Atomization image analysis method for pulmonary ventilation function evaluation

Abstract

The invention relates to the technical field of lung function evaluation, in particular to an atomization image analysis method for lung ventilation function evaluation. The method comprises the following steps: firstly, acquiring lung atomization images under various concentrations by changing the concentration of aerosol; based on image block level analysis, the lung is divided into different functional regions (such as lung lobe and airway regions), particle distribution difference characteristics of different image blocks are compared to quantify deposition indexes, and regional ventilation uniformity is reflected; sub-regions are further divided in combination with density clustering, the number change of atomized particles is analyzed across time points, and a metabolic rate index is calculated to represent local metabolic activity; and finally fusing the deposition index and the metabolic rate index to generate a ventilation obstruction evaluation value. According to the method, through space-time sequence multi-parameter coupling, the breathing mode and individual difference interference are effectively reduced, the ventilation function and the metabolism state are synchronously analyzed, a quantitative basis is provided for pulmonary ventilation disorder analysis, and evaluation comprehensiveness and clinical decision accuracy are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of pulmonary function assessment technology, and more specifically to a nebulized image analysis method for assessing pulmonary ventilation function. Background Technology

[0002] Pulmonary ventilation function is a dynamic indicator that measures the process of air entering and exiting the alveoli. It is an important means of diagnosing and monitoring respiratory diseases and is commonly used in the diagnosis and treatment of chronic obstructive pulmonary disease (COPD), asthma, pulmonary fibrosis, and other diseases. It is mainly achieved by generating aerosols of a specific particle size using a nebulizer and then capturing the distribution of these particles in the patient's lungs, thereby assisting doctors in assessing pulmonary ventilation function.

[0003] Existing technologies for assisting in the assessment of pulmonary ventilation function by measuring the distribution of nebulized particles in the lungs typically use a single concentration of aerosol and static global indicators (such as the FEV1 / FVC ratio). However, individual differences caused by variations in patients' inhalation methods and breathing patterns can lead to uneven deposition and distribution of a single concentration of nebulized tracer in the lungs. Furthermore, static global indicators are insufficient to quantify the dynamic deposition and metabolic processes of nebulized particles in the patient's lungs, resulting in reduced accuracy in assessing pulmonary ventilation function. Summary of the Invention

[0004] To address the technical problem that individual differences in inhalation methods and breathing patterns lead to uneven deposition and distribution of a single concentration of nebulized tracer in the lungs, and that static global indicators are insufficient to quantify the dynamic deposition and metabolic processes of nebulized particles in the patient's lungs, thus reducing the accuracy of pulmonary ventilation function assessment, this invention aims to provide a nebulized image analysis method for pulmonary ventilation function assessment. The specific technical solution adopted is as follows: The concentration of aerosols was changed in descending order over time, and lung nebulization images of each patient were obtained at different concentrations. Each lung nebulization image is segmented into several image blocks. Based on the differences in the distribution characteristics of nebulized particles between different image blocks, the particle distribution index of each image block is determined. The changing trend characteristics of the particle distribution index in the lung nebulization images of the same image block at different times are analyzed to determine the nebulized particle deposition index of each image block. In the first lung nebulization image, the nebulized particles in each image block are clustered based on the density characteristics of the nebulized particles, thereby dividing each image block into several sub-regions; based on the positional distribution of the sub-regions in each image block and the change characteristics of the number of nebulized particles in the corresponding region in the lung nebulization images at different times, the metabolic rate of the nebulized particles corresponding to each image block is determined. By integrating particle deposition indices and nebulized particle metabolic rates from each imaging block, a ventilation obstruction assessment value for each imaging block of the patient's lungs is obtained, which is used for auxiliary assessment of pulmonary ventilation function.

[0005] Furthermore, the method for obtaining the particle distribution index includes: In each lung nebulization image, the differences in the number of nebulized particles between all image blocks are analyzed to determine the first distribution factor for each image block; Based on the density difference characteristics of the fogged particles among all image blocks, the second distribution factor of each image block is determined; The normalized sum of the first and second distribution factors of each image block is used as the particle distribution index for each image block.

[0006] Furthermore, the method for obtaining the first distribution factor includes: In each lung nebulization image, the average number of nebulized particles in all image blocks is used as the baseline value. The normalized value of the difference between the number of nebulized particles in each image block and the baseline value is used as the first distribution factor for each image block.

[0007] Furthermore, the method for obtaining the second distribution factor includes: In each lung nebulization image, the ratio of the number of nebulized particles to the area of ​​each image block is used as the density factor. The mean of the density factors of all image blocks is used as the density baseline value. The normalized value of the difference between the density factor of each image block and the density baseline value is used as the second distribution factor of each image block.

