drift correction. Save all the spectra after background subtraction and background drift correction to form a corrected spectrum group Cor; The corrected spectrum group Cor is used to represent the spectrum result after background suppression is completed; It should be noted that: The background segment subtraction is defined as: Based on the background segment position and corresponding background response level recorded in the background fitting result Fit, the process of subtracting the background value from the spectral intensity value of each sampling point within the coverage area of ​​the background segment.

[0039] In practice, if a background fitting segment covers the m-th to n-th sampling points and the average spectral intensity corresponding to the background fitting segment is Bgv, then Bgv is subtracted from each spectral intensity value from the m-th to the n-th sampling point, and the result after subtraction is used as the new spectral intensity value after background segment subtraction.

[0040] The new spectral intensity sequence obtained by subtracting the background value point by point is defined as the processing result after the background fitting segment has completed the background segment subtraction.

[0041] The background drift correction is defined as: The process of flattening the overall baseline that still exists in the entire spectrum after the background segment has been removed, whether it is rising or falling.

[0042] During implementation, first select multiple retained spectral segments that were not marked as background fitting segments in the spectrum after background segment subtraction, and then read the starting and ending spectral intensity values ​​of each retained spectral segment. If the spectral intensity values ​​of multiple retained spectral bands increase or decrease as the band position increases, the line connecting the first and last retained spectral bands shall be used as the correction baseline. Then, subtract the corresponding baseline value of the sampling point on the correction baseline from the current spectral intensity value of each sampling point. The result after subtraction is defined as the spectral intensity value after background drift correction. The new spectrum obtained by flattening the entire spectrum through the correction baseline is defined as the spectrum result after background drift correction.

[0043] S3 further includes S32; S32. Perform residual response retention processing and reorganization on the calibration spectrum group Cor. Read each spectrum in the calibration spectrum group Cor, and check whether there are still remaining spectrum segments after the background segment subtraction and background drift correction are completed in each spectrum according to the original background fitting segment position recorded by the background fitting result Fit. Then, the original background fitting segment position in each spectrum is screened after background segment subtraction and background drift correction. The remaining spectrum segments that meet the conditions for continuous retention and spectral intensity retention are retained as residual response segments, and the remaining spectrum segments that do not meet the retention conditions are deleted. The specific filtering instructions are as follows: If a certain original background fitting segment still retains more than three consecutive sampling points after processing, and the average spectral intensity corresponding to the consecutive sampling points is higher than the average spectral intensity of the adjacent retained spectral segment to the left of the original background fitting segment and higher than the average spectral intensity of the adjacent retained spectral segment to the right of the original background fitting segment, then the remaining spectral segment corresponding to the consecutive sampling points is retained as the residual response segment. If a certain original background fitting segment position retains only one or two isolated sampling points after processing, or if the average spectral intensity corresponding to the remaining spectral segment is not higher than the average spectral intensity of the left adjacent retained spectral segment, or is not higher than the average spectral intensity of the right adjacent retained spectral segment, then the remaining spectral intensity value corresponding to the original background fitting segment position is set to zero. After the residual response of each spectrum is preserved, the corresponding spectra are grouped into the same sample subset according to the same sample number, and then sorted and organized in the same sample subset according to the band range number, the acquisition micro-area number, and the acquisition time sequence number. All spectra after residual response retention processing and sorting are constructed into a host background stripped spectrum set Sbg. The host background stripping spectrum set Sbg is used to represent a set of spectra that have had their host background and environmental background removed, while retaining candidate abnormal responses.

