Dual-modality imaging method and system based on photoacoustic spectroscopy and laser mass spectrometry

Abstract

The application relates to the field of fusion imaging technology, in particular to a bimodal imaging method and system based on photoacoustic spectroscopy and laser mass spectrometry, the method comprising the following steps: two solid-state lasers in a time-of-flight mass spectrometer emit pulsed laser beams in sequence, the first pulsed laser beam excites photoacoustic signals to obtain spatial structure information of a sample, and the first pulsed laser beam also serves as desorption laser to promote matrix and sample molecules to separate to form neutral molecule vapor cloud; the second pulsed laser beam emitted after a time delay is converted into vacuum ultraviolet laser through a gas cell three times frequency conversion, the neutral molecule vapor cloud is precisely ionized, and qualitative and positioning analysis of molecular composition are completed in combination with time-of-flight mass spectrometry. The system realizes automatic scanning of the sample through a three-dimensional platform sample feeder, a data processing unit synchronously collects and fuses photoacoustic time-domain signals and mass spectrometry ion signals, and finally generates a bimodal image with structure details and molecular distribution characteristics; the application can realize synchronous multi-dimensional analysis of the sample.

Description

Technical Field

[0001] This invention relates to the field of fusion imaging technology, specifically to a dual-modal imaging method and system based on photoacoustic spectroscopy and laser mass spectrometry. Background Technology

[0002] Imaging technology extracts spatially resolved information through the interaction of physical signals (sound, light, electricity, magnetism) with biological tissues and material media. Development in this field relies on interdisciplinary innovation in physics, chemistry, engineering, and information science, with the goal of achieving multi-scale, high-precision, and dynamic visualization from macroscopic to microscopic levels, and from structure to function. These technologies capture and process information using different physical principles and algorithms to obtain clearer and more detailed images.

[0003] Current imaging technology suffers from problems such as single detection dimension, low system integration, low sample utilization, insufficient resolution and sensitivity, and radiation risk. Traditional single-modal imaging technology has many inherent limitations, making it difficult to simultaneously correlate macroscopic structures and microscopic molecular composition, resulting in fragmented imaging information. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides a dual-modal imaging method and system based on photoacoustic spectroscopy and laser mass spectrometry. The aim is to integrate the high-resolution structural imaging capabilities of photoacoustic spectroscopy with the molecular recognition advantages of laser mass spectrometry to construct an integrated detection platform, enabling simultaneous multi-dimensional analysis of samples.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] On one hand, embodiments of the present invention provide a dual-modal imaging method based on photoacoustic spectroscopy and laser mass spectrometry, the method comprising the following steps:

[0007] After processing, the sample is fixed onto the matrix, and the mass spectrometer and three-dimensional platform sampler are turned on.

[0008] The matrix with the sample fixed is fixed on the injection rod of the mass spectrometer. The position of the injection rod is adjusted, the sampling and detection area is selected, and the sample enters the laser-equipped time-of-flight mass spectrometer for detection through the injection system.

[0009] The two solid-state lasers in the time-of-flight mass spectrometer emit pulsed lasers in sequence. The first pulsed laser beam is focused by the optical path system and enters the ionization chamber. It is then focused onto the sample by the measurement objective. The sample generates sound waves as it is excited by the laser. The sound signal is converted into an electrical signal by the ultrasonic probe. At the same time, the first pulsed laser beam acts as a desorption laser, causing the matrix to detach from the sample molecules and form a neutral molecular vapor cloud.

[0010] The second pulsed laser beam is emitted after a delay. After being tripled in the gas cell, it is converted into a vacuum ultraviolet laser. After being focused, it enters the ionization chamber and intersects with the first pulsed laser beam to ionize the neutral molecular vapor cloud. The ionized ions fly through the flight field and are detected by the microchannel plate.

[0011] The sample is scanned and detected as a whole. The photoacoustic time domain signal is collected by the transducer. After preprocessing, it is imported into the data processing unit to obtain photoacoustic data. After processing the photoacoustic data, the maximum value of the envelope signal is extracted and mapped to the photoacoustic image.

[0012] The signal detected by the microchannel plate is converted into an electrical signal by an oscilloscope, and the electrical signal is converted into a grayscale or pseudo-color image to form a molecular distribution map.

[0013] By fusing and analyzing the photoacoustic imaging map and the molecular distribution map, multimodal imaging information of the focused spot micro-region can be obtained.

[0014] Optionally, the step of sequentially emitting pulsed laser light from the two solid-state lasers in the time-of-flight mass spectrometer includes:

[0015] The solid-state laser emits pulsed laser light to generate the first pulsed laser beam. The first pulsed laser beam is focused by the constructed optical path system and enters the ionization chamber. It is then focused onto the sample under test by the measuring objective lens. The focusing direction is opposite to the direction of the ultrasonic probe. The sample generates sound waves as it is excited by the laser. The sound waves are detected by the ultrasonic probe and the sound signal is converted into an electrical signal.

[0016] Meanwhile, the first pulsed laser beam serves as the desorption laser in post-laser ionization mass spectrometry imaging, enabling optical path multiplexing. Through the desorption process, the matrix and sample molecules detach from the target surface and form a neutral molecular vapor cloud.

[0017] Optionally, the delayed emission of the second pulsed laser, after being third-harmonicized in a gas cell and converted into a vacuum ultraviolet laser, is focused and enters the ionization chamber where it intersects with the first pulsed laser, including:

[0018] A second pulsed laser is emitted after a delay following the desorption process. The second pulsed laser is converted into a vacuum ultraviolet laser after being tripled in the gas cell. The gas in the gas cell is a mixture of argon and xenon.

[0019] The first pulsed laser and the vacuum ultraviolet laser are separated by focusing with a plano-convex mirror at the light outlet of the gas cell. After being focused, the vacuum ultraviolet laser enters the ionization chamber parallel to the sample feed rod and intersects with the first pulsed laser.

[0020] Optionally, the overall scanning detection of the sample includes:

[0021] Select the target point, and let the electronic device control the three-dimensional stage to move the sample under test up and down near the focal point of the measuring objective lens. Starting from the marked point, first scan forward one column and then automatically move one step to the right, and then scan backward one column until all the test points on the sample under test are measured.

[0022] Optionally, the step of extracting the maximum value of the envelope signal after photoacoustic data processing and mapping it to a photoacoustic image includes:

[0023] The photoacoustic data is converted into a time-domain signal. The time-domain signal segment within a specified time window is extracted by the data processing unit. A rectangular window function is used to perform time-domain bandpass filtering on the time-domain signal segment to obtain the filtered time-domain signal.

[0024] The filtered time-domain signal is subjected to Hilbert transform to construct an analytic signal. The magnitude of the analytic signal is calculated as the envelope signal. The peak value of the envelope signal is mapped to a photoacoustic image according to its spatial location.

[0025] Optionally, the step of converting the electrical signal into a grayscale or pseudocolor image to form a molecular distribution map includes:

[0026] Calculate the peak area of ​​the signal peak where the target substance is located on the spectrum, convert the peak area at different acquisition positions into a matrix, and then use different colors to represent the values ​​in the matrix to generate grayscale or pseudo-color images to form a molecular distribution map.

[0027] Optionally, the step of fusing and analyzing the photoacoustic imaging map and the molecular distribution map to obtain multimodal imaging information of the focused spot micro-region includes:

[0028] Based on spatial registration algorithm and multimodal fusion strategy, target points of photoacoustic imaging map and slice anatomical landmarks of molecular distribution map are extracted. The photoacoustic imaging map and molecular distribution map are aligned by combining affine transformation matrix. The molecular distribution map is resampled by bilinear interpolation according to the transformation matrix and spatially aligned with the photoacoustic imaging map to obtain bimodal data.

[0029] Structural boundary information of the sample is extracted from the photoacoustic image, and molecular composition information of the sample is extracted from the molecular distribution map. The dual-modal data are superimposed to generate a dual-modal fused image that combines structural and molecular features.

[0030] On the other hand, embodiments of the present invention provide a dual-modal imaging system based on photoacoustic spectroscopy and laser mass spectrometry, comprising:

[0031] A sample fixation device is used to fix processed samples onto a matrix;

[0032] Mass spectrometer and three-dimensional platform sampler are used for sample introduction and detection;

[0033] The injection rod is used to hold the matrix containing the sample.

[0034] Two solid-state lasers are used to emit pulsed laser light;

[0035] Optical path system, used for focusing and transmitting laser light;

[0036] Ionization chamber, used for the interaction between laser and sample and the generation of ions;

[0037] An ultrasonic probe is used to detect the sound waves generated by a sample and convert them into electrical signals.

[0038] Microchannel plates are used to detect ions generated during ionization.

[0039] The data processing unit is used to process the acquired photoacoustic and mass spectrometry signals;

[0040] Electronic equipment used to fuse and analyze photoacoustic images with molecular distribution maps to obtain multimodal imaging information of the focused spot micro-region.