[0008] Furthermore, the method for obtaining the atomized particle deposition index includes: For any image patch, in the lung nebulization images at every two adjacent time points, the normalized value of the difference between the particulate matter distribution index at the later time point and the particulate matter distribution index at the previous time point is used as the particulate matter deposition factor of that image patch at the two adjacent time points. All particulate deposition factors of each image block are fitted with a straight line according to time series. The slope value of the fitted line is normalized and used as the deposition weight of each image block. The deposition weight of each image block is multiplied by the mean of all particulate deposition factors corresponding to each image block, and the normalized value of the product is used as the fog particle deposition index of each image block.

[0009] Furthermore, the method for obtaining the sub-region includes: In the first lung nebulization image, the nebulized particles in each image block were clustered based on the DBSCAN clustering algorithm to obtain all clusters. The minimum number of samples and the neighborhood radius were preset values. Each cluster is treated as a sub-region, thus obtaining all sub-regions corresponding to each image block.

[0010] Furthermore, the method for obtaining the metabolic rate of the atomized particles includes: In the first lung nebulized image, the positional relationship between sub-regions in each image block is analyzed to determine the metabolic rate confidence level corresponding to each sub-region; Analyze the changes in the number of nebulized particles in sub-regions of the first lung nebulization image and their corresponding regions in other lung nebulization images to determine the metabolic factors of nebulized particles in each sub-region. In each image block of the first lung nebulization image, the metabolic factors of nebulized particles are weighted and averaged using the metabolic rate confidence of the sub-region. The normalized value of the weighted result is used as the metabolic rate of nebulized particles for each image block.

[0011] Furthermore, the method for obtaining the metabolic rate confidence score includes: In the first lung nebulization image, the largest sub-region in each image block is taken as the core region; The normalized Euclidean distance between the centroid of each sub-region and the centroid of the core region is used as the metabolic rate confidence score for each sub-region.

[0012] Furthermore, the method for obtaining the metabolic factors of the atomized particles includes: Each sub-region in each image block of the first lung nebulization image is used as the corresponding region in other lung nebulization images as the comparison region; Each sub-region and its corresponding comparison regions are sorted in chronological order to obtain a sorted sequence. The sub-regions and comparison regions in the sorted sequence are collectively referred to as the target region. In the sorting sequence, in every two adjacent target regions, the ratio of the difference between the number of atomized particles in the previous target region and the number of atomized particles in the next target region to the time interval is used as a metabolic parameter. The mean value of all metabolic parameters corresponding to each sub-region in each image block of the first lung nebulization image is used as the nebulized particle metabolic factor for each sub-region.

[0013] Furthermore, the method for obtaining the ventilation obstruction assessment value includes: In the first lung nebulization image, the product of the nebulized particle deposition index and the nebulized particle metabolic rate corresponding to each image block is negatively correlated and normalized, and the result is used as the ventilation obstruction factor for each image block. In the first lung nebulized image, the mean value of the ventilation obstruction factor of all image blocks is used as the obstruction baseline value. The normalized value of the difference between the ventilation obstruction factor of each image block and the obstruction baseline value is used as the ventilation obstruction assessment value of each image block.

[0014] The present invention has the following beneficial effects: In this invention, by designing a descending concentration of aerosols, lung nebulization images of patients at different concentrations can be obtained. This more closely reflects the dynamic fluctuations of aerosol concentration during natural breathing and can correct for the influence of the patient's breathing pattern (such as deep inhalation or shallow, rapid breathing) on ​​particle deposition, reducing assessment bias caused by differences in inhalation methods. When a patient's lung ventilation function is normal, the distribution of nebulized particles will be more uniform. As aerosols are continuously inhaled, particle deposition will occur in the ventilation area. Therefore, by analyzing the image block level, the lungs are divided into different regions (such as the upper and lower lobes, central and peripheral airways), and then the distribution differences of nebulized particles between different image blocks and the changing trends in lung nebulization images at different concentrations can be analyzed. This allows for the quantification of nebulized particle deposition indices for each image block, serving as an indicator for assessing the ventilation function of the corresponding lung region. Furthermore, the atomized particles transformed from radioactive aerosols cannot be exhaled; they spread to the alveoli and are metabolized by the kidneys. Differences in cell types and local metabolic activity across different lung regions affect the distribution of atomized particles. Therefore, analyzing the metabolism of atomized particles in different lung regions can further confirm ventilation function. Thus, in the first lung atomization image, sub-regions are divided based on atomized particle density. The metabolic rate of atomized particles is determined based on the location distribution within each sub-region and the changes in the number of atomized particles within that sub-region over different time periods, serving as another indicator of ventilation function. Finally, the two indicators for ventilation function corresponding to each image block are fused to obtain a ventilation obstruction assessment value, assisting physicians in assessing lung ventilation function and effectively improving comprehensiveness and accuracy. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0016] Figure 1A flowchart of a nebulized image analysis method for assessing pulmonary ventilation function provided in one embodiment of the present invention; Figure 2 This is a flowchart illustrating a method for obtaining particle distribution indicators according to an embodiment of the present invention. Figure 3 This is a flowchart illustrating a method for obtaining the metabolic rate of atomized particles according to an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a nebulized image analysis method for assessing pulmonary ventilation function according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following describes in detail, with reference to the accompanying drawings, a specific scheme for a nebulized image analysis method for assessing pulmonary ventilation function provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of a nebulized image analysis method for assessing pulmonary ventilation function according to an embodiment of the present invention. The method includes the following steps: Step S1: Change the concentration of aerosol in descending order in time and acquire lung nebulization images for each patient at different concentrations.

