A Helicobacter pylori detection method based on mass spectrometry recognition

Abstract

The present application relates to the technical field of Helicobacter pylori detection, and specifically discloses a Helicobacter pylori detection method based on mass spectrometry identification, which first performs MALDI mass spectrometry acquisition on the sample to be tested to obtain its protein fingerprint spectrum, then performs pre-processing to remove noise and standardize the spectrum, further extracts the visual pattern features of the test features with discriminative ability, and simultaneously extracts the reference visual pattern features of each known strain from the constructed strain reference database, and finally, by introducing feature domain-based protein fingerprint spectrum pattern feature query response analysis to learn the complex matching relationship between the test features and the reference features to obtain the fingerprint spectrum visual pattern feature query response encoding representation, and finally determines the strain detection result based on this query response. This strategy based on visual pattern feature extraction and query network matching aims to make full use of the deep pattern information of mass spectrometry data to achieve more reliable Helicobacter pylori mass spectrometry identification.

Description

Technical Field

[0001] The present application relates to the technical field of Helicobacter pylori detection, and more specifically, to a Helicobacter pylori detection method based on mass spectrometry recognition. Background Art

[0002] Helicobacter pylori is a common Gram-negative bacillus that primarily colonizes the human gastric mucosa and is a major causative agent of a variety of upper gastrointestinal diseases, including chronic gastritis, peptic ulcers, gastric mucosa-associated lymphoid tissue lymphoma, and even gastric cancer. Given its widespread infection rate and serious health consequences, the development of rapid, accurate, and reliable H. pylori detection methods is of vital clinical and public health importance for early diagnosis, effective treatment, and the prevention of serious complications such as gastric cancer.

[0003] In recent years, matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) has been widely used in the field of microbial identification due to its rapidity, high throughput, and relatively low cost. Existing mass spectrometry-based microbial detection protocols typically collect a protein fingerprint of the target microorganism (either cultured or directly enriched from a sample) and then compare the resulting mass spectral peak list or entire spectrum with reference spectra of known bacterial species stored in a database to determine the species of the target microorganism. However, these traditional methods, which rely on peak matching or overall spectral similarity calculation, have certain limitations when applied to the detection of specific bacterial species such as Helicobacter pylori. For example, there may be subtle differences in protein expression between different Helicobacter pylori strains or between them and other commensal or pathogenic bacteria, resulting in overall similarity in the mass spectra but differences in key characteristic peaks. At the same time, factors such as sample preparation, culture conditions, and instrument status may also cause spectrum drift and fluctuations in peak intensity. These changes may interfere with the accuracy of traditional alignment algorithms, especially when dealing with low-quality spectra or distinguishing closely related species, which can easily lead to identification confusion or failure, affecting the sensitivity and specificity of detection.

[0004] Therefore, an optimized mass spectrometry-based H. pylori detection protocol is desired. Summary of the Invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide a Helicobacter pylori detection method based on mass spectrometry recognition, which adopts a strategy based on visual pattern feature extraction and query network matching, aiming to fully utilize the deep pattern information of mass spectrometry data to achieve more reliable Helicobacter pylori mass spectrometry recognition.

[0006] According to one aspect of the present application, a method for detecting Helicobacter pylori based on mass spectrometry recognition is provided, comprising: mixing a sample to be tested with a MALDI matrix and spotting the sample on a MALDI target plate, and using a MALDI-TOF mass spectrometer to collect a protein fingerprint spectrum; extracting a set of protein reference fingerprint spectra of known bacterial species from a bacterial species reference database; preprocessing the protein fingerprint spectrum to obtain a preprocessed protein fingerprint spectrum; extracting fingerprint spectrum visual pattern features from the preprocessed protein fingerprint spectrum to obtain visual pattern features of the protein fingerprint spectrum to be tested; extracting fingerprint spectrum visual pattern features from the set of protein reference fingerprint spectra of known bacterial species to obtain a set of protein fingerprint spectrum reference visual pattern features; performing a fingerprint spectrum pattern feature query on the visual pattern features of the protein fingerprint spectrum to be tested and the set of protein fingerprint spectrum reference visual pattern features to obtain a fingerprint spectrum visual pattern feature query response feature; and determining a bacterial species detection result based on the fingerprint spectrum visual pattern feature query response feature.

[0007] In a possible implementation, the preprocessing includes baseline correction, noise smoothing, peak identification, and alignment processing.

[0008] In one possible implementation, extracting fingerprint spectrum visual pattern features from the preprocessed protein fingerprint spectrum to obtain visual pattern features of the protein fingerprint spectrum to be tested includes: passing the preprocessed protein fingerprint spectrum through a fingerprint spectrum visual pattern feature extractor based on a void convolutional neural network model to obtain a visual pattern feature encoding vector of the protein fingerprint spectrum to be tested as the visual pattern feature of the protein fingerprint spectrum to be tested.

[0009] In one possible implementation, fingerprint spectrum visual pattern features are extracted from the set of protein reference fingerprint spectra of the known bacterial species to obtain a set of protein fingerprint spectrum reference visual pattern features, including: passing the protein reference fingerprint spectra of each known bacterial species in the set of protein reference fingerprint spectra of the known bacterial species through the fingerprint spectrum visual pattern feature extractor based on the void convolutional neural network model to obtain a set of protein fingerprint spectrum reference visual pattern feature encoding vectors as the set of protein fingerprint spectrum reference visual pattern features.

[0010] In a possible implementation, a fingerprint pattern feature query is performed on the set of the visual pattern features of the protein fingerprint spectrum to be tested and the protein fingerprint spectrum reference visual pattern features to obtain a fingerprint pattern feature query response feature, including: after feature enhancement of the visual pattern feature coding vector of the protein fingerprint spectrum to be tested, performing monomer semantic query coding on it and each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors to obtain a set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested; calculating the monomer semantic weight of each monomer semantic query score coding vector of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested to obtain a set of monomer query semantic self-attention weights of the visual pattern of the protein fingerprint spectrum to be tested; and aggregating the set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested based on the set of monomer query semantic self-attention weights of the visual pattern of the protein fingerprint spectrum to be tested to obtain the fingerprint spectrum visual pattern feature query response coding vector.

[0011] In a possible implementation, after the visual pattern feature coding vector of the protein fingerprint spectrum to be tested is feature enhanced, it is subjected to monomer semantic query coding together with each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors to obtain a set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested, including: performing feature enhancement based on deconvolution coding on the visual pattern feature coding vector of the protein fingerprint spectrum to be tested to obtain an enhanced visual pattern feature coding vector of the protein fingerprint spectrum to be tested, wherein the enhanced visual pattern feature coding vector of the protein fingerprint spectrum to be tested and each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors have the same feature scale; performing monomer semantic query coding on the enhanced visual pattern feature coding vector of the protein fingerprint spectrum to be tested and each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors to obtain a set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested.

