Fingerprint spectrogram generation method, device and system and computer readable storage medium

Abstract

The invention relates to a fingerprint spectrogram generation method, device and system and a computer readable storage medium, and relates to the technical field of image processing. The fingerprint spectrogram generation method comprises the following steps: determining intra-group features of each spectrogram of a plurality of spectrograms of lipid of a target type; determining inter-group features of the plurality of spectrograms; and according to the intra-group features and the inter-group features, generating a fingerprint spectrogram of the lipid of the target type.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technology, and in particular to a method, device and system for generating a fingerprint spectrum, and a computer-readable storage medium. Background Art

[0002] Mass spectrometry is an analytical method widely used in the fields of environment, biology, medicine, etc. Mass spectra of different substances can be obtained through mass spectrometry instruments. An important application of mass spectrometry technology is to identify substances by comparing mass spectrometry data with standard libraries and characteristic peak library information.

[0003] In the related art, a fingerprint spectrum library is constructed by obtaining the spectrum of a reference standard on a mass spectrometer. Summary of the invention

[0004] According to the first aspect of the present disclosure, a method for generating a fingerprint spectrum is provided, comprising: determining the intra-group features of each spectrum of multiple spectra of a target type of lipid; determining the inter-group features of the multiple spectra; and generating a fingerprint spectrum of the target type of lipid based on the intra-group features and the inter-group features.

[0005] In some embodiments, determining the intra-group features of each spectrum of the plurality of spectra of the target type of lipids comprises: extracting the intra-group features from each fingerprint spectrum using a first activation layer and a linear layer of a fingerprint spectrum generation model.

[0006] In some embodiments, determining the inter-group features of the plurality of spectra comprises: determining the inter-group features of the plurality of spectra based on the intra-group features of the plurality of spectra.

[0007] In some embodiments, determining the inter-group features of the multiple spectra based on the intra-group features of the multiple spectra includes: utilizing a multi-head attention layer of a fingerprint spectrogram generation model to determine the inter-group features of the multiple spectra based on the intra-group features of the multiple spectra.

[0008] In some embodiments, generating a fingerprint spectrum of the target type of lipids based on the intra-group features and the inter-group features includes: using a normalization layer of a fingerprint spectrum generation model to generate normalized features based on the intra-group features and the inter-group features; generating a fingerprint spectrum of the target type of lipids based on the normalized features.

[0009] In some embodiments, generating a fingerprint spectrum of the target type of lipids based on the normalized features includes: using the second activation layer of the fingerprint spectrum generation model to generate an activation feature based on the normalized features; generating a fingerprint spectrum of the target type of lipids based on the activation features.

[0010] In some embodiments, generating a fingerprint spectrum of the target type of lipids according to the activation features includes: using an upsampling layer of a fingerprint spectrum generation model to generate a fingerprint spectrum of the target type of lipids according to the activation features.

[0011] In some embodiments, the resolution of the multiple spectra is a first resolution, the resolution of the fingerprint spectrum is a second resolution, and the second resolution is greater than the first resolution.

[0012] According to a second aspect of the present disclosure, a device for generating a fingerprint spectrum is provided, comprising: an intra-group feature determination module, configured to determine the intra-group features of each spectrum of multiple spectra of a target type of lipid; an inter-group feature determination module, configured to determine the inter-group features of the multiple spectra; and a generation module, configured to generate a fingerprint spectrum of the target type of lipid based on the intra-group features and the inter-group features.

[0013] According to a third aspect of the present disclosure, a device for generating a fingerprint spectrum is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute a method for generating a fingerprint spectrum according to some embodiments of the present disclosure based on instructions stored in the memory.

[0014] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the instructions are executed by a processor, the method for generating a fingerprint spectrum according to some embodiments of the present disclosure is implemented.

[0015] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising computer program instructions, which, when executed by a processor, implement the method for generating a fingerprint spectrum according to some embodiments of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0017] The present disclosure can be more clearly understood from the following detailed description with reference to the accompanying drawings.