[0021] Pulmonary ventilation function is a core physiological indicator measuring the efficiency of air intake and exhaust and gas exchange in the alveoli. Its dynamic assessment has irreplaceable clinical value for the early diagnosis, efficacy monitoring, and prognosis of respiratory diseases such as chronic obstructive pulmonary disease (COPD), asthma, pulmonary fibrosis, and cystic fibrosis. Pulmonary ventilation function is usually assessed by evaluating the distribution of nebulized particles in the lungs. However, due to individual differences in patients' inhalation methods and breathing patterns, the deposition and distribution of a single concentration of nebulized tracer in the lungs can be uneven. Furthermore, nebulized particles undergo dynamic deposition and metabolism in the patient's lungs. Therefore, in this embodiment of the invention, different concentrations of radioactive aerosol are supplied to obtain nebulized images of the patient's lungs at different concentrations. In subsequent processes, by analyzing the distribution characteristics and changes between nebulized particles in the lung nebulized images at different concentrations, a more accurate assessment of pulmonary ventilation function can be achieved, assisting physicians in making more accurate judgments.

[0022] Specifically, methods for acquiring lung nebulization images can include: using 99mTc-DTPA radioactive aerosol as a tracer, which can be cleared from the alveolar region through transepithelial diffusion, and the absorbed dose is filtered out by the glomeruli in the kidneys, making it unlikely to cause harm to the human body. Providing nebulized tracers, i.e., radioactive aerosols, at different concentrations can create a gradient response, helping to distinguish between insufficient particle deposition caused by individual breathing patterns and true ventilation abnormalities caused by ventilation obstruction. Using only a single concentration results in similar particle deposition superposition effects, and the low contrast makes it difficult to distinguish whether the insufficient tracer dose is due to the patient's shallow breathing or whether pathological ventilation obstruction restricts gas exchange in the region. By supplying different concentrations of radioactive aerosols, the deposition dynamics in the lung region under varying concentrations can be observed. The safe dosage range for 99mTc-DTPA is 20-30 MBq. To enhance contrast by analyzing the distribution and deposition of nebulized particles in different lung regions through time-dynamic imaging, aerosols of different concentrations were prepared, with a total dose not exceeding 30 MBq. The concentration of 99mTc-DTPA was sequentially decreased, and a nebulizer was used to convert the different concentrations (inhaled in descending order of concentration) of 99mTc-DTPA solution into fine particles. The patient was placed in a supine position with arms behind their head, ensuring the probe was as close to the chest as possible. Inhalation was performed according to the predetermined inhalation protocol (deep inhalation, breath-holding for a certain period, and then exhalation as instructed by the doctor) to ensure the tracer was delivered to all lung regions as evenly as possible. Then, single-photon emission computed tomography (SPECT) was used to acquire images of the patient's lungs after inhaling the tracer. The nebulization time-series images were obtained during the patient's multiple breath-holding moments (intermediate moments) under the physician's instruction. The acquired image data was preprocessed, including but not limited to noise suppression and image enhancement, to facilitate the subsequent extraction of tracer distribution information. Thus, by changing the aerosol concentration in descending order in time, lung nebulization images of the patient at different concentrations can be obtained.

[0023] It should be noted that the preprocessing of image data is a well-known technique, and the specific process will not be elaborated here. The number of lung nebulization images can be adjusted according to the implementation scenario, but it must be ensured that there is at least one lung nebulization image at each concentration. At the same time, the concentration of aerosol can be set to 6 levels, and the specific settings can be adjusted according to the implementation scenario, without limitation here. In the embodiments of this invention, the collection and acquisition of patient information data are authorized by the relevant users, and the process does not violate relevant laws and regulations, nor does it violate public order and good morals.

[0024] Step S2: Segment each lung nebulization image into several image blocks; in each lung nebulization image, determine the particle distribution index of each image block based on the differences in the distribution characteristics of nebulized particles between different image blocks; analyze the changing trend characteristics of the particle distribution index in the lung nebulization images of the same image block at different times, and determine the nebulized particle deposition index of each image block.

[0025] The core of assessing pulmonary ventilation function by analyzing lung nebulization images lies in utilizing the transport and deposition characteristics of nebulized particles in the respiratory system to determine the ventilation function of different regions of the lungs. First, each lung nebulization image can be segmented into several image blocks.

[0026] Preferably, in one embodiment of the present invention, the method for acquiring image blocks includes: Each lung nebulization image is segmented based on a pre-trained neural network to obtain several image blocks, wherein the image blocks in this embodiment of the invention represent different lung regions.