[0012] In a possible implementation, the monomer semantic weights of each monomer semantic query score encoding vector of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested are calculated respectively to obtain a set of monomer query semantic self-attention weights of the visual pattern of the protein fingerprint spectrum to be tested, including: based on the feature set self-distribution characteristics of the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested, determining the monomer semantic matching degree of each monomer semantic query score encoding vector of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested to obtain a set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested; and inputting the set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested into a relationship gating agent module to obtain a set of monomer query semantic self-attention weights of the visual pattern of the protein fingerprint spectrum to be tested.

[0013] In a possible implementation, based on the self-distribution characteristics of the feature set of the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested, the monomer semantic matching degree of each monomer semantic query score encoding vector of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested is determined to obtain a set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested, including: performing inherent alignment calibration on each monomer semantic query score encoding vector of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested to obtain a set of calibrated monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested; and calculating the monomer semantic matching degree of each calibrated monomer semantic query score encoding vector of the visual pattern of the protein fingerprint spectrum to be tested based on the distribution characteristics of the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested to obtain the set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested.

[0014] In one possible implementation, based on the fingerprint spectrum visual pattern feature query response feature, determining the bacterial species detection result includes: passing the fingerprint spectrum visual pattern feature query response encoding vector through a classifier-based bacterial species detector to obtain a bacterial species detection result, and the bacterial species detection result is used to represent a bacterial species category label.

[0015] The present application provides a method for detecting Helicobacter pylori based on mass spectrometry identification, which first performs MALDI mass spectrometry on the sample to be tested to obtain its protein fingerprint spectrum, then removes noise and standardizes it through preprocessing, further extracts the visual pattern features of the test features with discriminative ability, and at the same time extracts the reference visual pattern features of each known strain from the constructed strain reference database. Finally, by introducing feature domain-based protein fingerprint spectrum pattern feature query response analysis to learn the complex matching relationship between the test features and the reference features to obtain the fingerprint spectrum visual pattern feature query response encoding representation, and finally determines the strain detection result based on this query response. This strategy based on visual pattern feature extraction and query network matching aims to make full use of the deep pattern information of mass spectrometry data to achieve more reliable Helicobacter pylori mass spectrometry identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings of the embodiments of the present application. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0017] Figure 1 This is a schematic flow chart of the Helicobacter pylori detection method based on mass spectrometry recognition according to an embodiment of the present application.

[0018] Figure 2 Schematic diagram of data flow of the Helicobacter pylori detection method based on mass spectrometry recognition in an embodiment of the present application.

[0019] Figure 3 This is a schematic flow chart of step S6 in the Helicobacter pylori detection method based on mass spectrometry recognition according to an embodiment of the present application.

[0020] Figure 4 This is a schematic flow chart of step S61 in the Helicobacter pylori detection method based on mass spectrometry recognition according to an embodiment of the present application.

[0021] Figure 5 This is a schematic flow chart of step S62 in the Helicobacter pylori detection method based on mass spectrometry recognition according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts also fall within the scope of protection of this application.

[0023] In response to the above technical problems, the technical solution of this application proposes a Helicobacter pylori detection method based on mass spectrometry recognition, which aims to improve the accuracy and robustness of mass spectrometry detection of Helicobacter pylori. Its core lies in no longer relying solely on discrete peak position and peak intensity information, but focusing on extracting fingerprint spectrum visual pattern features that better reflect the overall morphological and structural information of the spectrum from the protein fingerprint spectrum after pre-processing such as baseline correction, noise smoothing, peak identification and alignment. In other words, protein fingerprint spectrum, like an image, contains rich pattern information. Extracting its visual pattern features can capture the macroscopic structure of the spectrum and the characteristic distribution of key areas. Such features may be more resistant to interference caused by noise and subtle changes than the presence or precise position of a single peak, thereby more stably representing the identity of the bacterial species. Accordingly, the same operation is performed on the reference fingerprint spectra of known bacterial species in the bacterial species reference database to extract reference visual pattern features and construct a reference database based on visual pattern features. Furthermore, in order to achieve efficient and accurate matching, this solution learns the complex matching relationship between the test feature and the reference feature by introducing feature domain-based protein fingerprint spectrum pattern feature query response analysis. This query network goes beyond simple distance calculations or similarity scores to more intelligently determine which reference bacterial species' visual patterns the sample under test most closely resembles. It then outputs a query response encoding the visual pattern features of the fingerprint spectrum, and ultimately determines the bacterial species detection result based on this query response. This strategy, based on visual pattern feature extraction and query network matching, aims to fully utilize the deep pattern information in mass spectrometry data to achieve more reliable mass spectrometric identification of Helicobacter pylori.

[0024] In response to the above technical problems, the technical solution of the present application proposes a Helicobacter pylori detection method based on mass spectrometry recognition. Figure 1 is a schematic flow chart of a Helicobacter pylori detection method based on mass spectrometry recognition according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the Helicobacter pylori detection method based on mass spectrometry recognition according to an embodiment of the present application. Figure 1 and Figure 2As shown, the Helicobacter pylori detection method based on mass spectrometry recognition includes: S1, mixing the sample to be tested with the MALDI matrix and then spotting it on the MALDI target plate, and using a MALDI-TOF mass spectrometer to collect a protein fingerprint spectrum; S2, extracting a set of protein reference fingerprint spectra of known bacterial species from a bacterial species reference database; S3, preprocessing the protein fingerprint spectrum to obtain a preprocessed protein fingerprint spectrum; S4, extracting fingerprint spectrum visual pattern features from the preprocessed protein fingerprint spectrum to obtain visual pattern features of the protein fingerprint spectrum to be tested; S5, extracting fingerprint spectrum visual pattern features from the set of protein reference fingerprint spectra of known bacterial species to obtain a set of protein fingerprint spectrum reference visual pattern features; S6, performing a fingerprint spectrum pattern feature query on the visual pattern features of the protein fingerprint spectrum to be tested and the set of the protein fingerprint spectrum reference visual pattern features to obtain a fingerprint spectrum visual pattern feature query response feature; S7, determining the bacterial species detection result based on the fingerprint spectrum visual pattern feature query response feature.

