Method for evaluating lung function of patient in pneumonia rehabilitation period

Abstract

The invention discloses a lung function evaluation method for a patient in a pneumonia rehabilitation period, and relates to the technical field of rehabilitation management. Extracting a lung function feature set based on the image data; analyzing based on multi-dimensional parameter association, and establishing a dynamic mapping relationship between the lung function feature set and associated physiological parameters; identifying abnormal associated parameters through a dynamic mapping relation between the lung function feature set and the associated physiological parameters, and when the deviation between the image feature recovery degree and the corresponding physiological parameter recovery degree exceeds a preset threshold value, judging that functional abnormal parameters exist; and matching a potential complication database based on the abnormal parameter type, analyzing the relevance between the abnormal parameter and a preset disease feature model, and outputting secondary disease risk early warning. The method has the advantages that through cross-modal fusion of the lung image features and the dynamic physiological parameters, the early recognition precision of the pneumonia convalescence complication is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of rehabilitation management technology, specifically to a method for assessing lung function in patients recovering from pneumonia. Background Technology

[0002] In the management of patients recovering from pneumonia, analyzing the correlation between pulmonary imaging features and physiological parameters is a core challenge in assessing functional recovery. Traditional methods often assess imaging indicators and physiological parameters independently, such as quantitatively analyzing the degree of pulmonary fibrosis via CT scans or monitoring changes in blood oxygen levels alone. However, the dynamic synergistic relationship between the two has not been fully explored. Especially when there is a temporal misalignment between pulmonary structural damage and functional compensation, a single-dimensional assessment can easily mask potential risks, leading to missed diagnoses.

[0003] While existing technologies attempt to establish linear regression models between imaging and physiological parameters, they neglect the nonlinear dynamic coupling characteristics between the two during the rehabilitation process. For example, vital capacity recovery may lag behind the visible improvement in airway expansion in imaging, or there may be regional imbalances in respiratory rate abnormalities and the ratio of alveolar ventilation area. This fragmented analysis makes it difficult to accurately pinpoint the "image-function" recovery deviation and effectively identify hidden signals of early complications. Therefore, there is an urgent need to achieve quantitative identification of the correlation between imaging features and abnormal physiological parameters through cross-modal dynamic correlation modeling, providing an early warning basis for the risk of secondary diseases. Summary of the Invention

[0004] To address the aforementioned technical issues, this paper provides a method for assessing pulmonary function in patients recovering from pneumonia. This method overcomes the limitations of traditional methods where imaging indicators and physiological parameters are often assessed independently, such as quantitative analysis of pulmonary fibrosis via CT scans or monitoring changes in blood oxygen levels alone. The dynamic synergistic relationship between these two methods is not fully explored. Especially when there is a temporal misalignment between pulmonary structural damage and functional compensation, a single-dimensional assessment can easily mask potential risks, leading to missed diagnoses.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for assessing lung function in patients recovering from pneumonia, comprising: Acquire lung imaging data and associated physiological parameters of the patient, including at least blood oxygen saturation, respiratory rate, and vital capacity; Lung functional feature set was extracted based on imaging data, including lung tissue density distribution, airway dilatation, and alveolar ventilation area ratio; Based on multidimensional parameter correlation analysis, a dynamic mapping relationship between the lung function feature set and the associated physiological parameters is established; Abnormal associated parameters are identified by the dynamic mapping relationship between the lung function feature set and the associated physiological parameters. When the deviation between the recovery degree of the image feature and the recovery degree of the corresponding physiological parameter exceeds a preset threshold, it is determined that there are abnormal functional parameters. Based on the database of potential complications matched with abnormal parameter types, the correlation between abnormal parameters and preset disease feature models is analyzed, and a secondary disease risk warning is output.

[0006] Preferably, the step of establishing a dynamic mapping relationship between the lung function feature set and the associated physiological parameters based on multidimensional parameter correlation analysis specifically includes: Patients are classified based on their physical characteristics, which include at least gender, age, and medical history. Standard data on pneumonia recovery periods of patients with similar physical signs and attributes to the patients were selected from historical medical data and used as sample data. Based on sample data, a dynamic mapping relationship between the lung function feature set and related physiological parameters is established by using neural network learning training or by using data statistical analysis.