[0018] in:

[0019] Figure 1 A flowchart showing a method for generating a fingerprint spectrum according to some embodiments of the present disclosure;

[0020] Figure 2 A schematic diagram showing a model according to some embodiments of the present disclosure;

[0021] Figure 3 A schematic diagram showing a model according to some other embodiments of the present disclosure;

[0022] Figure 4 A block diagram showing a device for generating a fingerprint spectrum according to some embodiments of the present disclosure;

[0023] Figure 5 A block diagram showing a device for generating a fingerprint spectrum according to other embodiments of the present disclosure;

[0024] Figure 6 A block diagram showing a computer system for implementing some embodiments of the present disclosure;

[0025] Figure 7 A user interface of a generation system according to some embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0026] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.

[0027] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0028] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0029] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0030] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0031] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0032] In the related art, the fingerprint library is constructed by obtaining the spectrum of reference standard on a mass spectrometer. Since there are many kinds of lipid substances, the synthesis of reference standard is difficult and expensive, and only a small amount of lipid standards exist, it is difficult to construct a comprehensive lipid molecular fingerprint spectrum.

[0033] Lipids are a class of organic compounds composed of elements such as carbon, hydrogen and oxygen, usually containing long-chain hydrocarbons (fatty acids) and connected with polar groups (hydroxyl, phosphate, etc.). The standard representation of lipids consists of lipid subclass, total carbon number: number of double bonds, and carbon number of subchain: number of double bonds, such as PC 36:3 (PC 18:1_18:2). Due to the high cost of synthesis and the large number of lipids, even the same subclass has tens of millions of different side chain combinations, so there is a lack of comprehensive standards. Because a large number of lipids lack standard products, there is a lack of fingerprint spectra. Usually, the fingerprint spectrum under specific conditions is obtained by mass spectrometry of lipid standards.

[0034] According to the fingerprint spectrum generation method, device and system, and computer-readable storage medium of some embodiments of the present disclosure, the generation of fingerprint spectrum is achieved.

[0035] Figure 1 A flowchart of a method for generating a fingerprint spectrum according to some embodiments of the present disclosure is shown.

[0036] like Figure 1 As shown, the method for generating a fingerprint spectrum includes steps S1 to S3. In some embodiments, the method for generating a fingerprint spectrum is performed by a fingerprint spectrum generation model.

[0037] In step S1, the intra-group features of each of the multiple spectra of the target type of lipids are determined; in step S2, the inter-group features of the multiple spectra are determined; in step S3, a fingerprint spectrum of the target type of lipids is generated based on the intra-group features and the inter-group features.

[0038] For example, if you need to generate a fingerprint spectrum of a target type of lipid, you will obtain the experimental spectrum of the lipid sample of that type. Using the fingerprint spectrum generation model, learn the intra-group features within a single experimental spectrum and the inter-group features between multiple experimental spectra, and generate the fingerprint spectrum of the target type of lipid based on the intra-group features and inter-group features.

[0039] According to some fingerprint spectra generation methods disclosed in the present invention, the generation of fingerprint spectra is achieved by learning spectral features based on an artificial intelligence model, thereby realizing mass spectrometry identification of lipid substances.

[0040] The present disclosure utilizes the Kolmogorov-Arnold representation theorem to realize the generation of fingerprint spectra. The Kolmogorov-Arnold representation theorem indicates that a complex multivariable function can be represented as a combination of multiple single variable functions.

[0041] Lipids have high synthesis costs and a large number of them, and even the same subclass has tens of millions of different side chain combinations, so there is a lack of comprehensive standards. The present disclosure is based on deep learning and reversely constructs fingerprints. Based on the Kolmogorov-Arnold representation theorem, the present disclosure regards the final fingerprint spectrum as a linear representation of multiple experimental spectra, and its principle can be expressed as follows, where Φ(Y) represents the final generated fingerprint spectrum, X represents multiple experimental spectra, W and B represent parameters, and n represents the number of experiments.

[0042]

[0043] The present invention constructs a model for learning to generate fingerprint spectra under specific experimental conditions from input spectra to improve the accuracy and coverage of identification.

[0044] The following first introduces the extraction of intra-group features.