[0041] The beneficial effects of this invention are as follows: This invention discloses a dual-modal imaging method and system based on photoacoustic spectroscopy and laser mass spectrometry. By organically combining photoacoustic spectroscopy and laser mass spectrometry, this invention achieves optical path multiplexing design. The first pulsed laser beam serves simultaneously as the excitation source for photoacoustic imaging and the desorption laser for mass spectrometry imaging, effectively simplifying the system structure and improving the utilization rate of light energy. The introduction of vacuum ultraviolet laser enables efficient ionization of neutral molecular vapor clouds. Combined with the high-sensitivity detection of microchannel plates, this ensures the accuracy of molecular distribution information acquisition. The dual-modal image fusion analysis strategy breaks through the limitations of traditional single-imaging techniques. Through spatial registration and information superposition, the obtained multimodal imaging information of the focused spot micro-area can clearly present the microstructural features of the sample and accurately reflect the spatial distribution of molecules. This provides more comprehensive and reliable visual data support for the study of pathological mechanisms in biomedical samples and the analysis of micro-area components of materials, helping to promote in-depth exploration at the microscale in related fields. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating a dual-modal imaging method based on photoacoustic spectroscopy and laser mass spectrometry in an embodiment of the present invention.

[0044] Figure 2 This is an overall structural diagram of the vacuum imaging system in an embodiment of the present invention;

[0045] Figure 3 This is an overall structural diagram of the atmospheric pressure imaging system in an embodiment of the present invention;

[0046] Figure 4 This is a structural diagram of the time-of-flight mass spectrometer in an embodiment of the present invention;

[0047] Figure 5 This is an algorithm flowchart of the data processing unit in an embodiment of the present invention. Detailed Implementation

[0048] The following will provide a clear and complete description of the concept, specific structure, and technical effects of the present invention in conjunction with embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of the present invention. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0049] The technical terms used in this application will be explained below:

[0050] Laser Desorption / Post-Ionization (LDPI-MSI) is a mass spectrometry technique that combines laser desorption with a subsequent ionization step. It aims to improve ionization efficiency and detection sensitivity, and is particularly suitable for the deep analysis of substances that are difficult to ionize or have low abundance. Its core principle is to separate the sample desorption and ionization processes, optimizing analytical performance through the synergistic effect of two pulsed laser beams.

[0051] Desorption effect: In laser desorption / ionization technology, the desorption effect specifically refers to the process by which the matrix and sample molecules detach from the target surface and form a neutral molecular vapor cloud when the first pulsed laser acts on the sample. This process mainly produces neutral molecules rather than ions. This step is the basis for subsequent secondary ionization, and its efficiency directly affects the overall detection sensitivity.

[0052] Ionization technology: In time-of-flight mass spectrometry, ionization technology refers to the method of converting molecules or atoms in the sample to be tested into charged ions (ionization). This is the second step of LDPI-MSI and the core element that determines the detection sensitivity, resolution and applicability.

[0053] Electrospray desorption-ionization mass spectrometry (DESI-MSI) is a widely used atmospheric pressure imaging technique. Its core principle is to utilize the high focusing power of a laser to desorb molecules from the sample surface, then use the soft ionization properties of electrospray to convert the desorbed molecules into gaseous ions, which are subsequently detected by a mass spectrometer. Compared to traditional ionization methods, the controllability of the reaction is greatly improved.

[0054] MALDI imaging technology, short for matrix-assisted laser desorption / ionization mass spectrometry (MALDI-MSI), is a label-free mass spectrometry technique that can directly visualize the spatial distribution of molecules in tissues or cells in situ. Its core principle can be summarized as follows: biological tissue sections are co-crystallized with a matrix, and the matrix is ​​irradiated with a laser to absorb energy and ionize sample molecules. The mass-to-charge ratio and spatial coordinates of the ions are detected by a mass spectrometer, and finally, a spatial distribution image of the molecules is reconstructed. It is suitable for analyzing biological macromolecules and is currently the most commonly used technique in the field of mass spectrometry (MSI), especially in the study of molecular distribution in tissue sections.

[0055] Quenching effect: refers to the phenomenon where a process (such as luminescence, combustion, chemical reaction, etc.) is weakened or terminated due to external interference. In the field of fluorescence imaging, it refers to the phenomenon where the fluorescence signal is weakened or disappears due to interference from specific physical or chemical factors. Quenching leads to signal attenuation and a decrease in signal-to-noise ratio, and quenching must be suppressed to ensure image quality.

[0056] Time-domain signals: Time-domain signals are a core concept in signal processing, referring to mathematical functions or physical quantities that describe the change of signal amplitude over time, with time as the independent variable. Essentially, they directly observe the dynamic behavior of signals on the time axis and are the most intuitive representation of signals.

[0057] In related technologies, the imaging field faces the problems of "single detection dimension" and "low system integration." It requires high-resolution, multi-functional imaging of objects using highly integrated devices with minimal sample destructiveness. Currently, imaging technologies widely used in biomedical and non-life science research have several key limitations that restrict their ability to comprehensively understand the structural and functional information of samples. The shortcomings of existing technologies are mainly concentrated in the following four aspects:

[0058] The limitations of traditional single-dimensional imaging techniques are as follows: Traditional single-imaging technologies often provide only limited and partial local information, making it difficult to comprehensively analyze the complex mechanisms of life activities and achieve multi-scale dynamic observation from macroscopic structures to microscopic molecules. Photoacoustic imaging acquires the physical properties of deep tissues, such as structural deformation, material defects, thermoelastic parameters, and blood vessel distribution, but it cannot analyze specific molecular composition or chemical components. It lacks the ability to distinguish substances with similar chemical compositions and light absorption, and it lacks explanations of molecular mechanisms. Mass spectrometry can accurately identify molecular composition, i.e., the mass-to-charge ratio (m / z), and can analyze the components of pollutants, materials, and cells, but it generally uses a method of calibration at a specific point with a specific substance, lacking spatial positioning capabilities and correlation with physical environmental parameters. GC-MS / LC-MS (chromatography-mass spectrometry) only analyzes homogeneous samples and lacks spatial information. Other imaging techniques such as ultrasound and magnetic resonance imaging struggle to simultaneously capture the tissue structure, functional state, and molecular composition information of samples. Multimodal imaging, as a key imaging technology for enriching sample information, is becoming an important direction for development.

[0059] Low system integration: The three core problems caused by independent operation of equipment in existing technologies are: sample transfer damage, traditional processes require sample transfer across devices, which can easily cause dehydration, oxidation or structural damage, resulting in distortion of trace molecular data; spatial registration bottleneck, existing combined technologies require switching devices, resulting in registration errors and making it impossible to guarantee the spatial consistency of data in the same area, and instrument switching leads to low spatial registration accuracy; independent systems require multiple laser sources, which have low energy conversion efficiency and complex optical path calibration, and the equipment is bulky and time-consuming to operate.

[0060] Highly invasive and destructive: Current multimodal optical imaging technologies combined with MSI, such as fluorescence imaging, typically require exogenous labeling or genetic modification of samples. Fluorescent labeling targets molecules, interfering with the natural molecular state and failing to cover unknown targets. Furthermore, sample processing is complex, requiring multiple slicing, fixation, labeling, matrix coating, and staining processes, leading to the loss / degradation of small molecule metabolites, easily introducing errors. Staining and mounting media may contaminate samples, suppressing mass spectrometry signals and potentially generating false positive signals due to non-specific binding or signal attenuation due to quenching effects. Simultaneously, spectral overlap of different fluorescence channels during multicolor labeling can interfere with the accurate localization of target molecules. Moreover, strong laser excitation and high-intensity illumination may cause photobleaching and phototoxicity, damaging or altering the physiological state of live samples, making them unsuitable for long-term dynamic observation or studies requiring the preservation of sample integrity.

[0061] Radiation risks: Imaging technologies with strong penetrating power, such as computed tomography (CT) and positron emission tomography (PET), pose a potential risk of damage to living organisms or in situations requiring repeated imaging due to their ionizing radiation characteristics, which limits their widespread application frequency and scenarios.

[0062] To address the aforementioned technical bottlenecks, this invention proposes a dual-modal imaging method and system based on photoacoustic spectroscopy and laser mass spectrometry, whose design aims to systematically solve these key problems:

[0063] To address the issue of limited imaging information dimensions, this invention proposes a combined technique of photoacoustic imaging (PAI) and mass spectrometry imaging (MSI). By integrating PAI and MSI, this technique enables the collaborative acquisition of multimodal information from the same sample, achieving multi-dimensional information fusion and resolving the information gaps associated with single imaging modalities. This allows for comprehensive multi-dimensional characterization from macroscopic structures to microscopic molecules. Through deep multimodal collaboration, this invention significantly improves overall imaging depth, targeting accuracy, and molecular resolution capabilities. The entire system integrates structural imaging and molecular identification, acquiring multi-dimensional information and achieving multi-dimensional information fusion, providing an innovative tool for cross-scale imaging research of complex systems.