[0027] It should be noted that the neural network in this embodiment of the present invention can be a U-Net deep convolutional neural network, and the training process of the neural network is a well-known technology, which will not be described in detail here; in this embodiment of the present invention, it is assumed that the region segmentation results of all lung nebulization images are basically the same, that is, in different lung nebulization images, the image blocks can correspond one-to-one.

[0028] When a patient's lung ventilation function is normal, the nebulized particles in each image block of a lung nebulization image should be evenly distributed. However, if there is airway obstruction and poor ventilation, the nebulized particles cannot enter the obstructed area of ​​the lung with breathing, and there will be less nebulized particle deposition in this area. Therefore, the differences in the distribution of nebulized particles between different image blocks in each lung nebulization image can be compared to determine the particle distribution index of each image block and make a preliminary assessment of lung ventilation function.

[0029] Preferably, in one embodiment of the present invention, the method for obtaining the particle distribution index includes: Please see Figure 2 The diagram illustrates a method flowchart for obtaining particle distribution indicators according to an embodiment of the present invention. The method includes the following steps: Step S201: In each lung nebulization image, analyze the differences in the number of nebulized particles between all image blocks to determine the first distribution factor for each image block.

[0030] In each lung nebulization image, the average number of nebulized particles in all image blocks is used as the baseline value. The baseline value can quantify the average level of the number of nebulized particles in all image blocks of the entire lung nebulization image, providing a global reference standard.

[0031] Then, the difference between the number of atomized particles in each image block and the baseline value is calculated. A negative and smaller difference indicates that the number of atomized particles in the image block is less than the overall average, making the image block more likely to be an area with sparse atomized particles due to ventilation problems. Conversely, a positive and larger difference indicates that the atomized particle distribution in the image block is more normal and more likely to be within the normal range. Therefore, this difference is normalized to obtain the first distribution factor for each image block. Since the difference can be either positive or negative, a function can be used for normalization. .

[0032] Step S202: In each lung nebulization image, a second distribution factor for each image block is determined based on the density difference characteristics of nebulized particles among all image blocks.

[0033] In each lung nebulization image, the ratio of the number of nebulized particles to the area of ​​each image block is used as the density factor. The density factor can eliminate the influence of area differences on the number of nebulized particles and focus more on the density of nebulized particles per unit area.

[0034] Similarly, the mean density factor of all image blocks is used as the density baseline value, which serves as a global reference standard for density, quantifying the overall average level of nebulized particle density. Then, the difference between the density factor of each image block and the density baseline value is calculated. Since the distribution of nebulized particles in the image is more uniform when the patient's lung ventilation function is normal, and nebulized particles cannot enter the obstructed lung area with respiration when lung ventilation function is abnormal, a negative and smaller difference indicates that the number of nebulized particles in the image block is less than the overall average level. This suggests that the image block is more likely to be an area with sparse nebulized particles due to ventilation problems. Conversely, a positive and larger difference indicates a more normal distribution of nebulized particles in the image block, making it more likely to be a normal area. Therefore, this difference is normalized to obtain the second distribution factor for each image block. Since the difference can be positive or negative, a function can be used for normalization. .

[0035] Step S203: In each lung nebulization image, the first distribution factor and the second distribution factor of each image block are fused to obtain the particle distribution index of each image block.

[0036] Based on the analysis in steps S201 and S202 above, it is known that the larger the first distribution factor of an image block, the more normal the distribution of atomized particles in that image block. Similarly, the larger the second distribution factor of an image block, the more normal the distribution of atomized particles in that image block. Therefore, the normalized value of the sum of the first and second distribution factors of each image block is used as the particle distribution index for each image block. At this time, the smaller the particle distribution index of an image block, the more sparse the distribution of atomized particles is compared with other image blocks, and the more likely the image block is to have ventilation dysfunction, requiring higher attention. Normalization is a technique well known to those skilled in the art. The choice of normalization function can be linear normalization or standard normalization, etc., and the specific normalization method is not limited here.

[0037] When a patient initially inhales the maximum concentration of aerosol, there is a higher distribution of aerosol particles in normal areas of lung nebulization images, while fewer particles are found in obstructed areas. As the concentration of the tracer is further reduced, the distribution characteristics of aerosol particles across different lung nebulization images become more apparent. In normal areas, particles tend to accumulate and overlap, while in obstructed areas, the likelihood of low-concentration aerosols reaching these areas is lower, and particle deposition may only increase slightly or remain unchanged. Furthermore, due to aerosol metabolism, the amount of aerosol particles deposited may even decrease. Therefore, in this embodiment of the invention, continuously reducing the concentration of radioactive aerosols can improve the contrast between normal and obstructed areas, thereby increasing the accuracy of lung ventilation function assessment. The particle deposition pattern within the image blocks can then be used as an indicator to quantify ventilation function.