[0025] Specifically, in step S1, the sample to be tested is mixed with a MALDI matrix and then spotted onto a MALDI target plate. A protein fingerprint spectrum is then acquired using a MALDI-TOF mass spectrometer. It should be understood that in proteomics analysis, MALDI matrices (such as α-cyano-4-hydroxycinnamic acid) can effectively promote the ionization of peptides or proteins in the sample, suppressing interference and enhancing signal. The biological sample to be tested (such as wet cultured Helicobacter pylori cells) is suspended in a suitable solution at a specified concentration and then thoroughly mixed with a proportionately prepared MALDI matrix to ensure sufficient contact between the protein and the matrix crystals. Subsequently, an appropriate amount of the mixture (e.g., 1 μL) is dispensed into designated wells of the MALDI target plate and allowed to evaporate and dry naturally at room temperature, ensuring that the proteins / peptides are evenly embedded within the matrix crystals. This solid-phase spotting method can significantly improve the signal-to-noise ratio and spectral reproducibility of subsequent analyses. After sample preparation, the MALDI target plate is loaded into the MALDI-TOF mass spectrometer. The MALDI-TOF mass spectrometer scans the sample point by point according to preset parameters (such as laser energy and pulse frequency). Laser pulses illuminate the sample-matrix mixture. Matrix molecules absorb the excitation energy, desorbing and ionizing the proteins / peptides embedded within, generating time-of-flight distribution information for the charged ions. The spectrum is internally calibrated and recorded by an external data acquisition system, ultimately outputting a fingerprint mass spectrum for each sample tested. This spectrum is typically presented as a two-dimensional spectrum of m / z (mass-to-charge ratio) and intensity. For example, Helicobacter pylori is cultured, collected by centrifugation, diluted, and then mixed with a MALDI matrix solution at a 1:1 ratio. One microliter of this mixture is then precisely pipetted onto designated wells on a MALDI target plate, allowed to air dry to form a uniform film. Once the target plate is loaded, protein fingerprints are collected stepwise according to the pre-set target points.

[0026] Specifically, in step S2, a set of protein reference fingerprint spectra of known bacterial species is extracted from the bacterial species reference database. It should be understood that in clinical or actual microbial detection scenarios, Helicobacter pylori not only has a variety of strain type spectra, but is also easily interfered by multiple factors such as sample processing, culture conditions, and experimental instruments, resulting in protein fingerprint spectra showing a certain degree of spectrum drift, peak intensity fluctuations, and even the disappearance and addition of characteristic peaks between different batches or different laboratories. Traditional comparison algorithms based on peak or simple full-spectrum similarity are difficult to cope with the potential complex changes and subtle differences in protein spectra between Helicobacter pylori and its closely related species because they rely on discrete peak table information and surface feature distribution. Therefore, in this application, a set of protein fingerprint spectra of known bacterial species (including standard strains of Helicobacter pylori at all levels) is standardized and included in the reference database.

[0027] Specifically, in step S3, the protein fingerprint spectrum is preprocessed to obtain a preprocessed protein fingerprint spectrum. It should be understood that since the original MALDI-TOF mass spectrometry data is often interfered with by various factors, such as sample preparation, culture conditions, instrument status and other factors, the spectrum may cause baseline drift, contain a large amount of random noise, fluctuate in peak intensity and slightly shift in peak position (i.e., spectrum drift). These flaws in the original data will seriously interfere with subsequent analysis. Whether it is a traditional peak matching-based method or a method based on visual pattern features, relatively pure and standardized data input is required. Specifically, in the technical solution of the present application, the preprocessing includes baseline correction, noise smoothing, peak identification and alignment processing. Baseline correction aims to eliminate slowly changing background signals caused by the matrix or other non-peptide substances, ensuring that peak heights truly reflect peptide abundance; noise smoothing aims to reduce high-frequency random fluctuations in the spectrum, improve the signal-to-noise ratio, and make true protein peaks easier to detect; the purpose of peak identification (or peak extraction) is to accurately find the characteristic peaks representing protein or peptide fragments from the smoothed spectrum and determine their mass-to-charge ratio position and intensity information; and alignment processing is to correct for systematic peak position shifts caused by subtle instrument differences or calibration problems between different samples or multiple measurements of the same sample, ensuring that peaks of the same biomarker appear in comparable positions in different spectra.

[0028] Specifically, baseline correction, as the first step in mass spectrometry data processing, fundamentally aims to eliminate background interference signals in the spectrum. This background signal primarily originates from matrix carryover, ion suppression, and the accumulation of nonspecific ions during analysis. This can cause an upward shift or even a slowly fluctuating "false bottom" in the low-frequency region of the spectrum. Baseline correction algorithms include the rolling ball algorithm, TopHat filtering, and least-squares fitting. The rolling ball algorithm uses a specific window size along the laser time axis to simulate the lower bound of the background curve, thereby removing the baseline. For example, a window size of 200 Da is set around the m / z line. A baseline is fitted using the lower edge of a simulated rolling ball, closely following the spectral line and finding local minima. This baseline profile is then subtracted point by point from the original signal to restore the true peak height. TopHat filtering suppresses slowly varying components of the spectrum by constructing morphological structural elements. The least-squares method uses a polynomial fit to construct a background parabola, which is then corrected by subtracting this background function from the original spectrum. High-quality baseline correction ensures that signal intensity reflects only the true protein peak concentration, effectively suppressing interference from false signals.

[0029] After baseline correction, the remaining signal needs to be further smoothed. High-frequency noise in mass spectrometry data primarily manifests as random spikes, noise bands, and periodic minor fluctuations. Effective noise smoothing preserves the sharpness and contrast of true peak shapes while eliminating non-biologically relevant signal fluctuations. Common noise smoothing methods include moving average, Gaussian filtering, and Savitzky-Golay smoothing. Moving average calculates the mean over a fixed window (e.g., 5-11 points). It is suitable for high-frequency noise reduction but slightly reduces resolution. Gaussian filtering uses a Gaussian kernel to weight the average of the original signal, effectively filtering out sudden noise while preserving peak shape. Savitzky-Golay filtering uses a low-order polynomial to fit the signal within a local sliding window, achieving a balance between noise suppression and peak shape preservation. For MALDI-TOF proteomics data, for example, the window width is typically set to 7-15 points, and the filter order is set to 2-3. Adjustment of these parameters should be based on a balance between the specific spectral quality and the desired resolution. Through multiple experiments and empirical evaluation, a standardized signal trajectory with obvious peak structure, low noise background and natural and smooth peak shape was finally obtained.