[0064] To address the issue of low system integration, this invention innovatively integrates photoacoustic imaging (PAI) and mass spectrometry imaging (MSI) systems coaxially to achieve dual-modal imaging on a single platform. A vacuum-sealed cavity is proposed that is compatible with both the PAI laser window and the MSI ion source, supporting simultaneous imaging of the same sample in a vacuum environment and solving the spatial registration problem. The core breakthrough lies in embedding the ultrasonic transducer into the mass spectrometer cavity. An atmospheric pressure platform is proposed that is compatible with both PAI and MSI detection and imaging systems, and a single laser source is reused to simultaneously drive both imaging modes. Combined with a three-dimensional translation stage, in-situ scanning of the entire sample is achieved. First, single-shot in-situ detection eliminates dehydration, oxidation, or structural damage caused by sample transfer across devices, ensuring the integrity and reliability of trace molecular data. Second, the spatial registration bottleneck is overcome by utilizing coaxial optical path design and hardware integration, ensuring spatial consistency of dual-modal data at nanometer-micrometer level precision. Finally, the hardware structure is simplified through a laser reuse mechanism, avoiding energy loss from multiple laser sources and complex optical path calibration, significantly reducing equipment size, time costs, and operational complexity. This integrated system, with "hardware fusion + light source reuse" as its core, retains the advantages of PAI deep penetration and MSI high specificity molecular resolution while achieving a synergistic leap in data fidelity, spatial matching accuracy and system efficiency.

[0065] To address the issues of sample destructiveness and low utilization, the PAI mentioned in this invention is completely label-free, avoiding contamination of the mass spectrometry signal by staining agents and mounting media, and preserving the original molecular state. Simultaneously, PAI can capture the tissue structure of the sample, avoiding blind sampling by MSI. PAI utilizes non-invasive pulsed laser excitation to generate signals for imaging, avoiding the dependence on fluorescent labels and the resulting photodamage problems in fluorescence imaging, ensuring integrity, saving sample processing time, and improving detection efficiency. Although MSI may cause micro-destruction to the sample during sample pretreatment, it significantly reduces the damage to sample integrity compared to continuous fluorescence damage, especially beneficial for high-throughput, multi-round detection of ex vivo samples. This invention proposes PAI to guide MSI from "blind sampling" to "targeted deep analysis," cleverly combining the advantages of PAI's high contrast and MSI's good imaging penetration depth, improving deep tissue resolution, avoiding subjective bias of researchers, avoiding the low sample utilization and wasted time of full-sample scanning caused by blind MSI sampling, reducing sample damage, and improving data reproducibility.

[0066] To address the issues of insufficient resolution and sensitivity, this invention effectively improves the resolution of signal sampling and imaging through the selection of hardware modules and algorithm processing. On the hardware side, amplifiers and filters are added to amplify the ultrasonic signal, attenuate high-frequency noise, prevent sampling distortion, and effectively separate noise from photoacoustic signals, resulting in a purer signal. On the algorithm side, Hilbert transform is used to extract the envelope of the time-domain signal, improving the detection capability of weak signals. By adding a signal denoising technique based on singular value decomposition (SVD), the signal is filtered twice, effectively separating high-frequency noise from low-frequency photoacoustic signals, suppressing noise to achieve low background interference, and enhancing the signal-to-noise ratio. An independently developed algorithm based on Python and Matlab converts and stitches the mass spectrometry signal of each pixel into a two-dimensional mass distribution map, i.e., a mass spectrometry imaging map, where the relative intensity of ions at each pixel is represented by different shades of color. These hardware and algorithm improvements effectively overcome the bottleneck in imaging technology regarding resolution.

[0067] This invention utilizes a non-ionizing light source to avoid radiation risks. The core excitation source of PAI-MSI is non-ionizing radiation (pulsed laser within a safe wavelength range), which essentially avoids the risks posed by ionizing radiation such as X-rays or gamma rays. Therefore, the PAI-MSI combined technology provides a safer solution for research requiring high-precision, deep, molecular-level imaging, especially in biological tissue applications or radiation-sensitive scenarios.

[0068] In summary, the PAI-MSI technology of this invention, through innovative multi-dimensional information fusion, aims to overcome the main limitations of existing imaging technologies, such as limited information dimensions, low system integration, low sample utilization, insufficient resolution and sensitivity, and radiation risks. It can solve core problems such as multi-dimensional imaging sample information integration and molecular dynamic visualization. Through a technological breakthrough path of expanding the dimensions of imaging information, and with the help of an integrated hardware platform, it provides a more comprehensive, refined, and safer analytical tool for imaging research.

[0069] refer to Figure 1 ,like Figure 1 The image shows a dual-modal imaging method based on photoacoustic spectroscopy and laser mass spectrometry provided by an embodiment of the present invention. The method includes the following steps:

[0070] S100: After processing the sample, fix it on the matrix and turn on the mass spectrometer and the three-dimensional platform sampler.

[0071] S200: The matrix with the sample fixed is fixed on the injection rod of the mass spectrometer. The position of the injection rod is adjusted, the sampling and detection area is selected, and the sample enters the time-of-flight mass spectrometer with laser through the injection system for detection.

[0072] S300 enables two solid-state lasers in the time-of-flight mass spectrometer to emit pulsed lasers in sequence. The first pulsed laser beam is focused by the optical path system and enters the ionization chamber, and is then focused onto the sample by the measurement objective. The sample generates sound waves as it is excited by the laser. The sound signal is converted into an electrical signal by the ultrasonic probe. At the same time, the first pulsed laser beam acts as a desorption laser, causing the matrix to detach from the sample molecules and form a neutral molecular vapor cloud.

[0073] The S400 emits a second pulsed laser beam with a delay. After being tripled in the gas cell, it is converted into a vacuum ultraviolet laser. After being focused, it enters the ionization chamber and intersects with the first pulsed laser beam to ionize the neutral molecular vapor cloud. The ionized ions fly through the flight field and are detected by the microchannel plate.

[0074] S500 performs overall scanning detection of the sample, acquires photoacoustic time-domain signals through transducers, imports them into the data processing unit after preprocessing to obtain photoacoustic data, and extracts the maximum value of the envelope signal after processing the photoacoustic data to map it into a photoacoustic image.

[0075] S600: The signal detected by the microchannel board is converted into an electrical signal by an oscilloscope, and the electrical signal is converted into a grayscale or pseudo-color image to form a molecular distribution map.

[0076] The S700 integrates photoacoustic imaging with molecular distribution maps to obtain multimodal imaging information of the focused spot micro-region.

[0077] This invention provides a dual-modal imaging method based on photoacoustic spectroscopy and laser mass spectrometry. This method employs two solid-state lasers working collaboratively. The first pulsed laser not only excites photoacoustic signals to acquire spatial structural information of the sample but also acts as a desorption laser to cause the matrix to detach from sample molecules, forming a neutral molecular vapor cloud. The second pulsed laser, emitted with a delay, is converted into a vacuum ultraviolet laser via a gas cell after third harmonic generation. This laser precisely ionizes the neutral molecular vapor cloud, and combined with time-of-flight mass spectrometry, it performs qualitative and localization analysis of molecular components. The system automates sample scanning via a three-dimensional platform sampler. The data processing unit simultaneously acquires and fuses photoacoustic time-domain signals and mass spectrometry ion signals, ultimately generating a dual-modal image that combines structural details and molecular distribution characteristics. This effectively overcomes the technical bottleneck of traditional single-modal imaging, providing a novel analytical tool for in-depth research on biomedical samples.

[0078] In this embodiment, the spatiotemporal precision of the desorption and ionization processes is achieved through the coordinated operation of two solid-state lasers. The first pulsed laser, acting as the desorption laser, has its wavelength and energy density optimized according to the sample characteristics to ensure efficient desorption of sample molecules while avoiding excessive damage to the sample surface structure and molecular integrity. After being precisely focused by optical components such as mirrors and lens groups in the optical path system, this laser can form micron- or even submicron-sized spots on the sample surface, thereby achieving precise desorption of micro-regions of the sample and laying the foundation for subsequent high spatial resolution imaging. The time delay control of the second pulsed laser is crucial. By precisely adjusting the delay parameters, it can be ensured that the neutral molecular vapor cloud reaches the optimal density distribution state in the ionization region. At this point, the introduction of a vacuum ultraviolet laser after third harmonic conversion in the gas cell can utilize its high photon energy to achieve efficient soft ionization of neutral molecules. This stepwise desorption and ionization mechanism not only effectively avoids the problems of excessive molecular fragmentation and low ionization efficiency that may occur in traditional single-step ionization processes, but also helps to improve the ion yield and stability, thereby enhancing the sensitivity and signal-to-noise ratio of the entire mass spectrometry detection system. In terms of space, the cross-design of the two laser beams in the ionization chamber ensures that the neutral molecular vapor cloud can interact with the ionization laser to the maximum extent, further improving the ionization efficiency and enabling even trace amounts of analyte molecules to be effectively detected.