[0007] Preferably, establishing the dynamic mapping relationship between the lung function feature set and related physiological parameters using neural network learning and training specifically includes: Training is based on a dual-channel spatiotemporal attention network, which includes an image feature encoder and a physiological parameter encoder. The image feature encoder uses a three-dimensional convolutional neural network to extract multi-scale features from the lung function feature set, and the physiological parameter encoder uses a temporal convolutional network to analyze and correlate physiological parameter data to construct a model training dataset. The input features of the model training dataset are image features, and the output features are physiological parameters. The parameters of each node in the neural network are trained based on the model training dataset to obtain a dynamic mapping relationship model between the lung function feature set and the associated physiological parameters. The dynamic mapping relationship model takes the lung function feature set as input and the standard range of each associated physiological parameter as output.

[0008] Preferably, establishing a dynamic mapping relationship between the lung function feature set and related physiological parameters using statistical data analysis specifically includes: Historical medical data similar to the patient's current lung function characteristics are obtained from the sample data and labeled as target data; Based on the standard values ​​of each associated physiological parameter in the target data, the mean and standard deviation are calculated, and the mean minus 6 times the standard deviation is used as the lower limit of the maximum deviation of the associated physiological parameter.

[0009] Preferably, the step of identifying abnormal associated parameters through the dynamic mapping relationship between the lung function feature set and associated physiological parameters, and determining the existence of abnormal functional parameters when the deviation between the recovery degree of image features and the recovery degree of the corresponding physiological parameters exceeds a preset threshold, specifically includes: Standard evaluation thresholds for each associated physiological parameter are obtained based on the dynamic mapping relationship between the lung function feature set and the associated physiological parameters. The patient's measured values ​​of various related physiological parameters are compared with the standard evaluation thresholds for each related physiological parameter. If the measured value of a related physiological parameter exceeds the standard evaluation threshold, the related physiological parameter is judged to have recovered abnormally; otherwise, the related physiological parameter is judged to have recovered normally. If there are associated physiological parameters for abnormal recovery, then it is determined that there are abnormal function parameters, and the associated physiological parameters for abnormal recovery and their values ​​are output; otherwise, it is determined that there are no associated physiological parameters for abnormal recovery.

[0010] Preferably, the step of matching the abnormal parameter type with a potential complication database, analyzing the correlation between abnormal parameters and a preset disease feature model, and outputting a secondary disease risk warning specifically includes: Extract the associated physiological parameters that show abnormal recovery, and construct a feature set of abnormal associated physiological parameters; Several potential complications during pneumonia recovery were identified, and based on medical experience data, a sample-associated physiological parameter feature set for each potential complication was constructed. The similarity between the abnormal associated physiological parameter feature set and the sample associated physiological parameter feature set of potential complications is calculated. If the similarity is higher than the warning threshold, the corresponding potential complication is output and a secondary disease risk warning is output. Otherwise, no response is made.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention significantly improves the early identification accuracy of complications during the recovery period of pneumonia by fusing lung imaging features and dynamic physiological parameters across modalities. Specifically, the dynamic mapping relationship between the lung functional feature set constructed based on the spatiotemporal attention mechanism and the associated physiological parameters can accurately analyze the nonlinear dynamic correlation between imaging features such as alveolar ventilation area ratio and lung tissue density distribution and physiological indicators such as blood oxygen saturation and respiratory rate. When there is a significant deviation between the degree of structural recovery shown in imaging and the actual recovery process of physiological function indicators, such as CT images indicating inflammation absorption but blood oxygen saturation not rising synchronously, the system can automatically trigger an abnormal parameter warning mechanism. By matching a potential complication database and combining abnormal correlation patterns with a preset disease feature model for intelligent comparison, the system can quickly identify potential signals of specific complications. Compared with traditional single-dimensional assessment methods, this method incorporates the synergy between imaging structural repair and physiological function recovery into a quantitative analysis framework, effectively solving the problem of false recovery misjudgment, while shortening the cycle of secondary disease risk assessment, providing a critical time window for clinical intervention, thereby reducing the risk of lung function deterioration and secondary infection after severe pneumonia, and has significant clinical practical value. Attached Figure Description