[0030] After noise smoothing, peak identification is then performed. The goal is to precisely locate characteristic protein / peptide peaks within the spectrum and extract key information parameters (such as m / z value, peak height, peak area, and peak width). This peak identification process requires dynamic and adaptive adjustment based on spectral characteristics such as the overall signal-to-noise ratio, local noise level, and peak spacing. Common methods include threshold determination based on local maxima, peak shape discrimination based on first- and second-order derivatives, and multi-scale detection incorporating prior biological information. In practice, a sliding window is used to traverse the spectral data points. For each data point, the difference between the values of its adjacent points is calculated. A point is considered a legitimate peak if it exceeds the values of its left and right nearest neighbors and meets a preset signal-to-noise ratio threshold (e.g., S / N > 3), and its locally fitted peak width and area meet the protein peak distribution characteristics. For complex samples (such as the Helicobacter pylori proteome), signal processing techniques such as breakpoint detection and continuous peak analysis can be used to merge weak and spurious peaks from multiple measurements, improving peak identification accuracy. Peak identification parameters such as minimum peak height, minimum peak width, and minimum peak area need to be flexibly set according to the performance of the mass spectrometer and the specific experimental system.

[0031] After accurate peak identification, data from sample batches or multiple measurements of a single sample often exhibit slight shifts in the x-axis (m / z dimension) for the same protein peak due to ion source energy drift, timing errors, and other factors. To ensure effective comparison of peak information between spectra, and between the reference database and the sample under test, all spectra must be peak aligned. Peak alignment methods include global linear mapping, local nonlinear registration, and marker-based dynamic time normalization. Linear mapping assumes constant drift across the entire spectrum and calculates the m / z deviations of internal standard peaks or strong characteristic peaks across samples, using a univariate linear regression to assist in spectral alignment. For complex drift patterns, nonlinear registration can be employed. First, a group of co-occurring peaks (i.e., homologous marker peak groups) is identified in each spectrum. Dynamic time warping algorithms are then used to match the m / z sequences, achieving global and local non-isodistant mapping. In addition, existing advanced spectrum alignment toolkits all provide automated multi-spectrum batch alignment functions, which can use interpolation weighting, time domain normalization, waveform hierarchical clustering and other technologies at the bottom layer to ensure that the corresponding relationship of each identified peak in all samples or all database reference spectra is consistent.

[0032] In a specific embodiment, after collecting the protein fingerprint spectrum, the original profile data is first loaded into professional data processing software, and a rolling ball baseline correction with a window size of 15 Da is applied. Subsequently, a Savitzky-Golay filter with a width of 7 points is used for denoising. The maximum detection with a signal-to-noise ratio threshold of 3 is used to identify valid peaks. Then, 3-5 internal standard peaks are selected for linear / nonlinear alignment mapping to obtain the preprocessed protein fingerprint spectrum.

[0033] Specifically, in step S4, the visual pattern features of the fingerprint spectrum are extracted from the pre-treated protein fingerprint spectrum to obtain the visual pattern features of the protein fingerprint spectrum to be tested. It should be understood that since the traditional Helicobacter pylori detection method over-relies on discrete peak positions, peak intensity information or overall spectrum similarity, it is easily affected by spectrum drift, intensity fluctuations and subtle differences between strains caused by factors such as sample preparation and instrument status, resulting in insufficient identification accuracy and stability. Based on this, in order to no longer be limited to these variable information, but to dig out fingerprint spectrum visual pattern features that can better reflect the stable intrinsic structure of the spectrum, in the technical solution of the present application, fingerprint spectrum visual pattern features are further extracted from the pre-treated protein fingerprint spectrum to obtain the visual pattern feature encoding vector of the protein fingerprint spectrum to be tested. This fingerprint spectrum visual pattern feature is considered to be like an image, containing rich macroscopic structure and key area distribution information, more resistant to noise and slight changes, and thus more stably representing the identity of the strain. Therefore, by extracting the fingerprint spectrum visual pattern features, it is possible to extract features that can capture deep structural information such as its overall morphology, peak distribution pattern, and inter-peak relationship from the pre-treated protein fingerprint spectrum.

[0034] Specifically, in one embodiment, extracting fingerprint visual pattern features from the pre-processed protein fingerprint spectrum to obtain visual pattern features of the protein fingerprint spectrum to be tested comprises: passing the pre-processed protein fingerprint spectrum through a fingerprint visual pattern feature extractor based on a dilated convolutional neural network model to obtain a visual pattern feature encoding vector of the protein fingerprint spectrum to be tested as the visual pattern features of the protein fingerprint spectrum to be tested. The purpose of using models such as dilated convolutional neural networks is to utilize their ability to effectively expand the receptive field, capture long-range dependencies and multi-scale features when processing image-like data of fingerprint spectra, thereby more comprehensively capturing the complex global and local pattern features in the mass spectrum, rather than just focusing on isolated peaks. The ultimate goal is to convert these complex visual pattern features into a compact and information-rich mathematical representation, namely, the visual pattern feature encoding vector of the protein fingerprint spectrum to be tested. This encoding vector is a high-level summary and abstraction of the visual features of the original spectrum, aiming to maximize the retention of discriminative information while filtering out irrelevant variations.

[0035] Specifically, dilated convolution introduces an interval sampling mechanism that significantly expands the receptive field, balancing local peaks and global trends while maintaining constant kernel parameters. The network structure is typically stacked in multiple layers, each employing a different dilation rate to capture multi-scale feature information. In one specific embodiment, a fingerprint spectrum visual pattern feature extractor based on a dilated convolutional neural network model first sequentially implements three convolutional layers: a first convolutional layer (64 channels, kernel width 3, dilation rate 1), a second convolutional layer (128 channels, kernel width 3, dilation rate 2), and a third convolutional layer (256 channels, kernel width 3, dilation rate 4). Batch normalization and Reluctance Unit (ReLU) activation are inserted between each layer, and a global average pooling is applied as the output layer. The spatial feature vectors of all channels are averaged channel-wise and compressed into a 256-dimensional visual pattern feature encoding vector for the protein fingerprint spectrum to be tested. The training phase requires the participation of a large number of bacterial species fingerprints and their category labels from a reference database. Through supervised learning, a cross-entropy loss or triplet loss function is defined to guide the classification and segmentation of the feature encoding space. Parameter optimization uses adaptive gradient descent methods such as Adam, with learning rates set between 1e-3 and 1e-5, batch sizes between 32 and 128, and training epochs between 10 and 100, depending on the training set capacity and model convergence properties. During model training, mechanisms such as dropout and L2 regularization can be introduced to prevent overfitting. Data augmentation strategies such as spectral shifting and peak width perturbation can also be used to enrich data diversity. Finally, the parameters after training convergence are frozen and used as the fingerprint spectrum visual pattern feature extractor during the inference phase.