[0079] As an improvement to the above embodiment, step S300, which involves causing the two solid-state lasers in the time-of-flight mass spectrometer to sequentially emit pulsed lasers, includes:

[0080] The solid-state laser emits pulsed laser light to generate the first pulsed laser beam. The first pulsed laser beam is focused by the constructed optical path system and enters the ionization chamber. It is then focused onto the sample under test by the measuring objective lens. The focusing direction is opposite to the direction of the ultrasonic probe. The sample generates sound waves as it is excited by the laser. The sound waves are detected by the ultrasonic probe and the sound signal is converted into an electrical signal.

[0081] Meanwhile, the first pulsed laser beam serves as the desorption laser in post-laser ionization mass spectrometry imaging, enabling optical path multiplexing. Through the desorption process, the matrix and sample molecules detach from the target surface and form a neutral molecular vapor cloud.

[0082] In this embodiment, through optical path multiplexing design, the first pulsed laser beam performs the function of desorption laser while simultaneously exciting the photoacoustic signal, eliminating the need for an additional independent desorption light source and significantly simplifying the system's optical path structure. The opposing orientation of the focusing direction and the ultrasonic probe direction ensures that the acoustic signal is efficiently captured by the probe along the shortest path, reducing signal attenuation during propagation and improving the acquisition efficiency and signal-to-noise ratio of the photoacoustic signal. After the ultrasonic probe converts the acoustic signal into an electrical signal in real time, it is amplified by a preamplifier and then filtered by a low-pass filter to remove high-frequency noise. The preprocessed electrical signal is transmitted to an oscilloscope or data acquisition card, providing high-quality raw data for subsequent reconstruction of the photoacoustic image. When used as a desorption laser, its energy density must be strictly controlled within a threshold range. This ensures that the matrix and sample molecules effectively detach from the target surface to form a neutral molecular vapor cloud, while avoiding excessive energy that could lead to molecular structure breakage or sample surface carbonization. By adjusting the laser pulse width and repetition frequency, it can adapt to the desorption requirements of different types of samples (such as biological tissue sections, single-cell samples, etc.), ensuring the efficiency and molecular integrity of the desorption process.

[0083] As an improvement to the above embodiment, in step S400, the second pulsed laser beam is emitted with a delay, converted into a vacuum ultraviolet laser after being third-harmonicized in a gas cell, focused, and then enters the ionization chamber where it intersects with the first pulsed laser beam. This includes:

[0084] A second pulsed laser is emitted after a delay following the desorption process. The second pulsed laser is converted into a vacuum ultraviolet laser after being tripled in the gas cell. The gas in the gas cell is a mixture of argon and xenon.

[0085] The first pulsed laser and the vacuum ultraviolet laser are separated by focusing with a plano-convex mirror at the light outlet of the gas cell. After being focused, the vacuum ultraviolet laser enters the ionization chamber parallel to the sample feed rod and intersects with the first pulsed laser.

[0086] In this embodiment, by precisely controlling the delayed emission time of the second pulsed laser, it is coordinated with the first pulsed laser in timing, ensuring that the neutral molecular vapor cloud reaches its optimal spatial distribution in the ionization chamber before ionization. A mixture of argon and xenon gas is used as the nonlinear optical medium in the gas cell. By precisely controlling the gas ratio and pressure parameters, the laser's third-harmonic conversion can be efficiently achieved, converting the fundamental frequency laser into a vacuum ultraviolet laser. Utilizing the high photon energy characteristics of the vacuum ultraviolet laser, soft ionization of neutral molecules is achieved, effectively reducing the generation of molecular debris and improving ionization efficiency and detection sensitivity. A plano-convex mirror is placed at the light outlet of the gas cell, which not only focuses the converted vacuum ultraviolet laser but also uses its dispersion characteristics to separate the first pulsed laser from the vacuum ultraviolet laser, avoiding mutual interference between the two lasers and ensuring the purity of their respective optical paths. After being focused, the vacuum ultraviolet laser enters the ionization chamber parallel to the direction of the sample feed rod and intersects with the first pulsed laser beam. This spatial intersection design can maximize the interaction volume between the laser and the neutral molecular vapor cloud, allowing the neutral molecules to fully absorb the laser energy and undergo ionization, thereby improving the ion yield and providing a sufficient ion source for subsequent mass spectrometry detection.

[0087] As an improvement to the above embodiment, step S500, which involves performing a comprehensive scanning detection of the sample, includes:

[0088] Select the target point, and let the electronic device control the three-dimensional stage to move the sample under test up and down near the focal point of the measuring objective lens. Starting from the marked point, first scan forward one column and then automatically move one step to the right, and then scan backward one column until all the test points on the sample under test are measured.

[0089] In this embodiment, using preset target point coordinates, an electronic device drives a 3D stage to automatically scan the sample along a zigzag path. During the scanning process, the movement precision of the 3D stage is controlled at the micrometer level, ensuring the spatial accuracy of each sampling point. When the sample moves to the target position, the system triggers a laser emission and signal acquisition process, automatically switching to the next sampling point after completing the detection of the current point, until the entire preset detection area is covered. This automated scanning method not only avoids errors caused by manual operation but also further improves sample utilization and data acquisition speed by optimizing the scanning path (e.g., initially locating regions of high interest based on photoacoustic imaging and prioritizing the scanning of these regions to improve detection efficiency). During the scanning process, the transducer tracks the sample position changes in real time and dynamically adjusts the signal acquisition parameters to ensure that photoacoustic time-domain signals from different regions can be effectively captured.

[0090] As an improvement to the above embodiment, step S500, which involves extracting the maximum value of the envelope signal after photoacoustic data processing and mapping it to a photoacoustic imaging map, includes:

[0091] The photoacoustic data is converted into a time-domain signal. The time-domain signal segment within a specified time window is extracted by the data processing unit. A rectangular window function is used to perform time-domain bandpass filtering on the time-domain signal segment to obtain the filtered time-domain signal.

[0092] The filtered time-domain signal is subjected to Hilbert transform to construct an analytic signal. The magnitude of the analytic signal is calculated as the envelope signal. The peak value of the envelope signal is mapped to a photoacoustic image according to its spatial location.

[0093] In this embodiment, by extracting a time-domain signal segment within a specified time window, the photoacoustic signal generated by the sample can be effectively focused, eliminating interference from background noise and irrelevant signals. Time-domain bandpass filtering using a rectangular window function further highlights the photoacoustic signal characteristics within the target frequency band, suppressing the impact of high-frequency noise and low-frequency interference on signal quality. After time-domain filtering, an analytical signal is constructed from the filtered time-domain signal using Hilbert transform. The envelope signal is extracted by calculating the magnitude of the analytical signal. This envelope signal accurately reflects the amplitude variation characteristics of the photoacoustic signal, providing crucial data support for subsequent imaging. Finally, the extracted envelope signal peaks are mapped according to their corresponding spatial locations to generate a photoacoustic image with high contrast and high spatial resolution, clearly presenting the internal structural information and optical absorption characteristic distribution of the sample.

[0094] As an improvement to the above embodiment, step S600, converting the electrical signal into a grayscale or pseudo-color image to form a molecular distribution map, includes:

[0095] Calculate the peak area of ​​the signal peak where the target substance is located on the spectrum, convert the peak area at different acquisition positions into a matrix, and then use different colors to represent the values ​​in the matrix to generate grayscale or pseudo-color images to form a molecular distribution map.

[0096] In this embodiment, the characteristic ion peaks corresponding to the target molecules are identified by spectral analysis of the ion signals detected by the microchannel plate. The peak area of ​​the signal peak containing the target substance is calculated using an integral algorithm to characterize the relative ion intensity. The peak area data of different sampling points are arranged in spatial coordinates to construct a two-dimensional matrix, where each element corresponds to the molecular abundance of a specific micro-region on the sample surface. The matrix values ​​are converted into a visual image using grayscale mapping or pseudo-color encoding techniques: in grayscale mode, the brightness gradient reflects the difference in ion intensity, while in pseudo-color mode, the hue and saturation changes of colors such as red, green, and blue are used to quantify the molecular distribution. High-intensity areas are highlighted with warm colors, and low-intensity areas are presented with cool colors. Finally, a molecular distribution map that intuitively reflects the spatial distribution characteristics of the target molecules is generated, achieving visualized localization at the molecular level.

[0097] The acquisition location depends on the composition of the target substance, such as proteins, lipids, drug metabolites, alkaloids, etc. Different substances will appear in different positions in the mass spectrum, thus allowing for differentiation and imaging. That is, the molecule to be imaged is first acquired, and its mass-to-charge ratio is calculated to locate its specific position in the mass spectrum.