[0045] In some embodiments, determining the intra-group feature of each spectrum of multiple spectra of the target type of lipids includes: extracting the intra-group feature from each fingerprint spectrum using a first activation layer and a linear layer of a fingerprint spectrum generation model.

[0046] For example, the fingerprint spectrum generation model inputs multiple spectra at the same time, and based on the Kolmogrov-Arnold theorem, it realizes the feature extraction of the spectra, thereby realizing the intra-group feature learning. The network mainly learns the intrinsic features of a single spectrum in the input spectrum, maps the secondary spectrum of lipids to a fixed low-dimensional space of 512 rows, and the data is deflected by two Sigmoid activation functions (SiLU) and activated by a rectified linear unit (ReLU) and cut into eight matrices. The formula is as follows:

[0047] H1=SiLU(W1X1+B1)

[0048] H2=SiLU(W2H1+B2)

[0049] H3=ReLU(H2) (1)

[0050] Where X is the input matrix, W i is the weight matrix, B i is the deviation function, H i is the output of the corresponding layer.

[0051] Through the first activation layer and the linear layer, the data is mapped to the potential space dimension, which enables the extraction of data features and solves the input problem of different dimensions.

[0052] The following first introduces the extraction of inter-group features.

[0053] In some embodiments, the inter-group features of the plurality of spectra are determined based on the intra-group features of the plurality of spectra. For example, the inter-group features of the spectra are determined by analyzing and comparing the feature differences between different groups based on the intra-group features of the plurality of spectra.

[0054] In some embodiments, inter-group features of multiple spectra are determined based on intra-group features of multiple spectra, including: using a multi-head attention layer of a fingerprint spectrogram generation model to determine inter-group features of multiple spectra based on intra-group features of multiple spectra.

[0055] The multi-head attention part of the fingerprint spectrogram generation model includes, for example, 6 parallel encoders. Multiple encoders perform inter-group feature learning in parallel, and learn feature information between multiple spectrograms through the global feature learning characteristics of the multi-head self-attention mechanism. This part contains an eight-head self-supervision module and a network layer. The eight matrices after segmentation enter a multi-head self-supervision module respectively, and the formula is as follows.

[0056]

[0057] MultiHead(Q,K,V)=Concat(head1,...,head h )W O (2)

[0058] Where, Q = H3W Q , K=H3W k , V=H3W V , The head i Through the Concat function, the inter-group feature MultiHead (Q, K, V) is obtained.

[0059] In some embodiments, a fingerprint spectrum of lipids of a target type is generated based on intra-group features and inter-group features, including: using a normalization layer of a fingerprint spectrum generation model to generate normalized features based on intra-group features and inter-group features; generating a fingerprint spectrum of lipids of the target type based on the normalized features.

[0060] For example, after completing self-supervised learning, the feature matrix is ​​then subjected to a dropout function (Dropout) for overfitting control and a normalization layer (Norm) for output tidying up. The formula is as follows.

[0061] H4=LayerNorm(H3+Dropout(MultiHead(Q, K, V))) (3)

[0062] In some embodiments, a fingerprint spectrum of a target type of lipid is generated based on normalized features, including: using a second activation layer of a fingerprint spectrum generation model to generate activation features based on the normalized features; and generating a fingerprint spectrum of the target type of lipid based on the activation features.

[0063] For example, the result of the above process is formatted by the rectified linear unit (ReLU), and the formula is as follows.

[0064] H5=LayerNorm(H4+Dropout(ReLU(Layer(H4)))) (4)

[0065] After completing the entire encoder process, the data is processed through an activation function (SiLU) to achieve preliminary resolution improvement:

[0066] H6=SiLU(W3H5+B3) (5)

[0067] In some embodiments, generating a fingerprint spectrum of a target type of lipid based on the activation features includes: using an upsampling layer of a fingerprint spectrum generation model to generate a fingerprint spectrum of the target type of lipid based on the activation features.

[0068] After the above processing, the model has learned the features of multiple spectrograms, and can finally integrate the data and improve the resolution through upsampling and downsampling, as shown below.

[0069]

[0070] In some embodiments, the resolution of the multiple spectra is a first resolution, the resolution of the fingerprint spectrum is a second resolution, and the second resolution is greater than the first resolution. That is, through upsampling, high-resolution fingerprint spectrum regeneration is achieved.