[0044] In this embodiment, by implementing step S3 of the present invention, the host background fitting segment and the environment background fitting segment recorded in each spectrum in the background fitting result Fit can be precisely processed to achieve background segment subtraction, baseline drift correction, and retention of abnormal responses, thereby forming a complete host background stripping spectrum set Sbg. In specific operation, the spectral intensity value within the coverage area of ​​each background fitting segment is subtracted point by point from the corresponding average spectral intensity value to complete the background segment subtraction. For example, if the average spectral intensity of the background segment from sampling points 50 to 70 is 30, then the spectral intensity value of each sampling point in this segment is subtracted by 30 to form the subtracted spectrum. Subsequently, the retained spectral segments not marked as background are subjected to baseline flattening processing to eliminate overall upward or downward drift. For example, if the spectral intensity value of the first retained spectral segment is 20 and the last retained spectral segment is 35, then a corrected baseline is generated and the corresponding baseline value is subtracted from all sampling points to obtain the spectrum after background drift correction. Furthermore, when performing residual response retention processing on the calibrated spectrum group Cor, if the original background fitting segment still retains three or more consecutive points with an average spectral intensity higher than the average of the adjacent retained spectral segments after processing, then that segment is retained as a residual response segment; isolated points or those with an average spectral intensity not exceeding that of the adjacent segments are set to zero. After screening and sorting by sample number, band number, acquisition micro-area number, and acquisition time sequence number, the final constructed host background stripping spectrum set Sbg can retain candidate abnormal responses while removing host and environmental background interference. For example, for the spectrum of animal nasal swab samples at the nasal cavity sampling site, after removing the background from sampling points 120-140 of the original background fitting segment, if the spectral intensity of three consecutive points is higher than that of the adjacent segments, then it is retained as a residual response segment; while isolated sampling points are set to zero. By forming a host background stripping spectral set Sbg, subsequent physicochemical component extraction and cross-micro-region, cross-time series and cross-band correlation analysis can be performed directly on the spectrum after interference removal, thereby improving the detection accuracy and reliability of physicochemical components of low-load or weak-signal zoonotic pathogens, while reducing the risk of misjudgment caused by background interference.

[0045] Example 5 Please see Figure 3 Specifically: S4 includes S41; S41. Extract the component segments from each spectrum in the host background stripping spectrum set Sbg. Read each spectrum in the host background stripping spectrum set Sbg and find the local spectrum segments that continuously rise and then continuously fall, or continuously fall and then continuously rise, according to the change of spectral intensity value along the band position. The component segment extraction is defined as follows: The process of finding continuous spectral segments that can form complete local peaks or complete local valleys in a single spectrum and using these continuous spectral segments as candidate physicochemical component segments.

[0046] During implementation, starting from the first sampling point of the current spectrum, the changes in spectral intensity values ​​of adjacent sampling points are compared point by point along the band position; If the spectral intensity values ​​of multiple consecutive sampling points first increase and then decrease point by point, then the interval of these consecutive sampling points is defined as a local peak-shaped segment. If the spectral intensity values ​​of multiple consecutive sampling points decrease point by point and then increase point by point, then the interval of consecutive sampling points is defined as a local valley segment. Record the starting sampling point position, ending sampling point position, and peak or valley sampling point position of the local peak-shaped segment or local valley-shaped segment, and define the local peak-shaped segment or local valley-shaped segment as the candidate component segment after the component segment extraction is completed; For each local spectral band found, record the starting sampling point position, ending sampling point position, peak sampling point position, peak spectral intensity value, and local spectral band coverage length; Based on the band position of each local spectral segment, the corresponding local spectral segments are then classified into candidate segments for lipid components, candidate segments for protein components, candidate segments for nucleic acid-related components, or candidate segments for polysaccharide-related components, respectively. For local spectral segments in the same spectrum where the starting and ending sampling points overlap, retain the local spectral segments with higher peak intensity values ​​and delete the local spectral segments with lower peak intensity values. The local spectral segments that have been classified and filtered in each spectrum are saved together to form a candidate group set Seg; The candidate component set Seg is used to represent the candidate physicochemical component segments extracted from the host background stripping spectrum set Sbg.