[0038] However, since a patient's breathing pattern (breathing depth, respiratory flow rate, etc.) directly determines the delivery and deposition of the nebulized tracer in the lungs, even if the patient breathes at a fixed number of seconds under the guidance of a doctor, differences in breathing habits and vital capacity may lead to misjudgment, mistaking normal ventilation for abnormality or ignoring mild abnormalities. For example, if a patient has shallow breathing habits and is unable to inhale the nebulized tracer before reaching the breath-holding phase, the nebulized particles may deposit prematurely, resulting in a reduction of tracer in the terminal alveoli and being mistakenly identified as distal ventilation impairment. Therefore, if a uniform deposition standard is set for all patients based solely on the distribution of nebulized particles, the possibility of misjudgment is high. Individual patient differences need to be considered. Therefore, in this embodiment of the invention, the distribution and trend characteristics of nebulized particles in the patient's own time-series lung nebulization images are analyzed to determine the nebulized particle deposition index for each image block, thereby improving the accuracy of lung ventilation function assessment.

[0039] Preferably, in one embodiment of the present invention, the method for obtaining the atomized particle deposition index includes: For any image patch, in the lung nebulized images at every two adjacent time points, the difference between the particulate matter distribution index at the later time point and the particulate matter distribution index at the previous time point is calculated. Based on the aforementioned analysis, if the image patch is a region of pulmonary ventilation obstruction, then this difference should be negative; conversely, if the image patch is a normal lung region, then this difference should be positive. The normalized value of this difference is used as the particulate matter deposition factor for that image patch at two adjacent time points. The normalization method here can be a function. .

[0040] Thus, in all lung nebulization images of each patient, each image block has a particulate deposition factor at two adjacent time points.

[0041] Over time, as aerosols are continuously inhaled into normal lung areas, the particulate matter deposition factor of the same imaging patch tends to increase between adjacent time points. Therefore, a linear regression is performed on all particulate matter deposition factors of each imaging patch over time, and the slope of the fitted line is obtained. A positive slope indicates that particulate matter has deposited in the imaging patch over time, making it more likely to be a normal area. Conversely, a negative slope indicates that particulate matter has not deposited in the imaging patch over time, making it more likely to be a region of ventilatory obstruction. The normalized slope value is then used as the deposition weight for each imaging patch. The normalization method here can be a function. .

[0042] Based on the foregoing analysis and calculations, it is known that the larger the deposition weight of a certain image block, the more obvious the deposition phenomenon in that image block, and the more likely it is to be a normal area; furthermore, the larger the particulate matter deposition factor, the more likely it is to be a normal area. Therefore, in all lung nebulization images of each patient, the mean of all particulate matter deposition factors of the same image block is multiplied by its corresponding deposition weight, and the normalized value of the resulting product is used as the nebulization particulate matter deposition index for each image block. At this time, the larger the nebulization particulate matter deposition index, the more the deposition of nebulized particles in the image block conforms to the nebulization particulate matter deposition characteristics of a normal lung area, and the more likely it is to be a normal area. Normalization is a technique well known to those skilled in the art, and the choice of normalization function can be linear normalization or standard normalization, etc., and the specific normalization method is not limited here.

[0043] It should be noted that the method for obtaining the fitted straight line in the embodiments of the present invention can adopt the least squares method, which is a well-known technique, and the specific process will not be described in detail here.

[0044] At this point, the nebulized particle deposition index for each image block can be obtained, which can be used as an indicator to measure lung ventilation function.

[0045] Step S3: In the first lung nebulization image, the nebulized particles in each image block are clustered based on the density characteristics of the nebulized particles, thereby dividing each image block into several sub-regions; based on the positional distribution of the sub-regions in each image block and the change characteristics of the number of nebulized particles in the corresponding region in the lung nebulization image at different times, the metabolic rate of nebulized particles corresponding to each image block is determined.

[0046] The atomized particles transformed from radioactive aerosols cannot be exhaled and will spread to the alveoli and be metabolized by the kidneys. There are differences in cell types and local metabolic activity in different regions of the lungs. In normal areas of the lungs, ventilation is good, oxygen exchange is sufficient, and cellular energy metabolism is active. The metabolic rate is faster than in areas with obstructed ventilation. Therefore, analyzing the metabolic status of different regions of the lungs can further confirm their ventilation function.

[0047] To analyze the metabolism of aerosol particles in each image block, the density characteristics of aerosol particles in each image block of the lung aerosol image obtained from the initial inhalation of high-concentration aerosol can be clustered to obtain several sub-regions.

[0048] Preferably, in one embodiment of the present invention, the method for obtaining the sub-region includes: In the first lung nebulization image, the nebulized particles in each image block were clustered using the DBSCAN clustering algorithm to obtain all the clusters. The minimum number of samples and the neighborhood radius were preset values.

[0049] Each cluster is treated as a sub-region, thus obtaining all sub-regions corresponding to each image block.

[0050] It should be noted that the DBSCAN clustering algorithm is a well-known technology, and the specific process will not be described in detail here; the minimum number of samples is set to 8, and the neighborhood radius is set to 5 pixels. The specific values ​​can be adjusted according to the implementation scenario, and are not limited here.