[0071] According to some embodiments of the present disclosure, the Kolmogorov-Arnold representation theorem is combined with a convolutional network to extract local features within the data, and multi-head attention is further used to extract global features between data, thereby improving the feature extraction capability of the model and enabling better prediction results to be obtained in the analysis of complex data.

[0072] According to some embodiments of the present disclosure, a reverse omics model, namely a fingerprint spectrum generation model, is constructed. The fingerprint spectrum generation model extracts fragment information from low-resolution fingerprint spectra and regenerates high-resolution fingerprint spectra by learning intra-group and inter-group features of input spectra.

[0073] In some embodiments, the learning rate of the model is dynamically adjusted according to the different learning efficiencies of different data, thereby optimizing the model training process and significantly shortening the training time. At the same time, a convenient data batch processing function is provided, which is suitable for the processing and analysis of large-scale data.

[0074] Furthermore, according to some embodiments of the present disclosure, the weights of the model can be pre-trained, the model is robust, can be used for data on multiple platforms, and can be easily migrated between platforms.

[0075] Through the trained model, high-precision fingerprint spectra can be generated even when the input accuracy is low or the instrument is poor.

[0076] The fingerprint spectra obtained according to the fingerprint spectrum generation method of some embodiments of the present disclosure (referred to as predicted spectra in the table below) and the experimental results of lipid standards separately sampled on a mass spectrometer (referred to as independently collected spectra in the table below) were compared for spectral similarity, and the results are as follows.

[0077] Table 1

[0078]

[0079]

[0080] It can be seen from the above table that the generated fingerprint spectrum has a high degree of reducibility.

[0081] The method disclosed in the present invention can be applied to process data generated by common mass spectrometers on the market, such as Thermo Scientific Orbitrap Exploris 240MS system, Thermo Scientific Orbitrap Exploris 120MS system, Agilent 6546Q-TOF MS system, Xevo G2-XS Q-TOF MS system, SCIEX ZenoTOF 7600MS system, etc.

[0082] Figure 2 A schematic diagram showing a model according to some embodiments of the present disclosure.

[0083] like Figure 2 As shown, the model includes sub-models 1, 2, and 3, which are used to generate intra-group features, inter-group features, and fingerprint maps, respectively.

[0084] Figure 3 A schematic diagram showing a model according to some embodiments of the present disclosure.

[0085] Figure 3In , w and b represent network parameters, and Relu represents rectified linear unit. Multiple batches of spectrogram data are used as the input of the first network architecture (top). After a two-layer network, the spectrogram data is first mapped to the potential feature space, and then the features of multiple batches of spectrograms are obtained through the relu activation function. The obtained feature results are used as the input of the second network architecture (middle input). After a multi-head self-attention structure, the features between spectrograms are learned and the final output is obtained. The output result is input into the third network architecture (bottom input). After a two-layer network, the feature information is upsampled to a higher dimension, and then the dimensions are merged to finally generate and output a high-precision spectrogram.

[0086] Figure 4 A block diagram of a device for generating a fingerprint spectrum according to some embodiments of the present disclosure is shown.

[0087] like Figure 4 As shown, the fingerprint spectrum generation device includes an intra-group feature determination module 41, an inter-group feature determination module 42, and a generation module 43.

[0088] The intra-group feature determination module 41 is configured to determine the intra-group feature of each of the multiple spectra of the target lipid species, for example, by performing the following steps: Figure 1 Step S1 is shown.

[0089] The inter-group feature determination module 42 is configured to determine the inter-group features of the plurality of spectra, for example, by performing the following steps: Figure 1 Step S2 is shown.

[0090] The generation module 43 is configured to generate a fingerprint spectrum of the lipid of the target type according to the intra-group characteristics and the inter-group characteristics, for example, by performing the following steps: Figure 1 Step S3 shown.

[0091] In some embodiments, the intra-group feature determination module 41 is further configured to extract the intra-group feature from each fingerprint spectrum using the first activation layer and the linear layer of the fingerprint spectrum generation model.