[0036] Specifically, in step S5, fingerprint spectrum visual pattern features are extracted from the set of protein reference fingerprint spectra of the known bacterial species to obtain a set of protein fingerprint spectrum reference visual pattern features. It should be understood that in order to construct a high-quality reference benchmark that matches the feature representation of the sample to be tested and overcome the shortcomings of the traditional method that relies on discrete peak information and is easily interfered with, it is necessary not only to extract such robust features from the sample to be tested, but also to perform the same feature conversion processing on the known bacterial species reference database used for comparison. If the reference database still uses the traditional spectrum or peak list format, it will not be able to effectively compare and match with the visual pattern feature encoding vector of the sample to be tested, making the subsequent query network unable to function. Therefore, in the technical solution of the present application, fingerprint spectrum visual pattern features are further extracted from the set of protein reference fingerprint spectra of the known bacterial species to obtain a set of protein fingerprint spectrum reference visual pattern feature encoding vectors. Specifically, the raw or pre-processed protein fingerprints of a large number of known bacterial species (including different Helicobacter pylori strains and other related species) stored in a bacterial species reference database are converted into a standardized, visual pattern feature-based database using the same feature extraction mechanism used to process the test samples (for example, using the same atrous convolutional neural network model). Specifically, a corresponding protein fingerprint reference visual pattern feature encoding vector is generated for each reference bacterial species spectrum, and these vectors are combined to form a new, feature domain-based reference database. The goal of this database is to accurately and stably encode the characteristic identity of each known bacterial species, ensuring that its representation is consistent with the feature vector of the test sample, laying the foundation for subsequent pattern matching.

[0037] In one embodiment, fingerprint spectrum visual pattern features are extracted from the set of protein reference fingerprint spectra of the known bacterial species to obtain a set of protein fingerprint spectrum reference visual pattern features, including: the protein reference fingerprint spectra of each known bacterial species in the set of protein reference fingerprint spectra of the known bacterial species are respectively passed through the fingerprint spectrum visual pattern feature extractor based on the void convolutional neural network model to obtain a set of protein fingerprint spectrum reference visual pattern feature encoding vectors as the set of protein fingerprint spectrum reference visual pattern features.

[0038] Specifically, in step S6, a fingerprint spectrum pattern feature query is performed on the set of visual pattern features of the protein fingerprint spectrum to be tested and the set of visual pattern features of the protein fingerprint spectrum reference to obtain a fingerprint spectrum visual pattern feature query response feature. It should be understood that traditional methods based on peak list comparison or overall spectrum similarity calculation often stay at the surface feature similarity comparison, and it is difficult to effectively capture complex pattern changes caused by protein expression differences, sample processing or instrument fluctuations, and are easily ineffective when distinguishing closely related species or processing variant spectra. Therefore, it is not enough to simply extract more robust visual pattern features. An intelligent engine that can understand and compare these complex feature patterns is also required. Therefore, in the technical solution of the present application, the set of visual pattern feature encoding vectors of the protein fingerprint spectrum to be tested and the set of visual pattern feature encoding vectors of the protein fingerprint spectrum reference is further input into the fingerprint spectrum pattern feature query network to obtain a fingerprint spectrum visual pattern feature query response encoding vector. In particular, the fingerprint spectrum pattern feature query network can realize high-level and deep comparison between the visual pattern features of the sample to be tested and the numerous reference pattern features in the reference database, and finally condense a discriminant signal that can accurately reflect the attribution of the sample to be tested. This fingerprint pattern feature query network is capable of performing both single-unit semantic query and self-attention aggregation. First, it utilizes single-unit semantic query units to meticulously evaluate the semantic relevance between the test feature vector and each reference feature vector. Here, semantics refers to the species-specific biological information contained in the mass spectrometry visual pattern. This involves more than simply calculating a simple similarity score; rather, it generates a single-unit semantic query score encoding vector for the test protein's fingerprint visual pattern, encompassing multi-dimensional matching information. More importantly, the network then utilizes a self-attention aggregation strategy, taking into account the self-distributed nature of the feature set, to achieve context-aware match assessment. This means the network doesn't view each test-reference pair's match in isolation; instead, it considers the test sample within the context of the entire reference database, dynamically determining which reference patterns have the most significant and consistent matches with the test sample globally, and adjusting weights accordingly to effectively aggregate information. This fingerprint spectrum visual pattern feature query response encoding vector achieves a transition from scattered local feature matching to unified, global semantic judgment, providing a solid and highly confident basis for the subsequent determination of the final bacterial species detection results based on this vector, thereby strongly supporting the goal of more reliable and accurate Helicobacter pylori mass spectrometry identification aimed at achieving by the entire technical solution.

[0039] In one embodiment, Figure 3As shown, a fingerprint spectrum pattern feature query is performed on the set of the visual pattern features of the protein fingerprint spectrum to be tested and the visual pattern features of the protein fingerprint spectrum reference to obtain a fingerprint spectrum visual pattern feature query response feature, including: S61, after feature enhancement of the visual pattern feature coding vector of the protein fingerprint spectrum to be tested, it is subjected to monomer semantic query coding together with each protein fingerprint spectrum reference visual pattern feature coding vector in the set of the protein fingerprint spectrum reference visual pattern feature coding vector to obtain a set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested; S62, the monomer semantic weight of each monomer semantic query score coding vector of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested is calculated respectively to obtain a set of monomer query semantic self-attention weights of the visual pattern of the protein fingerprint spectrum to be tested; S63, based on the set of monomer query semantic self-attention weights of the visual pattern of the protein fingerprint spectrum to be tested, the set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested is aggregated to obtain the fingerprint spectrum visual pattern feature query response coding vector.

[0040] In one embodiment, Figure 4 As shown, after the visual pattern feature coding vector of the protein fingerprint spectrum to be tested is feature enhanced, it is subjected to monomer semantic query coding together with each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors to obtain a set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested, including: S611, feature enhancement based on deconvolution coding is performed on the visual pattern feature coding vector of the protein fingerprint spectrum to be tested to obtain an enhanced visual pattern feature coding vector of the protein fingerprint spectrum to be tested, wherein the enhanced visual pattern feature coding vector of the protein fingerprint spectrum to be tested has the same feature scale as each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors. Specifically, the process can be expressed by the formula:

[0041] ;

[0042] in, is the visual pattern feature encoding vector of the protein fingerprint spectrum to be tested, is deconvolution coding, is the deconvolution weight matrix, is the one-norm of the vector, To enhance the visual pattern feature encoding vector of the protein fingerprint spectrum to be tested.