[0098] As an improvement to the above embodiment, in step S700, the photoacoustic imaging map and the molecular distribution map are fused and analyzed to obtain multimodal imaging information of the focused spot micro-region, including:

[0099] Based on spatial registration algorithm and multimodal fusion strategy, target points of photoacoustic imaging map and slice anatomical landmarks of molecular distribution map are extracted. The photoacoustic imaging map and molecular distribution map are aligned by combining affine transformation matrix. The molecular distribution map is resampled by bilinear interpolation according to the transformation matrix and spatially aligned with the photoacoustic imaging map to obtain bimodal data.

[0100] Structural boundary information of the sample is extracted from the photoacoustic image, and molecular composition information of the sample is extracted from the molecular distribution map. The dual-modal data are superimposed to generate a dual-modal fused image that combines structural and molecular features.

[0101] In this embodiment, the photoacoustic imaging image and molecular distribution map are preprocessed using a spatial registration algorithm. First, target points with obvious structural features (such as tissue edges and vascular branch points) are extracted from the photoacoustic imaging image, and anatomical landmarks with known molecular expression (such as regions with high expression of specific proteins) are extracted from the molecular distribution map. The affine transformation matrix is ​​calculated using the least squares method to achieve precise alignment of the two images in the spatial coordinate system. For the molecular distribution map, bilinear interpolation is used for resampling based on the transformation matrix to ensure that its pixel resolution matches that of the photoacoustic imaging image, eliminating spatial misalignment caused by differences in scanning step size or imaging scale. During the data fusion stage, the system automatically segments the three-dimensional structural boundaries of the samples from the photoacoustic imaging image, including macroscopic morphological information such as cell contours and tissue layers; simultaneously, it extracts microscopic molecular composition data such as the expression intensity and distribution density of specific biomarkers from the molecular distribution map. The structural boundary information and molecular composition data are then superimposed pixel-wise according to preset weights. For example, the photoacoustic imaging image is used as the base layer, and the molecular distribution map is used as a transparent overlay layer. By adjusting the transparency parameters of the overlay layer, structural details and molecular distribution features are clearly presented in the fused image. The resulting dual-modal fusion image not only visually displays the spatial structure of the sample, but also simultaneously reflects the differences in molecular types and contents in different regions, providing multi-dimensional visualization evidence for studying the structure-function relationship of the sample. For example, in tumor tissue imaging, the morphological boundary of the tumor and the expression distribution of cancer cell-specific proteins can be observed simultaneously, which helps in pathological diagnosis and treatment target localization.

[0102] This invention overcomes the physical limitations of single imaging modalities through multimodal detection, system integration, and data fusion, promoting the complementarity of cross-physical mechanism technologies such as photoacoustic-mass spectrometry. Based on photoacoustic effects and ionization mass analysis, it achieves the integration of information across the entire chain, from the functional state of matter to the microscopic molecular composition.

[0103] In terms of detection dimensions: This invention, through the synergy of detection mechanisms, macroscopic localization of light absorption and precise analysis of mass-to-charge ratio, achieves the connection between macroscopic structure and microscopic molecular information, expanding the dimensions of imaging information and providing a "double-insurance" imaging scheme for imaging technology. Driven by the dual engines of physical field perception and molecular fingerprint decoding, a new multimodal imaging model is constructed, advancing the solution to the long-standing problem of integrating multidimensional imaging information in imaging technology, and providing a new research paradigm for analyzing complex system samples.

[0104] In terms of system integration: This invention innovates in aspects such as same-platform detection and coaxial optical path multiplexing, integrating a cavity that is compatible with PAI laser windows and MSI desorption sources, supporting point-by-point sequential imaging of the same sample, achieving single-platform integrated dual-modal imaging, and reducing equipment redundancy and operational complexity. Specifically, it solves the spatial registration problem, correlates multimodal imaging information; eliminates the need to move the sample, avoiding sample waste; simplifies the optical path, achieves optical path multiplexing, optimizes laser efficiency and cost; and improves sensitivity and resolution. This invention is applicable to fields such as life sciences, industrial testing, materials science, environmental monitoring, and cultural heritage protection, and is of groundbreaking significance.

[0105] The present invention solves the following technical problems in the prior art:

[0106] Single detection dimension: This invention solves the problem of missing information in single-modal imaging by combining photoacoustic imaging (PAI) and laser mass spectrometry imaging (MSI), and connects macroscopic structure with microscopic causes to expand the information dimension.

[0107] (1) Technical principle dimension: PAI-MSI technology has unique advantages in structural and functional imaging and molecular chemical information analysis. Its essence is the mathematical fusion of the acoustic wave propagation model based on the photoacoustic wave equation and the ionization model based on the mass spectrometry equation. Its core advantage stems from the deep complementarity of the physical mechanisms of the two technologies. First, the acoustic impedance and specific light wavelength absorption effect of photoacoustic technology reflect the structure of the imaged material. Then, the mass-to-charge ratio of mass spectrometry technology determines its specific composition, so that the structural and compositional signals are coordinated, realizing a multi-scale comprehensive characterization of the sample from macroscopic anatomy to microscopic molecular distribution.

[0108] (2) Spatial Dimension: The PAI-MSI combination unifies "where it occurs" and "what substance it is" within the same sample space, providing an analytical tool for multi-dimensional complex imaging systems. This invention forms a closed loop of PAI depth navigation and MSI molecular tracing in terms of spatial dimension, solving the problem of "where to measure," achieving precise mass spectrometry sampling, and avoiding the problems of low sample utilization and wasted time in full-sample scanning caused by blind MSI sampling. This invention, through dual-mode combination, enhances cross-scale penetration, improves spatial dimension targeting, and resolves the contradiction between imaging depth and molecular specificity.

[0109] (3) Resolution dimension: After the fusion of PAI-MSI technology, the synergy between the detection limit and signal intensity enhances low-abundance signals, solving the problem of "whether it can be detected". This invention can provide structural and functional information with micron-level spatial resolution. Photoacoustic imaging has a resolution of 10~100μm, but it is still difficult to identify low-abundance molecules. MSI spatial resolution is usually 400nm~50μm, and based on the mass-to-charge ratio detection principle, it has high specificity. It can perform precise and highly specific molecular identification through fragment ions, which can solve the ambiguity problem of photoacoustic imaging and make up for the lack of molecular specificity of photoacoustic imaging.

[0110] Low system integration: This invention integrates photoacoustic imaging (PAI) and laser mass spectrometry (MSI) systems on the same platform. For traditional vacuum MSI technology, it innovatively embeds the ultrasonic transducer of photoacoustic imaging into the mass spectrometer; for open-air atmospheric pressure MSI technology, it also achieves dual-mode joint imaging on the same platform under atmospheric pressure. This invention reuses a single laser path to drive dual-modal imaging, combined with a three-dimensional platform to achieve full sample scanning, realizing dual-modal imaging with sample integration on a single platform. This fundamentally solves the three core problems caused by the independent operation of equipment in existing technologies:

[0111] Eliminating sample transfer damage: Traditional coupled imaging techniques require sample transfer across devices, which can easily lead to dehydration, oxidation, or structural damage, resulting in distorted molecular data. This invention avoids sample movement through single-shot, in-situ detection, ensuring data integrity and reliability.

[0112] Overcoming the bottleneck of spatial registration: Existing combined technologies require equipment switching, leading to registration errors and failing to guarantee spatial consistency of data within the same region. Instrument switching also results in low spatial registration accuracy. This invention improves spatial registration accuracy through coaxial optical path design and hardware integration, ensuring precise matching of dual-modal data.

[0113] Optimizing laser efficiency and cost: Independent systems require multiple laser sources, resulting in low energy conversion efficiency, complex optical path calibration, and bulky, time-consuming equipment. This invention significantly reduces time and space costs through laser multiplexing and hardware simplification.

[0114] Low sensitivity and resolution: This invention effectively solves the problems of insufficient resolution and sensitivity through the selection of hardware modules and algorithm processing. On the hardware side, amplifiers and filters are added to amplify the ultrasonic signal, attenuate high-frequency noise, prevent sampling distortion, and effectively separate noise from photoacoustic signals, resulting in a purer signal. On the algorithm side, Hilbert transform is used to extract the envelope of the time-domain signal, improving the detection capability of weak signals; filtering effectively separates high-frequency noise from low-frequency photoacoustic signals, suppresses noise to achieve low background interference, and enhances the signal-to-noise ratio; and independently developed algorithms based on Python and Matlab are used to process and image the acquired signals. These hardware and algorithm improvements effectively overcome the bottleneck in imaging technology regarding resolution.

[0115] PAI-MSI technology represents a paradigm shift in imaging technology, moving from single-modality to multimodal integration and from structural observation to function-molecular correlation analysis. With breakthroughs in key areas such as probe synergy and algorithm optimization, it is expected to become a tool for studying the mechanisms of complex systems, helping researchers efficiently acquire multi-scale and multi-dimensional image data and driving the development of imaging technology towards dynamic, systematic, and quantitative approaches.