[0051] The metabolism of atomized particles can be specifically manifested as a decrease in the number of particles. However, since radioactive aerosols are continuously inhaled in this embodiment of the invention, new atomized particles will continuously accumulate in the upper trachea, while fewer new atomized particles will be deposited in the lower part near the alveoli, thus the impact is smaller. The metabolic situation calculated based on the number of atomized particles will be more accurate. Therefore, the location distribution of sub-regions in each image block of the lung atomization image can provide the confidence level for metabolic calculation. Thus, by combining this with the changes in the number of atomized particles in the corresponding regions of the sub-regions in different lung atomization images, the metabolic rate of atomized particles corresponding to each image block is determined.

[0052] Preferably, in one embodiment of the present invention, the method for obtaining the metabolic rate of atomized particles includes: Please see Figure 3 The diagram illustrates a method flowchart for obtaining the metabolic rate of atomized particles according to an embodiment of the present invention. The method includes the following steps: Step S301: In the first lung nebulized image, analyze the positional relationship between sub-regions in each image block to determine the metabolic rate confidence level corresponding to each sub-region.

[0053] Since the distribution of aerosol particles is related to ventilation, the upper trachea will come into contact with the aerosol inhaled by the patient the fastest. Therefore, in lung nebulization images, the distribution of aerosol particles will decrease from the upper end of the trachea to the lower end. Thus, in the first lung nebulization image, the largest sub-region in each image block is taken as the core region. The core region is the area with the densest distribution of aerosol particles, that is, the area closest to the upper end.

[0054] Then, in each image block of the first lung nebulization image, the Euclidean distance between the centroid of each sub-region and the centroid of the core region is calculated. Based on the aforementioned analysis, it is known that the lower part, closer to the alveolar region, has less newly deposited nebulized particles, thus the confidence level is higher, and therefore the larger the Euclidean distance, the higher the confidence level. Therefore, the normalized value of the Euclidean distance is used as the metabolic rate confidence level for each sub-region. Normalization is a technique well-known to those skilled in the art, and the choice of normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0055] It should be noted that, in this embodiment of the present invention, when calculating the Euclidean distance, a coordinate system can be constructed by taking the lower left corner of the first lung nebulization image as the origin, and the horizontal rightward direction and the vertical upward direction as the positive directions of the coordinate axes, so as to calculate the Euclidean distance.

[0056] Step S302: Analyze the changes in the number of nebulized particles in the sub-regions of the first nebulized image and their corresponding regions in other lung nebulized images, and determine the nebulized particle metabolic factors in each sub-region.

[0057] The metabolic status of nebulized particles can be characterized by the quantity of nebulized particles. Each sub-region in each image block of the first lung nebulized image can be used as the corresponding region in other lung nebulized images as the comparison region.

[0058] Then, each sub-region and its corresponding comparison regions are sorted in chronological order to obtain a sorted sequence. The sub-regions and comparison regions in the sorted sequence are collectively referred to as the target region, which corresponds to the same region at different times.

[0059] In the sorting sequence, within each pair of adjacent target regions, the ratio of the difference between the number of atomized particles in the previous target region and the number of atomized particles in the next target region to the time interval is used as a metabolic parameter. The larger the difference in the number of atomized particles, the greater the reduction in the number of atomized particles in the two adjacent target regions. The ratio of this difference to the time interval corresponding to the two adjacent target regions quantifies the rate of reduction in the number of particles, thus obtaining the metabolic parameter. In this case, the larger the metabolic parameter, the faster the metabolic rate.

[0060] Based on the aforementioned process, a metabolic parameter can be calculated between each pair of adjacent target regions. Finally, the mean value of all metabolic parameters corresponding to each sub-region in each image block of the first lung nebulization image is taken as the nebulized particle metabolic factor of each sub-region. It is easy to understand that the larger the nebulized particle metabolic factor, the faster the nebulized particle metabolic rate of the sub-region, and the higher the probability of normal ventilation function.

[0061] Step S303: In the first lung nebulized image, the metabolic rate confidence of the sub-region in each image block and the nebulized particle metabolic factor are fused to obtain the nebulized particle metabolic rate of each image block.

[0062] Based on the foregoing analysis, the higher the confidence level of the metabolic rate in a sub-region, the higher the confidence level of the calculated metabolic degree. The larger the metabolic factor of the atomized particles in the sub-region, the faster the metabolic rate of the atomized particles. Therefore, in each image block, the metabolic factor of the atomized particles is weighted and averaged using the metabolic rate confidence level of the sub-region, and the normalized value of the weighted result is used as the metabolic rate of the atomized particles in each image block. Then, in each image block, the metabolic rate confidence level of the sub-region is multiplied by the metabolic rate of the atomized particles, and the average level of the products corresponding to all sub-regions is normalized to obtain the metabolic rate of the atomized particles. The larger this value, the faster the metabolic rate of the particles in the corresponding image block, and the higher the probability that it is a region with normal ventilation function. Normalization is a technique well-known to those skilled in the art. The normalization function can be linear normalization or standard normalization, etc., and the specific normalization method is not limited here.