[0043] Specifically, the visual pattern feature encoding vector of the protein fingerprint spectrum to be tested is subjected to feature enhancement based on deconvolution coding. Its core effect is to greatly improve the expressive power and dimensional semantic richness of the features. The deconvolution coding (i.e., inverse convolution or transposed convolution) mechanism can achieve feature reconstruction and enhancement in high-dimensional space while retaining the original key spectral structural information by learning data-driven feature upsampling weights. Specifically, this operation can not only improve the spatial resolution of the features of the sample to be tested and strengthen the representation of microscopic patterns and global structures, but also adaptively enhance the content of the original encoding vector through a trainable weight system, effectively recovering, completing or highlighting the potential differences and discriminative features in the fingerprint spectrum that are not fully expressed by the superficial encoding. Compared with direct linear transformation or ordinary interpolation, the data-adaptive feature enhancement brought by deconvolution coding significantly expands the basis for comparing sample and database reference feature vectors, so that each vector is stretched into the same high-level feature space. In addition, the deconvolution process ensures that the enhanced vector is completely consistent with the visual pattern feature encoding vectors of all protein fingerprint spectra in the reference database in terms of characteristic scale, dimension and internal structure distribution. This standardization is not only a technical guarantee for subsequent fair matching, but also can significantly improve the generalization and robustness of the pattern recognition system.

[0044] S612, performing monomer semantic query encoding on each protein fingerprint reference visual pattern feature encoding vector in the set of the enhanced protein fingerprint spectrum to be tested and the protein fingerprint spectrum reference visual pattern feature encoding vector to obtain a set of monomer semantic query score encoding vectors of the protein fingerprint spectrum to be tested. Specifically, this process can be expressed as follows:

[0045] ;

[0046] ;

[0047] in, is the set of reference visual pattern feature encoding vectors of protein fingerprint spectrum, are the first, second, and third in the set of protein fingerprint reference visual pattern feature encoding vectors. and Protein fingerprint spectrum reference visual pattern feature encoding vector, and are the trainable weight matrix and the trainable bias vector, respectively. For vector concatenation operations, is the hyperbolic tangent function, For the The semantic query score encoding vector of the visual pattern monomer of the protein fingerprint spectrum to be tested.

[0048] Specifically, each protein fingerprint spectrum reference visual pattern feature encoding vector in the set of the enhanced protein fingerprint spectrum visual pattern feature encoding vector and the protein fingerprint spectrum reference visual pattern feature encoding vector is subjected to monomer semantic query encoding to obtain a set of monomer semantic query score encoding vectors of the protein fingerprint spectrum visual pattern to be tested. Here, the actual effect of the monomer semantic query encoding module is to automatically learn and quantify the deep semantic association and precise matching degree of the corresponding feature pairs composed of the enhanced protein fingerprint spectrum visual pattern feature encoding vector and each protein fingerprint spectrum reference visual pattern feature encoding vector through deep feature interaction of the neural network. For example, for different Helicobacter pylori strains or related species under the same disease background, although their protein fingerprint visual patterns are similar as a whole, the deep distribution structure, peak domain configuration or sub-block pattern will reflect rich differences. Through monomer semantic query, this series of semantic information can be encoded in a high-dimensional manner in the matching vector. The monomer semantic query score encoding vector of the protein fingerprint spectrum visual pattern to be tested is not a single score, but reflects the entire association pattern of the corresponding feature pair in the multi-semantic subspace, so that the pairing between the test spectrum and all database reference spectra presents a rich and detailed dynamic feature relationship distribution.

[0049] In one embodiment, Figure 5 As shown, the monomer semantic weights of each monomer semantic query score encoding vector of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested are calculated respectively to obtain a set of monomer query semantic self-attention weights of the visual pattern of the protein fingerprint spectrum to be tested, including: S621, based on the feature set self-distribution characteristics of the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested, determining the monomer semantic matching degree of each monomer semantic query score encoding vector of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested to obtain a set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested; S622, inputting the set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested into a relationship gated agent module to obtain a set of monomer query semantic self-attention weights of the visual pattern of the protein fingerprint spectrum to be tested.

[0050] In one embodiment, in step S621, based on the feature set self-distribution characteristics of the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested, the monomer semantic matching degree of each monomer semantic query score encoding vector of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested is determined to obtain a set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested, including: performing inherent alignment calibration on each monomer semantic query score encoding vector of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested to obtain a set of calibrated monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested. Specifically, this process can be expressed by the formula:

[0051] ;

[0052] ;

[0053] ;

[0054] ;

[0055] ;

[0056] in, is the visual pattern query potential vector of the protein fingerprint spectrum to be tested, is the projection weight matrix, Query the interactive encoding vector for the visual pattern of the protein fingerprint spectrum to be tested, is the coupling constant, calculated in the same way as for deconvolution enhancement to preserve symmetry, Query the covariance matrix for the visual pattern of the protein fingerprint spectrum to be tested, For the The semantic query score encoding vector of the visual pattern monomer of the protein fingerprint spectrum to be tested after calibration.

[0057] It should be understood that the single-unit semantic query encoding employs a direct feature concatenation approach to generate a single-unit semantic query score encoding vector for the visual pattern of the protein fingerprint spectrum under test by directly concatenating the enhanced feature encoding vector of the protein fingerprint spectrum under test with the feature encoding vector of the corresponding reference protein fingerprint spectrum in the database. This process is expected to significantly improve the representation accuracy of the single-unit semantic query score encoding vector for the visual pattern of the protein fingerprint spectrum under test by optimizing the dynamic feature representation from the original feature space to the semantic query encoding space, and by emphasizing the inherent alignment between the feature space and the semantic query encoding space.

[0058] Specifically, the visual pattern feature encoding vector of the protein fingerprint spectrum to be tested and the visual pattern feature encoding vectors of each protein fingerprint spectrum reference are combined through a cascade operation to form a dynamic eigenvector representation, namely the visual pattern query interaction encoding vector of the protein fingerprint spectrum to be tested. The visual pattern query interaction encoding vector of the protein fingerprint spectrum to be tested is further spatially mapped to obtain its projection representation to the semantic query encoding eigenstate, namely the visual pattern query potential vector of the protein fingerprint spectrum to be tested. On this basis, the visual pattern query potential vector of the protein fingerprint spectrum to be tested under the mapped representation is inherently aligned. This inherent alignment uses the same coupling constant setting as in the deconvolution feature enhancement step, effectively ensuring the symmetry and mathematical consistency of the entire mapping process. At the same time, by rationally constructing and solving the covariance matrix of the visual pattern query of the protein fingerprint spectrum to be tested, the covariance path is constrained under the canonical symmetry condition, thereby fixing the mapping norm. This mechanism improves the representation accuracy of the semantic query score encoding vector of the visual pattern of the protein fingerprint spectrum to be tested in the high-dimensional semantic space.