[0116] The following is a specific embodiment provided by the present invention:

[0117] The integrated system of this invention can be configured in two ways, respectively realizing vacuum imaging and atmospheric pressure imaging. The vacuum imaging system is generally composed of a photoacoustic imaging subsystem integrated into a mass spectrometry imaging system, while the atmospheric pressure imaging system combines an atmospheric pressure photoacoustic imaging-laser electrospray desorption platform with an ionization mass spectrometry imaging system. (Reference) Figure 2 , Figure 2 In this system, MCP stands for Microchannel Detector; TOF-MS for Time-of-Flight Mass Spectrometer; PAD for Photoacoustic Signal Detector; ER for Repulsion Region; AR for Acceleration Region; DR for Deflection Region; FR for Focusing Region; Valves for Pulse Valve; 3D-Stage for Three-Dimensional Platform; LD for Desorption Laser; LI for Ionization Laser; Pump for Pump; and SPM for Signal Processing Module. The vacuum imaging system includes: a digital time-delay generator, a time-of-flight mass spectrometer (sample introduction system, three-dimensional platform, vacuum system, gas cell, MCP microchannel plate), solid-state lasers (LD desorption laser, LI ionization laser), an optical path system (aperture, focusing lens, acquisition objective, etc.), a signal acquisition unit (oscilloscope, data acquisition card, ultrasonic transducer PAD), and a data processing module (amplification module, filtering module). The mass spectrometer is internally a vacuum system, and the digital time-delay generator, solid-state laser, microchannel plate, signal acquisition unit, and data processing module are sequentially and electrically connected.

[0118] Unlike vacuum imaging systems, open-type atmospheric pressure imaging systems have the sample stage located outside the mass spectrometer. The same platform enables desorption techniques for photoacoustic and laser mass spectrometry. The desorbed particles are then delivered into the mass spectrometer via electrospray or a carrier gas-heated capillary structure to complete the ionization step. The specific structure is as follows: Figure 3 As shown. Figure 3 In this context, IR-Laser represents an infrared laser source; MS inlet represents the mass spectrometer inlet; ESI emitter represents an electrospray emitter; Smaple represents the sample; X-Y translational stage represents the three-dimensional platform; PAD represents the photoacoustic signal detector; and SPM represents the signal processing module.

[0119] Laser: The lasers used in the system of this invention are: an IndiSpectra-Physics laser as the excitation source for photoacoustic imaging and a desorption source for LDPI-MSI; and a Continum miniature Nd:YAG laser Nimma-900 as the ionization source for LDPI-MSI.

[0120] Specifically, for laser ionization, we utilize an Nd:YAG pumped laser as the output laser source and employ a vacuum ultraviolet laser as its ionization source. The advantage of this approach is that the photon energy of the vacuum ultraviolet laser is relatively low, with a single photon energy of 10.5 eV, allowing for "soft" ionization, reducing the generation of fragment ions, and resulting in a simpler peak shape in the mass spectrum. After passing through a third-harmonic crystal, a 355 nm ultraviolet-visible laser is output. This 355 nm light is focused into an Xe-Ar mixer cell using a MgF2 focusing lens. Xenon (Xe) is used as the mixing gas, and argon (Ar) is used as the phase-matching gas. The nonlinear third-harmonic effect of these inert gases generates a 118 nm (10.5 eV) vacuum ultraviolet laser. The generated 118 nm light is then focused onto the center of the ionization chamber by a MgF2 focusing lens.

[0121] Timing control system: The time-of-flight mass spectrometer in this instrument uses the Stanford DG645. The DG645 has 5 output channels. It is used to provide unified external trigger control for the Q-switches of the laser, the lamp, the pulse valve, and the trigger signals of the oscilloscope. Each channel can independently control its turn-on time and pulse width.

[0122] During the experiment, channels AB were used to control the opening time and pulse width of the pulse valve, thereby controlling the amount of sample entering the vacuum chamber. Channels CD and EF were used to control the delay and pulse width of the Lamp and Q switches of the Nd:YAG laser, respectively. Channel GH was used to trigger the oscilloscope. The oscilloscope trigger signal was set to be turned on simultaneously with the laser's Q switch, so that the 0 point on the oscilloscope's horizontal axis represents the laser's emission time.

[0123] Time-of-flight mass spectrometer: Here we use the time-of-flight mass spectrometer (TOF-MS) that our laboratory built itself. This mass spectrometer uses a vacuum ultraviolet laser as an ionization source and uses "soft" ionization technology to ionize the sample to be tested. Essentially, ions are separated in the mass analyzer according to their mass-to-charge ratio (m / z). It has a built-in three-dimensional platform that can realize desorption and ionization of the sample in the entire coordinate system.

[0124] refer to Figure 4 A time-of-flight mass spectrometer (TOF-MS) typically comprises the following components: a vacuum system (pump), a sample introduction system, an extraction region, an acceleration region, a deflection region, a focusing region, a high-voltage power supply system, a field-free flight region, and an ion detector. The sample to be detected is introduced into the extraction region via the sample introduction system. In this region, the sample is ionized by an ionization source to form a stable ion beam. The ionized ions then enter the acceleration region, where they undergo uniform acceleration under the electric field generated by the high-voltage power supply system. Subsequently, the accelerated ions enter the field-free flight region, where they move at a constant velocity in a straight line. Finally, the ions are received by the detector (MCP detector), which generates a signal output. Appropriate processing of the detector output signal yields the mass spectrometry signal of the sample.

[0125] Furthermore, because air has a strong absorption effect on ions, a vacuum system is necessary to remove the air from the instrument; therefore, the mass spectrometer must be connected to the vacuum system. The vacuum system of a laboratory mass spectrometer is typically achieved by combining a turbomolecular pump with a rotary vane mechanical pump, a dry pump, a diaphragm pump, or other forestage rough vacuum pumps. In this invention, the time-of-flight field vacuum chamber is achieved using two CAS Instruments FF-200 molecular pumps and a Chengdu Nanguang Machinery 2XZ-4B mechanical pump.

[0126] In the final signal detection stage, the detector's output signal is input to an oscilloscope via a signal line. This invention uses a Tektronix DPO3032 oscilloscope for real-time signal observation. It also includes a Dongwen DW-MZ501 high-voltage power supply and voltage plates.

[0127] Photoacoustic Detector (PAD): The device for detecting photoacoustic signals in this invention is an ultrasonic transducer. Its core working principle is the piezoelectric effect, and piezoelectric ceramics are the key material for realizing this effect. When sound waves act on the piezoelectric ceramic, the crystal deforms under the sound pressure, thereby generating alternating electrical signals. These signals are received by the probe and transmitted to the instrument for amplification and processing. As the core of the transducer, it realizes efficient bidirectional conversion between electrical energy and acoustic energy, which is the foundation of the ultrasonic probe's function. Piezoelectric ceramics are the "heart" of ultrasonic technology; their unique piezoelectric effect enables precise conversion between acoustic energy and electrical energy.

[0128] Photoacoustic signal acquisition module: The self-built photoacoustic signal acquisition module preprocesses the photoacoustic time domain signal. The specific components include: a preamplifier with a gain of 40dB and a bandwidth of 50kHz~20MHz; a filter to remove background noise; and an oscilloscope to capture the original time domain signal in real time.

[0129] This invention is achieved through the following technical process:

[0130] This invention proposes a dual-modal imaging method and system based on photoacoustic spectroscopy and laser mass spectrometry. It utilizes an integrated vacuum system to detect and image the sample, a photoacoustic signal detection system to detect and image the ultrasonic signal generated by laser excitation of the sample, and a mass spectrometry detection system to perform mass spectrometry imaging on the charged molecules and atoms generated by desorption and ionization of the sample. Then, by fusing and comparing the data from the two detections, multimodal imaging of the sample under test can be achieved.

[0131] The specific process of this invention includes the following steps:

[0132] (1) After processing, the sample is fixed on the matrix (preferably, the matrix of LDI-MSI can be carbon, so that more analytes are vaporized during the desorption process. Similarly, if MALDI technology is used, the sample will be processed according to different samples. This process is not the focus of this patent).

[0133] (2) Turn on the mass spectrometer (vacuum pump, molecular pump, etc.) and the three-dimensional platform sampler;

[0134] (3) Taking vacuum environment imaging (LDPI-MSI) as an example: the matrix with the sample fixed is fixed on the injection rod of the mass spectrometer;

[0135] (4) Adjust the position of the injection rod, select the sampling and detection area, and enter the laser-equipped time-of-flight mass spectrometer for detection through the injection system;

[0136] (5) The two solid-state lasers in the time-of-flight mass spectrometer used in this invention emit pulsed lasers in sequence. The solid-state laser emits pulsed lasers to generate the first 1064nm laser beam, which is focused by the constructed optical path system and enters the ionization chamber. It is then focused by the measuring objective lens onto the sample to be tested, with the direction facing the ultrasonic probe. The sample generates sound waves as it is excited by the laser, and the sound signal is converted into an electrical signal by the ultrasonic probe.

[0137] (6) At the same time, the first pulsed laser also serves as the desorption laser in post-laser ionization mass spectrometry imaging, achieving optical path reuse. This desorption step causes the matrix and sample molecules to detach from the target surface and form a neutral molecular vapor cloud.