[0092] In some embodiments, the inter-group feature determination module 42 is further configured to determine the inter-group features of the plurality of spectra according to the intra-group features of the plurality of spectra.

[0093] In some embodiments, the inter-group feature determination module 42 is further configured to: utilize the multi-head attention layer of the fingerprint spectrogram generation model to determine the inter-group features of the multiple spectrograms based on the intra-group features of the multiple spectrograms.

[0094] In some embodiments, the generation module 43 is further configured to: generate normalized features based on intra-group features and inter-group features using a normalization layer of the fingerprint spectrum generation model; and generate a fingerprint spectrum of the target type of lipids based on the normalized features.

[0095] In some embodiments, the generation module 43 is further configured to: generate activation features based on the normalized features using the second activation layer of the fingerprint spectrum generation model; and generate a fingerprint spectrum of the target type of lipids based on the activation features.

[0096] In some embodiments, the generation module 43 is further configured to: generate a fingerprint spectrum of the target type of lipids according to the activated features using the upsampling layer of the fingerprint spectrum generation model.

[0097] In some embodiments, the resolution of the multiple spectra is a first resolution, the resolution of the fingerprint spectrum is a second resolution, and the second resolution is greater than the first resolution.

[0098] Figure 5 A block diagram of a device for generating a fingerprint spectrum according to some other embodiments of the present disclosure is shown.

[0099] like Figure 5 As shown, the fingerprint spectrum generating device 5 includes a memory 51; and a processor 52 coupled to the memory 51, and the memory 51 is used to store instructions for executing the fingerprint spectrum generating method. The processor 52 is configured to execute the fingerprint spectrum generating method in any of the embodiments of the present disclosure based on the instructions stored in the memory 51.

[0100] Figure 6 A block diagram of a computer system for implementing some embodiments of the present disclosure is shown.

[0101] like Figure 6 As shown, the computer system 60 may be embodied in the form of a general purpose computing device. The computer system 60 includes a memory 610, a processor 620, and a bus 600 that connects the various system components.

[0102] The memory 610 may include, for example, a system memory, a non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs. The system memory may include a volatile storage medium, such as a random access memory (RAM) and / or a cache memory. The non-volatile storage medium may store, for example, instructions for executing the method for generating a fingerprint spectrum in any of the embodiments of the present disclosure. The non-volatile storage medium includes, but is not limited to, a disk memory, an optical memory, a flash memory, etc.

[0103] The processor 620 can be implemented by a general processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistors, etc. Discrete hardware components. Accordingly, each module such as the judgment module and the determination module can be implemented by a central processing unit (CPU) running instructions in a memory that execute corresponding steps, or can be implemented by a dedicated circuit that executes corresponding steps.

[0104] The bus 600 may use any of a variety of bus architectures, including, but not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, and a Peripheral Component Interconnect (PCI) bus.

[0105] The computer system 60 may also include an input / output interface 630, a network interface 640, a storage interface 650, etc. These interfaces 630, 640, 650, the memory 610, and the processor 620 may be connected via a bus 600. The input / output interface 630 may provide a connection interface for input / output devices such as a display, a mouse, and a keyboard. The network interface 640 may provide a connection interface for various networked devices. The storage interface 650 may provide a connection interface for external storage devices such as a floppy disk, a USB flash drive, and an SD card.

[0106] According to some embodiments of the present disclosure, a system for generating a fingerprint spectrum is also provided, including: a device for generating a fingerprint spectrum according to some embodiments of the present disclosure.

[0107] Figure 7 A user interface of a generation system according to some embodiments of the present disclosure is shown.

[0108] Combine the following Figure 7 , which introduces the operations of the user interface.

[0109] First, organize the secondary spectra of 15 standard reference lipids in multiple samples into a matrix format and save them as a csv file. Then, enter the LipidIN-ReverseLipidomics module of the generation system, left-click analysis, upload the csv file, put it in the data folder, compress it to get data.zip, left-click upload, select data.zip to upload, and wait for the upload to complete. Left-click working, wait for the operation to end, left-click operation records to find result.zip, and left-click download to get the operation results.