[0047] S4 also includes S42; S42. Perform cross-micro-region, cross-time series and cross-band association discrimination on each candidate component segment in the candidate component set Seg. First, classify all candidate component segments in the candidate component set Seg into the same sample subset according to the same sample number. In the same sample subset, candidate component segments that belong to the same component category and whose peak sampling point positions differ by no more than a preset number of points are merged and compared. For each candidate component segment, the number of times it appears in different acquisition micro-regions, the number of times it appears repeatedly in different acquisition time sequences, and the number of times it is repeatedly detected in different band ranges are counted. When any candidate component segment simultaneously meets the preset number of micro-region occurrences, preset number of time-series repetitions, and preset number of band detections, the candidate component segment is determined to be a stable candidate component; If any candidate component segment does not meet any of the preset micro-area occurrence count, preset time sequence repetition count, and preset band detection count, the candidate component segment is determined to be an occasional interference component and deleted. Candidate component segments that are all determined to be stable candidate components in the same sample subset are saved together to form the component discrimination set Cmp; The component discrimination set Cmp is used to represent the abnormal results of physicochemical components in the sample under test after repeated verification across micro-regions, time series and bands; It should be noted that: Cross-micro-region association discrimination is defined as: The process of comparing the repeated occurrences of candidate component segments of the same component category extracted from different micro-regions of the same sample.

[0048] In practice, if candidate component segments belonging to the same component category in the same sample and whose peak sampling point positions differ by no more than a preset number of points are detected in two or more different sampling micro-regions, then the candidate component segments are defined as satisfying cross-micro-region association discrimination. If the candidate component segment appears only in a single acquisition micro-region, then the candidate component segment is defined as not satisfying the cross-micro-region association discrimination.

[0049] Cross-temporal association discrimination is defined as: The process of comparing the repeated occurrences of candidate component segments of the same component category extracted from the same sample at different collection times.

[0050] In practice, if candidate component segments belonging to the same component category and whose peak sampling point positions differ by no more than a preset number of points in the same sample are detected in two or more different sampling time sequences, then the candidate component segments are defined as satisfying cross-time sequence association discrimination. If the candidate component segment appears only in a single acquisition time series, then the candidate component segment is defined as not satisfying the cross-time series association discrimination.

[0051] Cross-band correlation discrimination is defined as: The process of repeatedly detecting and comparing candidate component segments of the same component category extracted from the same sample in different spectral ranges.

[0052] In practice, if candidate component segments belonging to the same component category in the same sample and whose peak sampling point positions differ by no more than a preset number of points are detected in two or more different spectral ranges, then the candidate component segments are defined as satisfying cross-spectral correlation discrimination. If the candidate component segment appears only in a single band range, then the candidate component segment is defined as not satisfying the cross-band association discrimination.

[0053] Stable candidate components are defined as: Candidate component segments that simultaneously satisfy cross-micro-area correlation discrimination, cross-time series correlation discrimination, and cross-band correlation discrimination.

[0054] In practice, if a candidate component segment simultaneously satisfies two or more different acquisition micro-areas, two or more different acquisition timings, and two or more different band ranges within the same sample, then the candidate component segment is defined as a stable candidate component. If a candidate component segment does not simultaneously meet the above three conditions, then the candidate component segment is defined as an occasional interference component.

[0055] S5 includes S51; S51. Determine whether there are abnormal physicochemical components related to zoonotic pathogens in the sample to be tested based on the component discrimination set Cmp. Among them, each stable candidate component segment in the component discrimination set Cmp is read, and the number of micro-regions, the number of acquisition time series, and the number of band ranges appearing in the same sample are counted respectively. If any stable candidate component segment appears twice or more in each of the three dimensions of micro-region, acquisition time sequence and band range, it is determined that the sample to be tested has abnormal physicochemical components related to zoonotic pathogens. Otherwise, the sample is determined to have no abnormal physicochemical components related to zoonotic pathogens; Based on the judgment results, a pathogen judgment result Res is generated, representing the final detection status; It should be noted that: Setting the number of occurrences of stable candidate component segments in the three dimensions of micro-area, acquisition time sequence, and band range to at least twice is to ensure that the components have stable responses at different locations, at different times, and in different bands, thereby eliminating sporadic noise and improving the reliability of judgment, while retaining the ability to detect low-load pathogens.