[0059] Based on the distribution characteristics of the set of monomer semantic query score encoding vectors of the visual pattern of the calibrated protein fingerprint spectrum to be tested, the monomer semantic matching degree of each monomer semantic query score encoding vector of the visual pattern of the calibrated protein fingerprint spectrum to be tested is calculated to obtain the set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested. Specifically, the process can be expressed by the formula:

[0060] ;

[0061] in, for The number of For the A calibrated protein fingerprint spectrum visual pattern monomer semantic query score encoding vector, Indicated Transpose the vector, is the natural exponential function with base e, for function, is the semantic matching degree of the visual pattern monomer of the protein fingerprint spectrum to be tested.

[0062] Specifically, the calculation of the monomer semantic matching degree is based not only on the numerical characteristics of the monomer semantic query score encoding vector of the visual pattern of the calibrated protein fingerprint spectrum to be tested, but also on the distribution characteristics of the set of monomer semantic query score encoding vectors of the visual pattern of the calibrated protein fingerprint spectrum to be tested. In this way, those feature pairs that can significantly distinguish the test spectrum from the reference database and have high biological relevance and discriminant efficiency will be given a higher matching degree, while the influence of noisy, redundant or weakly discriminative feature pairs will be automatically weakened. Ultimately, all monomer semantic matching degrees together constitute the set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested.

[0063] Specifically, the set of semantic matching degrees of the visual pattern monomers of the protein fingerprint spectrum to be tested is input into the relational gating agent module to obtain the set of query semantic self-attention weights of the visual pattern monomers of the protein fingerprint spectrum to be tested. The calculation process can be expressed as follows:

[0064] ;

[0065] in, is the mask function, is a preset threshold, such as 0.5. Of course, this is just an example and can be adjusted according to actual conditions. Query the semantic self-attention weight for the visual pattern monomer of the protein fingerprint spectrum to be tested.

[0066] Specifically, the Relationship Gating Proxy Module, leveraging a trainable gating mechanism, effectively retains the semantic matches of individual visual patterns in the protein fingerprint spectrum most relevant to the identification of target bacteria, such as Helicobacter pylori, while automatically suppressing the negative impact of accidental high scores, weak correlations, or redundant pairings on the overall decision outcome. Based on contextual understanding of the global feature space, the Relationship Gating Proxy Module dynamically weights the semantic matches of individual visual patterns in the protein fingerprint spectrum in subsequent self-attention aggregation, ensuring that the final focus of the self-attention mechanism closely matches the actual identification requirements.

[0067] Specifically, the calculation process of aggregating the set of semantic query score encoding vectors of the visual pattern monomers of the protein fingerprint spectrum to be tested based on the set of semantic self-attention weights of the monomer query of the visual pattern of the protein fingerprint spectrum to be tested to obtain the fingerprint spectrum visual pattern feature query response encoding vector can be expressed as follows:

[0068] ;

[0069] in, Encode vector for fingerprint spectrum visual pattern feature query response.

[0070] First, through deep modeling of the feature domain, a single semantic query score encoding vector is obtained between each visual pattern of the protein fingerprint spectrum to be tested and the reference pattern in the database. Leveraging the contextual distribution characteristics of the feature set and a relational gating proxy mechanism, an adaptive single-query semantic self-attention weight is assigned to each encoding vector. These weights reflect the actual contribution and discriminative power of each encoding vector in the overall feature representation, effectively highlighting highly discriminative matching information while weakening or suppressing noise and redundancy. On this basis, the aggregation step uses the self-attention weight as a weighting coefficient to dynamically linearly combine all single semantic query score encoding vectors. During the aggregation process, not only is the local matching information at the microscopic level weighted and integrated according to its importance, but also a global semantic abstraction and consistent expression of the entire visual structure of the protein fingerprint spectrum is achieved.

[0071] Specifically, in step S7, the strain detection result is determined based on the fingerprint spectrum visual pattern feature query response feature. In one embodiment, the strain detection result is determined based on the fingerprint spectrum visual pattern feature query response feature, including: passing the fingerprint spectrum visual pattern feature query response encoding vector through a strain detector based on a classifier to obtain a strain detection result, and the strain detection result is used to represent the strain category label. Specifically, the Softmax classifier is widely used in the final multi-classification discrimination module after deep learning features because of its rigorous structure, strong differentiability, and suitability for multi-class probability modeling. In a specific embodiment of the present application, the fingerprint spectrum visual pattern feature query response encoding vector is input into the Softmax classifier. If the fingerprint spectrum visual pattern feature query response encoding vector is a high-dimensional vector of fixed length, and there are m strain categories in the system reference database, then for each category i to be discriminated, the Softmax classifier first multiplies the learnable weight vector of the category, adds the corresponding bias, obtains the corresponding score, and further obtains the probability distribution through exponential operation, and finally outputs the category corresponding to the maximum pi as the strain detection result, and this probability distribution can be directly regarded as an indicator of the credibility of the model.

[0072] In summary, the Helicobacter pylori detection method based on mass spectrometry recognition provided by the present application first performs MALDI mass spectrometry on the sample to be tested to obtain its protein fingerprint spectrum, then removes noise and standardizes it through preprocessing, further extracts the visual pattern features of the features to be tested that have the ability to discriminate, and at the same time extracts the reference visual pattern features of each known strain from the constructed strain reference database, and finally, by introducing the feature domain-based protein fingerprint spectrum pattern feature query response analysis to learn the complex matching relationship between the features to be tested and the reference features to obtain the fingerprint spectrum visual pattern feature query response encoding representation, and finally determines the strain detection result based on this query response. This strategy based on visual pattern feature extraction and query network matching aims to make full use of the deep pattern information of mass spectrometry data to achieve more reliable Helicobacter pylori mass spectrometry identification.

[0073] The basic principles of the present application have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in this application are merely illustrative and not restrictive, and should not be construed as necessarily possessed by each embodiment of the present application. Furthermore, the specific details of the above embodiments are provided for illustrative purposes and to facilitate understanding, rather than as limitations. These details do not limit the present application to necessarily being implemented using these specific details.