[0012] Figure 1 The flowchart of the lung function assessment method for pneumonia recovery patients proposed in Example 1 is shown below. Figure 2 This is a flowchart of the method for establishing a dynamic mapping relationship between lung function feature sets and associated physiological parameters, as proposed in Example 2. Figure 3 This is a flowchart of the method for establishing a dynamic mapping relationship between lung function feature sets and associated physiological parameters, as proposed in Example 3. Figure 4 This is a flowchart of the method for identifying abnormal correlation parameters proposed in Example 4; Figure 5 This is a flowchart of the method for analyzing the correlation between abnormal parameters and a preset disease feature model proposed in Example 4; Figure 6 This is an architecture diagram of the electronic devices in this solution; Figure 7 This is a schematic diagram of the computer-readable storage medium structure in this scheme.

[0013] The numbers on the map are: 500 - Electronic device; 501 - Bus; 502 - CPU; 503 - ROM; 504 - RAM; 505 - Communication port; 506 - Input / output component; 507 - Hard disk; 508 - User interface; 600 - Computer-readable storage medium. Detailed Implementation

[0014] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0015] Example 1:

[0016] Reference Figure 1 As shown in this embodiment, a method for assessing lung function in patients recovering from pneumonia is proposed, including: Obtain the patient's lung imaging data and related physiological parameters, including at least blood oxygen saturation, respiratory rate, and vital capacity; By integrating multimodal data sources, such as lung images from CT and MRI scans and real-time physiological parameters monitored by wearable devices, the comprehensiveness and dynamism of the assessment are ensured. Blood oxygen saturation reflects oxygenation capacity, while respiratory rate and vital capacity directly characterize ventilation function. Combining these three metrics allows for a multidimensional capture of the respiratory status of patients in the recovery period. Imaging data provides direct evidence of structural damage, while dynamic physiological parameters reflect the progress of functional recovery, laying a data foundation for subsequent correlation modeling.

[0017] Lung functional feature set was extracted based on imaging data, including lung tissue density distribution, airway dilatation, and alveolar ventilation area ratio; Quantitative analysis of lung images was performed using 3D reconstruction algorithms and image segmentation techniques. Lung tissue density distribution was used to assess the degree of inflammation absorption or fibrosis, airway expandability quantified airway resistance through changes in tube diameter, and alveolar ventilation area ratio measured the proportion of effectively ventilated area. These features construct an imaging functional assessment system from three levels: structural damage, airway dynamics, and gas exchange efficiency, providing key indicators for cross-modal correlation analysis.

[0018] Based on multidimensional parameter correlation analysis, a dynamic mapping relationship between the lung function feature set and the associated physiological parameters is established; Abnormal associated parameters are identified by the dynamic mapping relationship between lung function feature set and associated physiological parameters. When the deviation between the recovery degree of image features and the recovery degree of corresponding physiological parameters exceeds a preset threshold, it is determined that there are abnormal functional parameters. The expected recovery range of physiological parameters is generated based on a dynamic mapping relationship. If the measured values ​​consistently deviate from the range, such as when imaging shows that airway expansion is within the target range but vital capacity recovery is insufficient, an abnormality marker is triggered. The deviation threshold is dynamically adjusted through historical rehabilitation data statistics or clinical expert consensus, balancing sensitivity and specificity. Abnormal parameters may indicate that local structural repair has not translated into overall functional improvement, or that there are hidden complications interfering.

[0019] Based on the database of potential complications matched with abnormal parameter types, the correlation between abnormal parameters and preset disease feature models is analyzed, and a secondary disease risk warning is output.

[0020] The potential complication database integrates common secondary conditions following pneumonia, such as typical patterns of pleural effusion. For example, abnormal lung tissue density accompanied by elevated respiratory rate may be associated with interstitial lung disease. The system matches abnormal parameter combinations with disease characteristics through similarity calculations to generate risk levels and specific warning suggestions to assist clinical decision-making.