[0056] In this embodiment, by implementing steps S4 and S5 of the present invention, candidate physicochemical component segments can be accurately extracted from the host background stripped spectral set Sbg after background stripping, and repeated verification can be performed across micro-regions, time series, and spectral bands. This forms a component discrimination set Cmp and generates the final pathogen determination result Res, enabling reliable detection of physicochemical components of low-load or weak-signal zoonotic pathogens. In specific operation, local spectral segments that rise and then fall continuously or fall and then rise continuously in each spectrum are defined as candidate component segments, and the starting sampling point, ending sampling point, peak / valley sampling point, peak spectral intensity, and coverage length are recorded. Subsequently, the candidate component segments are classified and grouped into the candidate component set Seg according to lipids, proteins, nucleic acids, and polysaccharides. In cross-microregion, cross-time series, and cross-band correlation discrimination, candidate component segments of the same component category with peak sampling point positions differing by no more than a preset number of points are identified as stable candidate components if they appear in more than two different microregions, more than two different acquisition times, and more than two different band ranges; otherwise, they are identified as occasional interfering components and deleted. For example, in the spectral map of the nasal cavity sampling site of the same animal nasal swab sample, if a candidate segment of a protein component is detected in three microregions, two acquisition times, and two band ranges, it is identified as a stable candidate component and retained; if it appears only in a single microregion, it is identified as an occasional interfering component and removed. After forming the component discrimination set Cmp, the repetition of each stable candidate component in the three dimensions of microregion, acquisition time series, and band range is statistically analyzed. If any stable candidate component appears twice or more in each of the three dimensions, it is determined that the sample to be tested has abnormal physicochemical components related to zoonotic pathogens, and the detection conclusion is output through the pathogen determination result Res. This method not only eliminates occasional interfering components and noise signals while retaining the true abnormal physicochemical components, but also provides a reliable basis for the detection of low-load pathogens. At the same time, it ensures consistency and repeatability across micro-regions, time series, and wavebands, thereby significantly improving the accuracy and reliability of detection and reducing the risk of misjudgment.

[0057] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended technical solutions and their equivalents.