[0074] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0075] In addition, the functional modules in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0076] Finally, it should be noted that the above description has been provided for the purpose of illustration and description. In addition, the above embodiments are intended only to illustrate the technical solutions of the present application and are not intended to be limiting. Although the present application has been described in detail with reference to preferred embodiments, those skilled in the art will appreciate that the technical solutions of the present application may be modified or replaced with equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A method for detecting Helicobacter pylori based on mass spectrometry, characterized in that: include: The sample to be tested is mixed with the MALDI matrix and spotted on the MALDI target plate, and the protein fingerprint spectrum is collected using a MALDI-TOF mass spectrometer; Extracting a collection of protein reference fingerprint spectra of known bacterial species from a bacterial species reference database; preprocessing the protein fingerprint spectrum to obtain a preprocessed protein fingerprint spectrum; Extracting fingerprint spectrum visual pattern features from the preprocessed protein fingerprint spectrum to obtain visual pattern features of the protein fingerprint spectrum to be tested, comprising: passing the preprocessed protein fingerprint spectrum through a fingerprint spectrum visual pattern feature extractor based on a dilated convolutional neural network model to obtain a visual pattern feature encoding vector of the protein fingerprint spectrum to be tested as the visual pattern feature of the protein fingerprint spectrum to be tested; Extracting fingerprint spectrum visual pattern features from the set of protein reference fingerprint spectra of the known bacterial species to obtain a set of protein fingerprint spectrum reference visual pattern features, comprising: passing the protein reference fingerprint spectrum of each known bacterial species in the set of protein reference fingerprint spectra of the known bacterial species through the fingerprint spectrum visual pattern feature extractor based on the void convolutional neural network model to obtain a set of protein fingerprint spectrum reference visual pattern feature encoding vectors as the set of protein fingerprint spectrum reference visual pattern features; Performing a fingerprint spectrum pattern feature query on the set of the visual pattern feature of the protein fingerprint spectrum to be tested and the visual pattern feature of the protein fingerprint spectrum reference to obtain a fingerprint spectrum visual pattern feature query response feature; Determining a bacterial species detection result based on the fingerprint spectrum visual pattern feature query response feature; Wherein, performing a fingerprint spectrum pattern feature query on the set of the visual pattern feature of the protein fingerprint spectrum to be tested and the visual pattern feature of the reference protein fingerprint spectrum to obtain a fingerprint spectrum visual pattern feature query response feature includes: After feature enhancement is performed on the visual pattern feature coding vector of the protein fingerprint spectrum to be tested, monomer semantic query coding is performed on the vector and each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors to obtain a set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested; Calculating the monomer semantic weight of each monomer semantic query score encoding vector of the protein fingerprint spectrum visual pattern to be tested in the set of monomer semantic query score encoding vectors of the protein fingerprint spectrum visual pattern to be tested to obtain a set of monomer semantic self-attention weights of the protein fingerprint spectrum visual pattern to be tested; The set of monomer semantic query score encoding vectors of the protein fingerprint spectrum visual pattern to be tested is aggregated based on the set of monomer semantic self-attention weights of the protein fingerprint spectrum visual pattern to be tested to obtain the fingerprint spectrum visual pattern feature query response encoding vector.

2. The method for detecting Helicobacter pylori based on mass spectrometry recognition according to claim 1, characterized in that: The preprocessing includes baseline correction, noise smoothing, peak identification and alignment.

3. The method for detecting Helicobacter pylori based on mass spectrometry recognition according to claim 2, characterized in that: After feature enhancement is performed on the visual pattern feature coding vector of the protein fingerprint spectrum to be tested, monomer semantic query coding is performed on the vector and each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors to obtain a set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested, including: Performing feature enhancement based on deconvolution coding on the visual pattern feature coding vector of the protein fingerprint spectrum to be tested to obtain an enhanced visual pattern feature coding vector of the protein fingerprint spectrum to be tested, wherein the enhanced visual pattern feature coding vector of the protein fingerprint spectrum to be tested has the same feature scale as each protein fingerprint spectrum reference visual pattern feature coding vector in the set of protein fingerprint spectrum reference visual pattern feature coding vectors; Each protein fingerprint spectrum reference visual pattern feature coding vector in the set of the enhanced protein fingerprint spectrum visual pattern feature coding vector to be tested and the protein fingerprint spectrum reference visual pattern feature coding vector is subjected to monomer semantic query coding to obtain a set of monomer semantic query score coding vectors of the visual pattern of the protein fingerprint spectrum to be tested.

4. The method for detecting Helicobacter pylori based on mass spectrometry recognition according to claim 3, characterized in that: Calculating the monomer semantic weight of each monomer semantic query score encoding vector of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the protein fingerprint spectrum to be tested respectively to obtain a set of monomer semantic self-attention weights of the protein fingerprint spectrum to be tested visual pattern, including: Determining the monomer semantic matching degree of each protein fingerprint spectrum visual pattern monomer semantic query score encoding vector in the set of protein fingerprint spectrum visual pattern monomer semantic query score encoding vectors based on the feature set self-distribution characteristics of the set of protein fingerprint spectrum visual pattern monomer semantic query score encoding vectors to obtain a set of protein fingerprint spectrum visual pattern monomer semantic matching degrees; The set of semantic matching degrees of the visual pattern monomers of the protein fingerprint spectrum to be tested is input into the relationship gate agent module to obtain the set of query semantic self-attention weights of the visual pattern monomers of the protein fingerprint spectrum to be tested.

5. The method for detecting Helicobacter pylori based on mass spectrometry recognition according to claim 4, characterized in that: The method comprises determining the monomer semantic matching degree of each of the monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested in the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested based on the feature set self-distribution property of the set of monomer semantic query score encoding vectors of the visual pattern of the protein fingerprint spectrum to be tested to obtain the set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested, comprising: performing inherent alignment calibration on each of the protein fingerprint spectrum visual pattern monomer semantic query score encoding vectors in the set of protein fingerprint spectrum visual pattern monomer semantic query score encoding vectors to obtain a set of calibrated protein fingerprint spectrum visual pattern monomer semantic query score encoding vectors; Based on the distribution characteristics of the set of monomer semantic query score encoding vectors of the calibrated visual pattern of the protein fingerprint spectrum to be tested, the monomer semantic matching degree of each calibrated visual pattern monomer semantic query score encoding vector of the protein fingerprint spectrum to be tested is calculated to obtain the set of monomer semantic matching degrees of the visual pattern of the protein fingerprint spectrum to be tested.

6. The method for detecting Helicobacter pylori based on mass spectrometry recognition according to claim 5, characterized in that: Based on the fingerprint spectrum visual pattern feature query response feature, determining the bacterial species detection result includes: passing the fingerprint spectrum visual pattern feature query response encoding vector through a classifier-based bacterial species detector to obtain a bacterial species detection result, and the bacterial species detection result is used to represent a bacterial species category label.

Images