[0110] Here, various aspects of the present disclosure are described with reference to flowcharts and / or block diagrams of methods, devices, and computer program products according to embodiments of the present disclosure. It should be understood that each frame of the flowchart and / or block diagram and the combination of frames can be implemented by computer-readable program instructions.

[0111] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable device to produce a machine, so that the processor executes the instructions to produce means for implementing the functions specified in one or more blocks in the flowchart and / or block diagram.

[0112] These computer-readable program instructions can also be readable and stored in a computer-readable memory, which cause the computer to work in a specific manner to produce a product, including instructions for implementing the functions specified in one or more blocks in the flowchart and / or block diagram.

[0113] The present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects.

[0114] The generation of fingerprint spectra is achieved through the generation method, system and device of fingerprint spectra and computer-readable storage medium in the above embodiments.

[0115] So far, the method and device for generating a fingerprint spectrum according to the present disclosure, and the computer-readable storage medium have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

Claims

1. A method for generating a fingerprint spectrum, comprising: determining within-group features of each spectrum of the plurality of spectra of the target species of lipids; determining inter-group characteristics of the plurality of spectra; A fingerprint spectrum of the lipids of the target type is generated according to the intra-group features and inter-group features.

2. The method for generating a fingerprint spectrum according to claim 1, wherein: The step of determining the intra-group features of each spectrum of the plurality of spectra of the target lipid species comprises: The intra-group features are extracted from each fingerprint spectrum using the first activation layer and the linear layer of the fingerprint spectrum generation model.

3. The method for generating a fingerprint spectrum according to claim 1, wherein: Determining the inter-group features of the plurality of spectra includes: According to the intra-group features of the plurality of spectra, inter-group features of the plurality of spectra are determined.

4. The method for generating a fingerprint spectrum according to claim 3, wherein: Determining the inter-group features of the plurality of spectra according to the intra-group features of the plurality of spectra includes: The multi-head attention layer of the fingerprint spectrogram generation model is used to determine the inter-group features of the multiple spectrograms according to the intra-group features of the multiple spectrograms.

5. The method for generating a fingerprint spectrum according to claim 1, wherein: The step of generating a fingerprint spectrum of the lipids of the target type according to the intra-group features and the inter-group features comprises: Generate normalized features according to the intra-group features and inter-group features by using a normalization layer of the fingerprint spectrum generation model; Based on the normalized features, a fingerprint spectrum of the lipids of the target type is generated.

6. The method for generating a fingerprint spectrum according to claim 5, wherein: Generating a fingerprint spectrum of the target lipid species according to the normalized features comprises: Generate an activation feature based on the normalized feature using the second activation layer of the fingerprint spectrum generation model; Based on the activation features, a fingerprint spectrum of the target type of lipids is generated.

7. The method for generating a fingerprint spectrum according to claim 6, wherein: The step of generating a fingerprint spectrum of the lipid of the target type according to the activation feature comprises: The fingerprint spectrum of the target lipid species is generated according to the activation features using the upsampling layer of the fingerprint spectrum generation model.

8. The method for generating a fingerprint spectrum according to claim 7, wherein: The resolutions of the multiple spectra are a first resolution, the resolution of the fingerprint spectrum is a second resolution, and the second resolution is greater than the first resolution.

9. A fingerprint spectrum generating device, comprising: an intra-group feature determination module configured to determine an intra-group feature of each of a plurality of spectra of a target type of lipid; An inter-group feature determination module, configured to determine the inter-group features of the plurality of spectra; A generating module is configured to generate a fingerprint spectrum of the lipids of the target type according to the intra-group features and inter-group features.

10. A fingerprint spectrum generating device, comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the method for generating a fingerprint spectrum according to any one of claims 1 to 8 based on instructions stored in the memory.

11. A fingerprint spectrum generation system, comprising: A device for generating a fingerprint spectrum according to claim 9 or 10.

12. A computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implements the method for generating a fingerprint spectrum according to any one of claims 1 to 8.

13. A computer program product, comprising computer program instructions, which, when executed by a processor, implement the method for generating a fingerprint spectrum according to any one of claims 1 to 8.

Images