Claims

1. A method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy, characterized in that: Includes the following steps: S1. Collect multidimensional spectral data of the sample to be tested under different detection dimensions, and perform correlation and organization to construct a multidimensional original spectral set Msp; S2. Perform hierarchical identification and fitting of host background and environmental background on the multidimensional original spectrum set Msp, and generate background fitting result Fit; S3. Based on the background fitting results Fit, perform background segment subtraction, background drift correction and residual response retention processing on the multidimensional original spectrum set Msp to construct the host background stripping spectrum set Sbg. S3 includes S31; S31. Read the host background fitting segment and environment background fitting segment corresponding to each spectrum in the background fitting result Fit, and locate the spectrum segment to be processed in the corresponding spectrum according to the background segment position recorded in each background fitting segment. For each background fitting segment obtained by positioning, according to the average spectral intensity of the background fitting segment, the corresponding background value is subtracted point by point from the spectral intensity value of each sampling point within the coverage area of ​​the background fitting segment to complete the background segment subtraction. After the background segment is removed, the starting and ending spectral intensity values ​​of each retained spectral segment are read based on the retained spectral segments that are not marked as background fitting segments in the current spectrum. The baseline flattening process is then performed on the entire spectrum according to the spectral intensity change trend between adjacent retained spectral segments to eliminate the overall upward or downward drift caused by background residue and complete the background drift correction. Save all the spectra after background subtraction and background drift correction to form a corrected spectrum group Cor; S3 further includes S32; S32. Perform residual response retention processing and reorganization on the calibration spectrum group Cor. Read each spectrum in the calibration spectrum group Cor, and check whether there are still remaining spectrum segments after the background segment subtraction and background drift correction are completed in each spectrum according to the original background fitting segment position recorded by the background fitting result Fit. Then, the remaining spectral segments after background segment subtraction and background drift correction are screened for the original background fitting segment position in each spectrum. The remaining spectral segments that meet the conditions of continuous retention and spectral intensity retention are retained as residual response segments, and the remaining spectral segments that do not meet the retention conditions are deleted. After the residual response of each spectrum is preserved, the corresponding spectra are grouped into the same sample subset according to the same sample number, and then sorted and organized in the same sample subset according to the band range number, the acquisition micro-area number, and the acquisition time sequence number. All spectra after residual response retention processing and sorting are constructed into a host background stripped spectrum set Sbg. S4. Extract physicochemical components from the host background stripped spectral set Sbg and perform cross-micro-region, cross-temporal, and cross-band correlation discrimination to generate the component discrimination set Cmp; S4 includes S41; S41. Extract the component segments from each spectrum in the host background stripping spectrum set Sbg. Read each spectrum in the host background stripping spectrum set Sbg and find the local spectrum segments that continuously rise and then continuously fall, or continuously fall and then continuously rise, according to the change of spectral intensity value along the band position. For each local spectral band found, record the starting sampling point position, ending sampling point position, peak sampling point position, peak spectral intensity value, and local spectral band coverage length; Based on the band position of each local spectral segment, the corresponding local spectral segments are then classified into candidate segments for lipid components, candidate segments for protein components, candidate segments for nucleic acid-related components, or candidate segments for polysaccharide-related components, respectively. For local spectral segments in the same spectrum where the starting and ending sampling points overlap, retain the local spectral segments with higher peak intensity values ​​and delete the local spectral segments with lower peak intensity values. The local spectral segments that have been classified and filtered in each spectrum are saved together to form a candidate group set Seg; S4 also includes S42; S42. Perform cross-micro-region, cross-time series and cross-band association discrimination on each candidate component segment in the candidate component set Seg. First, classify all candidate component segments in the candidate component set Seg into the same sample subset according to the same sample number. In the same sample subset, candidate component segments that belong to the same component category and whose peak sampling point positions differ by no more than a preset number of points are merged and compared. For each candidate component segment, the number of times it appears in different acquisition micro-regions, the number of times it appears repeatedly in different acquisition time sequences, and the number of times it is repeatedly detected in different band ranges are counted. When any candidate component segment simultaneously meets the preset number of micro-region occurrences, preset number of time-series repetitions, and preset number of band detections, the candidate component segment is determined to be a stable candidate component; If any candidate component segment does not meet any of the preset micro-area occurrence count, preset time sequence repetition count, and preset band detection count, the candidate component segment is determined to be an occasional interference component and deleted. Candidate component segments that are all determined to be stable candidate components in the same sample subset are saved together to form the component discrimination set Cmp; S5. Based on the component discrimination set Cmp, determine whether there are abnormal physicochemical components related to zoonotic pathogens in the sample to be tested, and generate pathogen determination result Res.

2. The method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy according to claim 1, characterized in that: S1 includes S11; S11. First, determine the sample number corresponding to the same sample to be tested, and determine the band range, acquisition micro-area location and acquisition sequence of the sample to be tested according to the preset detection scheme; During the band range acquisition process, the same sample to be tested is acquired in multiple preset band ranges, so that the same sample to be tested can output multiple spectra with different band ranges. During the micro-area acquisition process, the detection area of ​​the same sample to be tested is divided into multiple micro-areas according to a preset position sequence, and a spectrum is acquired in each micro-area, so that the same sample to be tested outputs multiple spectra with different acquisition positions. During the acquisition time sequence, the same sample to be tested is acquired multiple times in sequence according to a preset time interval, so that the same sample to be tested outputs multiple spectra with different acquisition times. When each spectrum is acquired, the sample number, band range number, acquisition micro-region number, and acquisition time sequence number corresponding to the spectrum are recorded simultaneously. The spectra obtained from the same sample under all preset band ranges, all preset acquisition micro-regions, and all preset acquisition time sequences are collected and saved to form the raw spectrum group.