[0021] Example 2:

[0022] Reference Figure 2 As shown in Example 1, this example uses multidimensional parameter correlation analysis to establish a dynamic mapping relationship between the lung function feature set and the associated physiological parameters. The steps are as follows: Patients are classified based on their physical characteristics, which include at least gender, age, and medical history. Standard data on pneumonia recovery periods of patients with similar physical signs and attributes to the patients were selected from historical medical data and used as sample data. Based on sample data, a dynamic mapping relationship between lung function feature set and related physiological parameters is established by using neural network learning and training. Specifically, the steps for learning and training a neural network are as follows: Training is based on a dual-channel spatiotemporal attention network, which includes an image feature encoder and a physiological parameter encoder. The image feature encoder uses a three-dimensional convolutional neural network to extract multi-scale features from the lung function feature set, while the physiological parameter encoder uses a temporal convolutional network to analyze and correlate physiological parameter data to construct a model training dataset. The input features of the model training dataset are image features, and the output features are physiological parameters. The parameters of each node in the neural network are trained based on the model training dataset to obtain a dynamic mapping relationship model between the lung function feature set and the associated physiological parameters. The dynamic mapping relationship model takes the lung function feature set as input and the standard range of each associated physiological parameter as output.

[0023] By integrating individual patient vital signs and attributes with cross-modal data for dynamic modeling, the personalized accuracy of pulmonary function assessment and complication early warning efficacy for patients in the pneumonia recovery period are significantly improved. Specifically, based on patient classification by gender, age, and medical history records, and screening of historical similar case data, a highly adaptable sample training set is constructed to ensure that the model fully learns the specific correlation patterns between imaging features and physiological parameters in different vital sign groups. Through a dual-channel spatiotemporal attention network design, the image feature encoder uses a three-dimensional convolutional neural network to extract multi-scale spatial structural information, such as changes in alveolar microstructure, while the physiological parameter encoder uses a temporal convolutional network to capture the dynamic evolution trends of indicators such as blood oxygen saturation and respiratory rate. Combined with the cross-modal attention mechanism, the nonlinear coupling relationship between the two in the spatiotemporal dimension is accurately quantified. The dynamically mapped relationship model generated by the training can output individualized physiological parameter standard ranges. For example, for elderly patients with a history of chronic obstructive pulmonary disease, the expected correlation threshold between airway expansion and vital capacity is automatically corrected, effectively overcoming the generalization limitations of traditional population standardization models. The resulting abnormal correlation identification can not only indicate the synergistic deviation between local image restoration and overall functional recovery, but also screen for high-risk secondary diseases through a complication pattern library associated with physical signs and attributes. For example, it can identify the tendency of pulmonary fibrosis in elderly patients or the risk of infection recurrence in people with a history of chronic diseases. This provides clinical decision support that is both accurate and interpretable, shortens the rehabilitation assessment cycle, and reduces the rate of misdiagnosis and missed diagnosis.

[0024] Example 3:

[0025] Reference Figure 3 As shown in Example 1, this example uses multidimensional parameter correlation analysis to establish a dynamic mapping relationship between the lung function feature set and the associated physiological parameters. The steps are as follows: Patients are classified based on their physical characteristics, which include at least gender, age, and medical history. Standard data on pneumonia recovery periods of patients with similar physical signs and attributes to the patients were selected from historical medical data and used as sample data. Based on sample data, a dynamic mapping relationship between lung function feature set and related physiological parameters was established using statistical data analysis. Specifically, the steps for data statistical analysis are as follows: Historical medical data similar to the patient's current lung function characteristics are obtained from the sample data and labeled as target data; Based on the standard values ​​of each associated physiological parameter in the target data, the mean and standard deviation are calculated, and the mean minus 6 times the standard deviation is used as the lower limit of the maximum deviation of the associated physiological parameter.

[0026] The specific calculation formula is as follows: ; in, This represents the lower limit of the maximum deviation of the associated physiological parameters. This represents the average of the associated physiological parameters. This represents the standard deviation of the associated physiological parameters.