[0063] Thus, the metabolic rate of nebulized particles in each image block of the first lung nebulization image can be obtained, serving as another indicator for assessing the pulmonary ventilation function of the corresponding region.

[0064] Step S4: Fuse the particle deposition index and nebulized particle metabolic rate of each image block to obtain the ventilation obstruction assessment value of each image block of the patient's lungs, which is used for auxiliary assessment of lung ventilation function.

[0065] Based on the aforementioned steps, two indicators can be obtained to measure the lung ventilation function of the corresponding region of the lung in the imaging block. Here, the two indicators can be fused to obtain the ventilation obstruction assessment value of each imaging block in the patient's lungs, which can be used to help doctors conduct auxiliary assessments of lung ventilation function.

[0066] Preferably, in one embodiment of the present invention, the method for obtaining the ventilation obstruction assessment value includes: A higher nebulized particle deposition index for a given image patch indicates that the deposition of nebulized particles in that patch closely resembles the characteristics of nebulized particle deposition in a normal lung region, making it more likely to be a normal region. Conversely, a higher nebulized particle metabolic rate for a given image patch indicates a faster metabolic rate of particles, increasing the likelihood of it being a region with normal ventilation function. Therefore, both the nebulized particle deposition index and the nebulized particle metabolic rate are negatively correlated with the ventilation obstruction assessment value. Thus, in the first lung nebulization image, the product of the nebulized particle deposition index and the nebulized particle metabolic rate for each image patch is negatively correlated, normalized, and used as the ventilation obstruction factor for each image patch. A higher ventilation obstruction factor indicates a higher likelihood of impaired lung ventilation function in that image patch, requiring greater attention from physicians. This negative correlation mapping and normalization can be achieved using the formula... ,in, Let x represent an exponential function with the natural constant e as the base, and let x represent the independent variable.

[0067] In the first nebulized lung image, the mean of the ventilation obstruction factor across all image patches is used as the baseline obstruction value. The normalized difference between the ventilation obstruction factor of each image patch and the baseline obstruction value is then used as the ventilation obstruction assessment value for each image patch. A larger difference indicates a higher degree of deviation of the ventilation obstruction factor from the average level, resulting in a higher ventilation obstruction assessment value and a greater likelihood of abnormalities in the corresponding lung region. The normalization here uses a function... .

[0068] At this point, the ventilation obstruction assessment value of each image block in the first lung nebulization image can be obtained. The larger the value, the higher the probability of an abnormality in the corresponding lung region. Therefore, all image blocks are arranged in descending order according to the ventilation obstruction assessment value to obtain the assessment sequence. In the assessment sequence, the earlier the image block is, the greater the probability of an abnormality, and the higher the priority that the doctor needs to focus on. The lung regions are displayed to the doctor in the order of the assessment sequence to assist the doctor in making a more accurate assessment of lung ventilation function.

[0069] In summary, in this embodiment of the invention, by designing aerosol concentration in descending order, lung nebulization images of patients at different concentrations can be obtained. This more closely reflects the dynamic fluctuations of aerosol concentration during natural breathing and can correct for the influence of the patient's breathing pattern (such as deep inhalation or shallow, rapid breathing) on ​​particle deposition, reducing assessment bias caused by differences in inhalation methods. When a patient's lung ventilation function is normal, the distribution of nebulized particles will be more uniform. As aerosols are continuously inhaled, particle deposition will occur in the ventilation area. Therefore, by analyzing the image block level, the lung is divided into different regions (such as the upper and lower lobes, central and peripheral airways), and then the distribution differences of nebulized particles between different image blocks and the changing trends in lung nebulization images at different concentrations can quantify the nebulized particle deposition index of each image block, serving as an indicator for assessing the ventilation function of the corresponding lung region. Furthermore, the atomized particles transformed from radioactive aerosols cannot be exhaled; they spread to the alveoli and are metabolized by the kidneys. Differences in cell types and local metabolic activity across different lung regions affect the distribution of atomized particles. Therefore, analyzing the metabolism of atomized particles in different lung regions can further confirm ventilation function. Thus, in the first lung atomization image, sub-regions are divided based on atomized particle density. The metabolic rate of atomized particles is determined based on the location distribution within each sub-region and the changes in the number of atomized particles within that sub-region over different time periods, serving as another indicator of ventilation function. Finally, the two indicators for ventilation function corresponding to each image block are fused to obtain a ventilation obstruction assessment value, assisting physicians in assessing lung ventilation function and effectively improving comprehensiveness and accuracy.