3. The method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy according to claim 2, characterized in that: S1 further includes S12; S12. Read each spectrum in the raw spectrum group and generate a unique spectrum number for each spectrum according to the combination of sample number, band range number, acquisition micro-area number and acquisition time sequence number. After generating a unique spectrum number, the sample type, sampling location, sampling time, band range, acquisition micro-area location, and acquisition sequence of each spectrum are written into the record information of that spectrum; After the recording information is written, all spectra belonging to the same test sample are grouped into the same sample subset according to the same sample number; Then, within the same sample subset, all spectra are arranged in the order of their band range numbers, in the order of their acquisition micro-region numbers, and in the order of their acquisition time sequence numbers, so that any spectra can be clearly identified as belonging to the same sample, the same band range, the same acquisition micro-region, or the same acquisition time sequence. After completing the generation of unique spectrum numbers, writing of record information, and grouping and sorting, all the spectra are constructed into a multidimensional original spectrum set Msp.

4. The method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy according to claim 3, characterized in that: S2 includes S21; S21. First, read each spectrum in the multidimensional original spectrum set Msp, and determine the host background reference range that the spectrum should be matched first according to the sample type and sampling location corresponding to the spectrum. Based on the host background reference range, each spectral segment in the spectrum is compared sequentially to identify the spectral segments that correspond to the response locations of host-derived proteins, host-derived lipids, host metabolic residues, and host cell debris, and these segments are marked as candidate segments for host background. After marking the candidate segments of the host background, the remaining segments in the spectrum are compared according to the sampling scenario and sampling carrier corresponding to the spectrum. The spectrum segments that are consistent with the response positions of environmental contact surface residues, sampling carrier residues and non-pathogen organic residues are identified and marked as candidate segments of the environmental background. For each candidate host background and candidate environmental background segment identified in the spectrum, record the start position, end position, band range, and intensity change of the corresponding segment. The host background candidate segments and environmental background candidate segments corresponding to all spectra in the multidimensional original spectrum set Msp are collected and saved to form the background recognition set Rec.

5. The method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy according to claim 4, characterized in that: S2 further includes S22; S22. Read the host background candidate segment and environment background candidate segment corresponding to each spectrum in the background recognition set Rec, and arrange them according to the position order of the spectrum segments within the same spectrum; Adjacent and continuously changing candidate segments of the host background are merged to form a fitted segment of the host background. Adjacent environmental background candidate segments with consistent and continuous changing trends are merged to form a fitted environmental background segment. After merging the fitted segments, calculate the coverage length, average spectral intensity, and spectral intensity fluctuation of each host background fitted segment in the corresponding spectral map, and calculate the coverage length, average spectral intensity, and spectral intensity fluctuation of each environmental background fitted segment in the corresponding spectral map. Then, the host background fitting segment and the environment background fitting segment in each spectrum are written into the corresponding record, so that each spectrum forms a background record containing the background segment position, background segment length, background segment type and background response degree; All background records corresponding to the spectra are collected and saved, and a background fitting result Fit is generated.

6. The method for detecting the physicochemical components of zoonotic pathogens based on multidimensional spectroscopy according to claim 1, characterized in that: S5 includes S51; S51. Determine whether there are abnormal physicochemical components related to zoonotic pathogens in the sample to be tested based on the component discrimination set Cmp. Among them, each stable candidate component segment in the component discrimination set Cmp is read, and the number of micro-regions, the number of acquisition time series, and the number of band ranges appearing in the same sample are counted respectively. If any stable candidate component segment appears twice or more in each of the three dimensions of micro-region, acquisition time sequence and band range, it is determined that the sample to be tested has abnormal physicochemical components related to zoonotic pathogens. Otherwise, the sample is determined to have no abnormal physicochemical components related to zoonotic pathogens; The pathogen determination result Res is generated based on the determination result, representing the final detection status.

Images