[0027] Based on the user's situation, a more similar group of rehabilitation patients is selected. Combining the influence of the rehabilitation patient group on physiological parameter characteristics, the 6-sigma criterion is adopted, that is, the lower limit threshold of physiological parameter recovery is determined by subtracting 6 standard deviations from the mean. Based on the normal distribution criterion, the probability that the physiological parameters of patients in a normal recovery state are above the lower limit threshold is 99.99%. Based on the Chebyshev lower limit calculation, the probability that the physiological parameters of patients in a normal recovery state are above the lower limit threshold is 98.61%. This method can quickly determine the standard recovery indicators of patients.

[0028] This embodiment simplifies the complex model training process, achieving both efficiency and reliability in assessing the recovery of physiological parameters in pneumonia patients during the recovery period through data statistical analysis and individualized group matching strategies. Specifically, based on patient vital signs and attributes, standard data from historically similar groups are selected to construct a highly correlated target dataset, ensuring that the statistical model is highly adapted to the current patient's recovery characteristics. The lower limit threshold for physiological parameter recovery is dynamically calculated using the six-sigma criterion. When the data follows a normal distribution, it can cover the physiological parameter range of 99.99% of normal recovering patients. For non-normally distributed scenarios, Chebyshev's inequality ensures that at least 98.61% of normal values ​​are above the threshold, thus rapidly establishing statistically strongly correlated anomaly discrimination criteria without the need for complex model training. For example, for the vital capacity index of elderly COPD patients, personalized thresholds are calculated using group data, effectively avoiding misjudgments caused by overly broad or strict group standards. This method not only significantly shortens the abnormal parameter identification cycle, but also is compatible with the data distribution characteristics of different sign groups, providing statistically interpretable quantitative evidence for the deviation between imaging features and physiological parameter recovery, thereby accurately locating early signals of complications such as pulmonary fibrosis and restrictive ventilation disorders, significantly reducing the risk of missed diagnosis, and providing efficient and robust individualized rehabilitation assessment support for clinical practice.

[0029] Example 4:

[0030] Reference Figure 4 As shown, this embodiment, based on Embodiment 1, further proposes to identify abnormal associated parameters through the dynamic mapping relationship between lung function feature sets and associated physiological parameters. When the deviation between the recovery degree of image features and the recovery degree of the corresponding physiological parameters exceeds a preset threshold, it is determined that there are abnormal functional parameters, specifically including: Standard evaluation thresholds for each associated physiological parameter are obtained based on the dynamic mapping relationship between the lung function feature set and the associated physiological parameters. The patient's measured values ​​of various related physiological parameters are compared with the standard evaluation thresholds for each related physiological parameter. If the measured value of a related physiological parameter exceeds the standard evaluation threshold, the related physiological parameter is judged to have recovered abnormally; otherwise, the related physiological parameter is judged to have recovered normally. If there are associated physiological parameters for abnormal recovery, then it is determined that there are abnormal function parameters, and the associated physiological parameters for abnormal recovery and their values ​​are output; otherwise, it is determined that there are no associated physiological parameters for abnormal recovery.

[0031] By employing dynamic mapping relationships and precise threshold determination mechanisms, the sensitivity and clinical applicability of identifying functional abnormalities in pneumonia recovery patients are significantly improved. Specifically, individualized standard evaluation thresholds generated based on the dynamic mapping relationship between pulmonary functional feature sets and associated physiological parameters fully integrate patient vital signs and the distribution patterns of historical similar group data. For example, the vital capacity threshold for elderly patients is dynamically adjusted according to their airway expansion characteristics. Compared to traditional static threshold methods, this avoids misjudgments caused by mismatches between group standards and individual recovery trajectories. By comparing the deviation between measured physiological parameters and standard thresholds in real time, such as when measured blood oxygen saturation values ​​are lower than the model's expected lower limit, the system can quickly identify asynchronous abnormalities in imaging features and physiological function recovery. For instance, if the alveolar ventilation area ratio meets the standard but the respiratory rate is still higher than the threshold, it suggests the possible presence of occult airway hyperresponsiveness or local ventilation / perfusion mismatch. The targeted output of abnormal parameters provides clinicians with intuitive quantitative evidence, supporting their rapid identification of the risk of complications such as pulmonary fibrosis progression, secondary infections, or restrictive ventilatory dysfunction, thereby shortening the diagnostic decision-making cycle. This method incorporates the dynamic synergy between image structure repair and physiological function recovery into the abnormality criterion, effectively solving the "false negative" problem in traditional assessment, reducing the rate of missed diagnosis due to functional compensation masking true damage, and providing key technical support for precise intervention and prognosis management.