[0138] (7) After desorption, after a delay of 40-50us, the second pulsed laser is emitted. The second pulsed laser is converted into a vacuum ultraviolet laser after being tripled by the gas cell. The gas in the gas cell is a mixture of argon and xenon with a volume ratio of 1:8 to 1:10. The first pulsed laser and the vacuum ultraviolet laser are separated by focusing with a plano-convex mirror at the light outlet of the gas cell. The wavelength is 118nm. After focusing, the vacuum ultraviolet laser enters the ionization chamber parallel to the direction of the sample feed rod and crosses with the first pulsed laser. The gas cloud (neutral molecular vapor cloud) generated after the first pulsed laser is desorbed is then ionized. The ionized ions fly through the flight field and are detected by the microchannel plate.

[0139] (8) Scan the entire sample. The scanning steps are as follows: Select the target point, and use the electronic device to control the three-dimensional stage to move the sample under test up and down near the focal point of the measuring objective lens. Starting from the marked point, scan forward one column and then move automatically one step to the right. Then scan backward one column. Repeat this process until all the test points on the sample under test are measured.

[0140] (9) Similarly, if atmospheric pressure imaging is used, the sample feed rod in steps (3) and (4) is not needed. The sample platform is placed outside the mass spectrometer. The specific implementation method is as follows: 1. Step (5); 2. Step (6); 3. Sample injection: The sample injection method is changed to laser desorption followed by transmission through a heated capillary (300℃) to directly capture the desorbed neutral molecules and prevent condensation. With the assistance of heating and carrier gas, nitrogen carrier gas (flow rate 1~3L / min) pushes the molecules to move towards the ionization region and enter the mass spectrometer for detection. 4. (7) The subsequent steps are the same as above.

[0141] (10) In step (5), the time domain signal of photoacoustic is collected by the transducer, and then the signal is filtered and amplified by the hardware module. The pre-processed data is then imported into the data processing unit for further processing to obtain photoacoustic data.

[0142] (11) The detected photoacoustic signal data is processed by the visualization computing software based on Python, which is developed independently, to obtain effective signals. The large amount of photoacoustic time domain data collected can be automatically calculated according to the preset algorithm. It only takes tens of seconds to process tens of thousands of data and obtain the corresponding envelope signal, frequency domain signal and other data for analysis.

[0143] (12) Extract the maximum value of the acquired envelope signal, map the different envelope values ​​at different locations to an image, and obtain the photoacoustic image of the entire sample;

[0144] (13) The signal detected by the microchannel plate in step (7) is converted into an electrical signal by an oscilloscope;

[0145] (14) The ion intensity at a specific m / z is converted into a grayscale or pseudocolor image using software to form a molecular distribution map. Preferably, based on Matlab software, the algorithm logic is as follows: calculate the peak area at a certain position on the spectrum, convert these peak areas into a matrix according to the acquisition position, and then use different colors to represent these numbers to obtain a mass spectrometry imaging image.

[0146] (15) Fusion analysis stage: The electronic device fuses and analyzes the sample information of the laser focusing micro-area measured by the photoacoustic detection system and the spot position with the mass spectrometry information measured by the mass spectrometry detection system, and then obtains the multimodal imaging information of the focusing spot micro-area;

[0147] refer to Figure 5 The algorithm for the data processing unit mentioned in step (7) is as follows:

[0148] (7.1) Multidimensional signal processing: including time domain processing, raw signal truncation, and Hilbert envelope calculation.

[0149] The specific steps are as follows: construct an analytic signal for the time-domain signal x(t), perform orthogonal demodulation, and extract the envelope signal.

[0150] ;

[0151] Where H is the Hilbert transform operator, which is equivalent in the frequency domain to: ; For Fourier transform operators;

[0152] ;

[0153] Where x(t) is the time-domain signal, For the analytical signal, the real part is the original time-domain signal x(t), and the imaginary part is the Hilbert transform of the time-domain signal x(t). Let y(t) be the envelope signal, and y(t) be the result of the Hilbert transform. These are the time variable, the imaginary unit, and the angular frequency, respectively. This is the Fourier transform result of the time-domain signal x(t).

[0154] Then, a rectangular window function is used to implement time-domain bandpass filtering.

[0155] In this embodiment, a time window is used. Extracting the time-domain signal:

[0156] ;

[0157] in, These represent the start and end times of the time-domain signal truncation, respectively. `rect` is a rectangular window function used to select the effective time-domain signal segment to eliminate noise interference. The truncated time-domain signal segment. The process will then proceed to the Hilbert envelope calculation stage. By constructing an analytical signal z(t), orthogonal demodulation of the time-domain signal segment will be achieved, and finally, the envelope signal e(t) reflecting the intensity change of the photoacoustic signal will be extracted, providing a foundation for subsequent photoacoustic imaging data processing.

[0158] (7.2) Calculate the sampling rate based on time difference:

[0159] ;

[0160] ;

[0161] in, Indicates the sampling rate. Let t[i] and t[i-1] be the time interval between adjacent sampling points, respectively, and let t[i] and t[i-1] be the time values ​​corresponding to the i-th and (i-1)-th sampling points. This indicates calculating the average. This represents averaging the differences between adjacent time points t[i] and t[i-1] in a time series. By calculating the average of the time intervals between all adjacent sampling points in the time series and then taking its reciprocal, the actual sampling rate of the system can be obtained, ensuring the accuracy of the time axis in subsequent data processing. In practice, it is necessary to first perform linear fitting on the timestamp data of the original time-domain signal to eliminate time drift caused by equipment errors, and then perform time interval calculations. The calculation is performed to improve the accuracy of the sampling rate calculation.

[0162] (7.3) Batch processing mechanism:

[0163] Perform time-domain and envelope data integration. Automatically scan all CSV files in a specified directory, complete folder traversal, skip files with insufficient data or incorrect formats, and finally aggregate the results, storing the cross-file data uniformly. The specific steps are: starting with the time series of the first file... Using the base column, for the k-th file signal ,satisfy: ;Require Otherwise, skip.

[0164] in, This represents the time series of the k-th file; The baseline time series is typically the time series of the first file successfully read; len() represents the length of the time series, i.e. Refers to the base time series The total number of time points included; This refers to the time series of the k-th file. The total number of time points included; N is the length of the base time series, i.e. M represents the signal matrix, and the signal matrix M has 100 rows. The number of columns in the signal matrix M is equal to the total number of files; Indicates the index of a time series; This represents the signal value of the element at time point i in the k-th file, corresponding to the signal matrix M. , Let be the signal value of the element in the i-th row and k-th column of the signal matrix M. The core effect of this series of operations is to automatically align and integrate time-domain data across multiple files. If the time series lengths are different, it indicates a problem with the data in the file, and data acquisition is stopped; there is no need to participate in the above batch processing.

[0165] (7.4) Visualization output:

[0166] Combined chart: displaying the raw signal and its envelope together;

[0167] High-resolution saving: 300dpi, PNG format output;

[0168] (7.5) Map the obtained envelope signal peaks to an image.

[0169] The specific steps of the fusion analysis in step (13) are as follows:

[0170] (13.1) Feature point matching: Based on spatial registration algorithm and multimodal fusion strategy, PAI target points and MSI slice anatomical landmarks are extracted. The images are aligned by affine transformation matrix. The MSI images are resampled by bilinear interpolation according to transformation matrix T and aligned with the PAI structure map space.

[0171] (13.2) Generate a dual-modal correlation report. PAI provides structural boundaries and MSI provides specific molecular information. Overlay the reports to generate a fused image.

[0172] Compared with the prior art, the present invention has the following technical advantages:

[0173] (1) Sample information content: Achieving double insurance for multi-dimensional information fusion imaging. The PAI-MSI combined technology proposed in this invention adopts dual-modal tandem analysis of the same sample to achieve collaborative acquisition of multi-modal information of the same sample. It has richer information content and lower false positive probability than the current single-modal imaging technology, and promotes the solution to the problem that current imaging technology cannot simultaneously realize the analysis from "macroscopic structural anomalies" to "microscopic molecular mechanisms".

[0174] (2) Imaging functionality: Solving the problems of sample destructiveness and low utilization. This invention proposes PAI-guided MSI from "blind sampling" to "targeted deep mining", which cleverly combines the advantages of PAI's high contrast and MSI's good imaging penetration depth, resolving the contradiction between imaging depth and molecular specificity, and avoiding the problems of low sample utilization and wasted time in full sample scanning caused by traditional blind sampling; at the same time, the PAI mentioned in this invention is completely label-free, avoiding the contamination of mass spectrometry signals by staining agents and mounting media, and preserving the original molecular state.