[0070] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A nebulized image analysis method for assessing pulmonary ventilation function, characterized in that, The method includes: The concentration of aerosols was changed in descending order over time, and lung nebulization images of each patient were obtained at different concentrations. Each lung nebulization image is segmented into several image blocks. Based on the differences in the distribution characteristics of nebulized particles between different image blocks, the particle distribution index of each image block is determined. The changing trend characteristics of the particle distribution index in the lung nebulization images of the same image block at different times are analyzed to determine the nebulized particle deposition index of each image block. In the first lung nebulization image, the nebulized particles in each image block are clustered based on the density characteristics of the nebulized particles, thereby dividing each image block into several sub-regions; based on the positional distribution of the sub-regions in each image block and the change characteristics of the number of nebulized particles in the corresponding region in the lung nebulization images at different times, the metabolic rate of the nebulized particles corresponding to each image block is determined. By integrating particle deposition indices and nebulized particle metabolic rates from each imaging block, a ventilation obstruction assessment value for each imaging block of the patient's lungs is obtained, which is used for auxiliary assessment of pulmonary ventilation function.

2. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 1, characterized in that, The method for obtaining the particle distribution index includes: In each lung nebulization image, the differences in the number of nebulized particles between all image blocks are analyzed to determine the first distribution factor for each image block; Based on the density difference characteristics of the fogged particles among all image blocks, the second distribution factor of each image block is determined; The normalized sum of the first and second distribution factors of each image block is used as the particle distribution index for each image block.

3. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 2, characterized in that, The method for obtaining the first distribution factor includes: In each lung nebulization image, the average number of nebulized particles in all image blocks is used as the baseline value. The normalized value of the difference between the number of nebulized particles in each image block and the baseline value is used as the first distribution factor for each image block.

4. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 2, characterized in that, The methods for obtaining the second distribution factor include: In each lung nebulization image, the ratio of the number of nebulized particles to the area of ​​each image block is used as the density factor. The mean density factor of all image blocks is used as the density baseline value. The normalized value of the difference between the density factor of each image block and the density baseline value is used as the second distribution factor of each image block.

5. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 1, characterized in that, The method for obtaining the atomized particle deposition index includes: For any image patch, in the lung nebulization images at every two adjacent time points, the normalized value of the difference between the particulate matter distribution index at the later time point and the particulate matter distribution index at the previous time point is used as the particulate matter deposition factor of that image patch at the two adjacent time points. All particulate deposition factors of each image block are fitted with a straight line according to time series. The slope value of the fitted line is normalized and used as the deposition weight of each image block. The deposition weight of each image block is multiplied by the mean of all particulate deposition factors corresponding to each image block, and the normalized value of the product is used as the fog particle deposition index of each image block.

6. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 1, characterized in that, The method for obtaining the sub-region includes: In the first lung nebulization image, the nebulized particles in each image block were clustered based on the DBSCAN clustering algorithm to obtain all clusters. The minimum number of samples and the neighborhood radius were preset values. Each cluster is treated as a sub-region, thus obtaining all sub-regions corresponding to each image block.

7. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 1, characterized in that, The method for obtaining the metabolic rate of the atomized particles includes: In the first lung nebulized image, the positional relationship between sub-regions in each image block is analyzed to determine the metabolic rate confidence level corresponding to each sub-region; Analyze the changes in the number of nebulized particles in sub-regions of the first lung nebulization image and their corresponding regions in other lung nebulization images to determine the metabolic factors of nebulized particles in each sub-region. In each image block of the first lung nebulization image, the metabolic factors of nebulized particles are weighted and averaged using the metabolic rate confidence of the sub-region. The normalized value of the weighted result is used as the metabolic rate of nebulized particles for each image block.

8. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 7, characterized in that, The method for obtaining the metabolic rate confidence score includes: In the first lung nebulization image, the largest sub-region in each image block is taken as the core region; The normalized Euclidean distance between the centroid of each sub-region and the centroid of the core region is used as the metabolic rate confidence score for each sub-region.

9. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 7, characterized in that, The method for obtaining the metabolic factors of the atomized particles includes: Each sub-region in each image block of the first lung nebulization image is used as the corresponding region in other lung nebulization images as the comparison region; Each sub-region and its corresponding comparison regions are sorted in chronological order to obtain a sorted sequence. The sub-regions and comparison regions in the sorted sequence are collectively referred to as the target region. In the sorting sequence, in every two adjacent target regions, the ratio of the difference between the number of atomized particles in the previous target region and the number of atomized particles in the next target region to the time interval is used as a metabolic parameter. The mean value of all metabolic parameters corresponding to each sub-region in each image block of the first lung nebulization image is used as the nebulized particle metabolic factor for each sub-region.

10. The nebulized image analysis method for assessing pulmonary ventilation function according to claim 1, characterized in that, The method for obtaining the ventilation obstruction assessment value includes: In the first lung nebulization image, the product of the nebulized particle deposition index and the nebulized particle metabolic rate corresponding to each image block is negatively correlated and normalized, and the result is used as the ventilation obstruction factor for each image block. In the first lung nebulized image, the mean value of the ventilation obstruction factor of all image blocks is used as the obstruction baseline value. The normalized value of the difference between the ventilation obstruction factor of each image block and the obstruction baseline value is used as the ventilation obstruction assessment value of each image block.

Images