[0032] Reference Figure 5 As shown, based on Embodiment 1, this embodiment further proposes to analyze the correlation between abnormal parameters and a preset disease feature model by matching the abnormal parameter type with a potential complication database, and output a secondary disease risk warning, specifically including: Extract the associated physiological parameters that show abnormal recovery, and construct a feature set of abnormal associated physiological parameters; Several potential complications during pneumonia recovery were identified, and based on medical experience data, a sample-associated physiological parameter feature set for each potential complication was constructed. The similarity between the abnormal associated physiological parameter feature set and the sample associated physiological parameter feature set of potential complications is calculated. If the similarity is higher than the warning threshold, the corresponding potential complication is output and a secondary disease risk warning is output. Otherwise, no response is made.

[0033] A feature set of potential complications, constructed based on medical experience data, covers typical abnormal physiological parameter patterns of common secondary diseases such as pulmonary fibrosis and pleural effusion. For example, persistently low vital capacity coupled with fluctuating blood oxygen saturation may indicate the progression of interstitial lung disease. By extracting abnormally associated physiological parameter feature sets from patients and calculating similarity with preset disease feature models, the system can quickly identify the correlation strength between specific parameter combinations and target complications. When the similarity exceeds a dynamic warning threshold, a targeted warning is automatically triggered. For example, the combination of abnormally elevated respiratory rate and regional imbalance in alveolar ventilation area ratio can accurately match the disease feature model of local pulmonary ventilation impairment. This method transforms clinical expert experience into quantifiable feature matching rules, effectively solving the problems of low efficiency and large subjective bias in manual screening. At the same time, the dynamic threshold mechanism avoids over-warning, ensuring early targeted identification of high-risk complications. Clinicians can prioritize screening for high-risk diseases based on the warning results, shorten the diagnosis cycle, and intervene in advance, thereby reducing the risk of irreversible lung function damage after severe pneumonia and providing intelligent decision support for patient rehabilitation management.

[0034] Furthermore, the method according to the embodiments of this application can also be achieved by means of... Figure 6 The architecture of the electronic device shown is used to implement this. For example... Figure 6 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, ROM 503, RAM 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store a method for assessing lung function in patients recovering from pneumonia, as provided in this application. The electronic device 500 may also include a user interface 508. Of course, Figure 6 The architecture shown is merely exemplary and can be omitted as needed when implementing different devices. Figure 6 One or more components in the illustrated electronic device.

[0035] Figure 7 This is a schematic diagram of a computer-readable storage medium structure provided in one embodiment of this application. Figure 7The diagram illustrates a computer-readable storage medium 600 according to one embodiment of this application. The computer-readable storage medium 600 stores computer-readable instructions. When executed by a processor, the computer-readable instructions can perform a method for assessing lung function in a pneumonia recovery patient according to an embodiment of this application, as described with reference to the above figures. The computer-readable storage medium 600 includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0036] In summary, the advantages of this invention are: by fusing lung imaging features and dynamic physiological parameters across modalities, it significantly improves the early identification accuracy of complications during the pneumonia recovery period. Specifically, the dynamic mapping relationship between the lung functional feature set constructed based on the spatiotemporal attention mechanism and the associated physiological parameters can accurately analyze the nonlinear dynamic correlation between imaging features such as alveolar ventilation area ratio and lung tissue density distribution and physiological indicators such as blood oxygen saturation and respiratory rate. When there is a significant deviation between the degree of structural recovery shown in imaging and the actual recovery process of physiological function indicators, such as CT images indicating inflammation absorption but blood oxygen saturation not rising synchronously, the system can automatically trigger an abnormal parameter warning mechanism. By matching a potential complication database and combining intelligent comparison of abnormal correlation patterns with preset disease feature models, the system can quickly identify potential signals of specific complications. Compared with traditional single-dimensional assessment methods, this method incorporates the synergy between imaging structural repair and physiological function recovery into a quantitative analysis framework, effectively solving the problem of false recovery misjudgment, while shortening the cycle of secondary disease risk assessment, providing a critical time window for clinical intervention, thereby reducing the risk of lung function deterioration and secondary infection after severe pneumonia, and has significant clinical practical value.