[0175] (3) In terms of system integration: This invention achieves single-platform integrated dual-modal imaging through the innovation of system integration (same-platform detection + coaxial optical path multiplexing). The integrated system is used to perform in-situ detection and imaging of samples, reducing equipment redundancy and operational complexity. Specifically, it solves the spatial registration problem and correlates multimodal imaging information; it eliminates the need to move the sample, avoiding sample waste; it simplifies the optical path, achieves optical path multiplexing, and optimizes laser efficiency and cost; and it improves sensitivity and resolution.

[0176] (4) This invention effectively breaks through the bottleneck of imaging technology in terms of resolution through hardware and algorithm improvements. Through data processing algorithms, the peak value of the photoacoustic signal envelope signal and the peak area of ​​the mass spectrometry signal are used to improve the imaging quality and perform dual-mode fusion analysis. Through hardware module processing, this invention effectively separates noise and photoacoustic signals, improves the detection capability of weak signals, achieves low background interference, and enhances the signal-to-noise ratio.

[0177] (5) The PAI mentioned in this invention uses non-invasive pulsed laser excitation to generate signals for imaging, avoiding the dependence on fluorescent markers and the resulting light damage problems in fluorescence imaging, ensuring integrity, using a non-ionizing light source, avoiding radiation risks, and is safer than other MSI multimodal imaging technologies.

[0178] This invention also provides a dual-modal imaging system based on photoacoustic spectroscopy and laser mass spectrometry, comprising:

[0179] A sample fixation device is used to fix processed samples onto a matrix;

[0180] Mass spectrometer and three-dimensional platform sampler are used for sample introduction and detection;

[0181] The injection rod is used to hold the matrix containing the sample.

[0182] Two solid-state lasers are used to emit pulsed laser light;

[0183] Optical path system, used for focusing and transmitting laser light;

[0184] Ionization chamber, used for the interaction between laser and sample and the generation of ions;

[0185] An ultrasonic probe is used to detect the sound waves generated by a sample and convert them into electrical signals.

[0186] Microchannel plates are used to detect ions generated during ionization.

[0187] The data processing unit is used to process the acquired photoacoustic and mass spectrometry signals;

[0188] Electronic equipment used to fuse and analyze photoacoustic images with molecular distribution maps to obtain multimodal imaging information of the focused spot micro-region.

[0189] The content of the above method embodiments is applicable to this embodiment. The specific functions implemented in this embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments. Therefore, they will not be repeated here.

[0190] Although the description of this disclosure has been quite detailed and particularly focused on several of the described embodiments, it is not intended to limit itself to any of these details or embodiments or any particular embodiment, but should be considered as effectively covering the intended scope of this disclosure by referring to the appended claims and taking into account the broad possible interpretations of these claims provided by the prior art. Furthermore, the foregoing description of this disclosure with respect to embodiments foreseeable by the inventors is intended to provide a useful description, and non-substantial modifications to this disclosure that have not yet been foreseen may still represent equivalent modifications.

Claims

1. A dual-modal imaging method based on photoacoustic spectroscopy and laser mass spectrometry, characterized in that, The method includes the following steps: After processing, the sample is fixed onto the matrix, and the mass spectrometer and three-dimensional platform sampler are turned on. The matrix with the sample fixed is fixed on the injection rod of the mass spectrometer. The position of the injection rod is adjusted, the sampling and detection area is selected, and the sample enters the laser-equipped time-of-flight mass spectrometer for detection through the injection system. The two solid-state lasers in the time-of-flight mass spectrometer emit pulsed lasers in sequence. The first pulsed laser beam is focused by the optical path system and enters the ionization chamber. It is then focused onto the sample by the measurement objective. The sample generates sound waves as it is excited by the laser. The sound signal is converted into an electrical signal by the ultrasonic probe. At the same time, the first pulsed laser beam acts as a desorption laser, causing the matrix to detach from the sample molecules and form a neutral molecular vapor cloud. The second pulsed laser beam is emitted after a delay. After being tripled in the gas cell, it is converted into a vacuum ultraviolet laser. After being focused, it enters the ionization chamber and intersects with the first pulsed laser beam to ionize the neutral molecular vapor cloud. The ionized ions fly through the flight field and are detected by the microchannel plate. The sample is scanned and detected as a whole. The photoacoustic time domain signal is collected by the transducer. After preprocessing, it is imported into the data processing unit to obtain photoacoustic data. After processing the photoacoustic data, the maximum value of the envelope signal is extracted and mapped to the photoacoustic image. The signal detected by the microchannel plate is converted into an electrical signal by an oscilloscope, and the electrical signal is converted into a grayscale or pseudocolor image to form a molecular distribution map. By fusing and analyzing the photoacoustic imaging map and the molecular distribution map, multimodal imaging information of the focused spot micro-region can be obtained; The overall scanning and detection of the sample includes: Select the target point, and let the electronic device control the three-dimensional stage to move the sample under test up and down near the focal point of the measuring objective lens. Starting from the marked point, first scan forward one column and then automatically move one step to the right, and then scan backward one column until all the test points on the sample under test are measured. The step of extracting the maximum value of the envelope signal after photoacoustic data processing and mapping it to a photoacoustic image includes: The photoacoustic data is converted into a time-domain signal. The time-domain signal segment within a specified time window is extracted by the data processing unit. A rectangular window function is used to perform time-domain bandpass filtering on the time-domain signal segment to obtain the filtered time-domain signal. The filtered time-domain signal is subjected to Hilbert transform to construct an analytic signal. The magnitude of the analytic signal is calculated as the envelope signal. The peak value of the envelope signal is mapped to a photoacoustic image according to its spatial location.

2. The method according to claim 1, characterized in that, The method of causing the two solid-state lasers in the time-of-flight mass spectrometer to emit pulsed lasers sequentially includes: The solid-state laser emits pulsed laser light to generate the first pulsed laser beam. The first pulsed laser beam is focused by the constructed optical path system and enters the ionization chamber. It is then focused onto the sample under test by the measuring objective lens. The focusing direction is opposite to the direction of the ultrasonic probe. The sample generates sound waves as it is excited by the laser. The sound waves are detected by the ultrasonic probe and the sound signal is converted into an electrical signal. Meanwhile, the first pulsed laser beam serves as the desorption laser in post-laser ionization mass spectrometry imaging, enabling optical path multiplexing. Through the desorption process, the matrix and sample molecules detach from the target surface and form a neutral molecular vapor cloud.

3. The method according to claim 1, characterized in that, The delayed emission of the second pulsed laser beam, after being third-harmonicized in the gas cell, is converted into a vacuum ultraviolet laser, which, after being focused, enters the ionization chamber and intersects with the first pulsed laser beam, including: A second pulsed laser is emitted after a delay following the desorption process. The second pulsed laser is converted into a vacuum ultraviolet laser after being tripled in the gas cell. The gas in the gas cell is a mixture of argon and xenon. The first pulsed laser and the vacuum ultraviolet laser are separated by focusing with a plano-convex mirror at the light outlet of the gas cell. After being focused, the vacuum ultraviolet laser enters the ionization chamber parallel to the sample feed rod and intersects with the first pulsed laser.

4. The method according to claim 1, characterized in that, The process of converting the electrical signal into a grayscale or pseudo-color image to form a molecular distribution map includes: Calculate the peak area of ​​the signal peak where the target substance is located on the spectrum, convert the peak area at different acquisition positions into a matrix, and then use different colors to represent the values ​​in the matrix to generate grayscale or pseudo-color images to form a molecular distribution map.

5. The method according to claim 1, characterized in that, The process of fusing and analyzing photoacoustic images with molecular distribution maps to obtain multimodal imaging information of the focused spot micro-region includes: Based on spatial registration algorithm and multimodal fusion strategy, target points of photoacoustic imaging map and slice anatomical landmarks of molecular distribution map are extracted. The photoacoustic imaging map and molecular distribution map are aligned by combining affine transformation matrix. The molecular distribution map is resampled by bilinear interpolation according to the transformation matrix and spatially aligned with the photoacoustic imaging map to obtain bimodal data. Structural boundary information of the sample is extracted from the photoacoustic image, and molecular composition information of the sample is extracted from the molecular distribution map. The dual-modal data are superimposed to generate a dual-modal fused image that combines structural and molecular features.

6. A dual-modal imaging system based on photoacoustic spectroscopy and laser mass spectrometry, characterized in that, The method applied to any one of claims 1 to 5 includes: A sample fixation device is used to fix processed samples onto a matrix; Mass spectrometer and three-dimensional platform sampler are used for sample introduction and detection; The injection rod is used to hold the matrix containing the sample. Two solid-state lasers are used to emit pulsed laser light; Optical path system, used for focusing and transmitting laser light; Ionization chamber, used for the interaction between laser and sample and the generation of ions; An ultrasonic probe is used to detect the sound waves generated by a sample and convert them into electrical signals. Microchannel plates are used to detect ions generated during ionization. The data processing unit is used to process the acquired photoacoustic and mass spectrometry signals; Electronic equipment used to fuse and analyze photoacoustic images with molecular distribution maps to obtain multimodal imaging information of the focused spot micro-region.

Images