[0037] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

[0038] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0039] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

Claims

1. A method for assessing lung function in patients recovering from pneumonia, characterized in that, include: Acquire lung imaging data and associated physiological parameters of the patient, including at least blood oxygen saturation, respiratory rate, and vital capacity; Lung functional feature set was extracted based on imaging data, including lung tissue density distribution, airway dilatation, and alveolar ventilation area ratio; Based on multidimensional parameter correlation analysis, a dynamic mapping relationship between the lung function feature set and the associated physiological parameters is established; Abnormal associated parameters are identified by the dynamic mapping relationship between the lung function feature set and the associated physiological parameters. When the deviation between the recovery degree of the image feature and the recovery degree of the corresponding physiological parameter exceeds a preset threshold, it is determined that there are abnormal functional parameters. Based on the database of potential complications matched with abnormal parameter types, the correlation between abnormal parameters and preset disease feature models is analyzed, and a secondary disease risk warning is output. The analysis based on multidimensional parameter correlation, establishing a dynamic mapping relationship between the lung function feature set and the associated physiological parameters, specifically includes: Patients are classified based on their physical characteristics, which include at least gender, age, and medical history. Standard data on pneumonia recovery periods of patients with similar physical signs and attributes to the patients were selected from historical medical data and used as sample data. Based on sample data, a dynamic mapping relationship between the lung function feature set and related physiological parameters is established by using neural network learning training or by using data statistical analysis.

2. The method for assessing lung function in patients recovering from pneumonia according to claim 1, characterized in that, Establishing a dynamic mapping relationship between the lung function feature set and associated physiological parameters using neural network learning and training specifically includes: Training is based on a dual-channel spatiotemporal attention network, which includes an image feature encoder and a physiological parameter encoder. The image feature encoder uses a three-dimensional convolutional neural network to extract multi-scale features from the lung function feature set, and the physiological parameter encoder uses a temporal convolutional network to analyze and correlate physiological parameter data to construct a model training dataset. The input features of the model training dataset are image features, and the output features are physiological parameters. The parameters of each node in the neural network are trained based on the model training dataset to obtain a dynamic mapping relationship model between the lung function feature set and the associated physiological parameters. The dynamic mapping relationship model takes the lung function feature set as input and the standard range of each associated physiological parameter as output.

3. The method for assessing lung function in patients recovering from pneumonia according to claim 2, characterized in that, Establishing a dynamic mapping relationship between the lung function feature set and associated physiological parameters using statistical data analysis specifically includes: Historical medical data similar to the patient's current lung function characteristics are obtained from the sample data and labeled as target data; Based on the standard values ​​of each associated physiological parameter in the target data, the mean and standard deviation are calculated, and the mean minus 6 times the standard deviation is used as the lower limit of the maximum deviation of the associated physiological parameter.

4. A method for assessing lung function in patients recovering from pneumonia according to any one of claims 2-3, characterized in that, The step of identifying abnormal associated parameters through the dynamic mapping relationship between the lung function feature set and associated physiological parameters, and determining the existence of abnormal functional parameters when the deviation between the recovery degree of image features and the recovery degree of the corresponding physiological parameters exceeds a preset threshold, specifically includes: Standard evaluation thresholds for each associated physiological parameter are obtained based on the dynamic mapping relationship between the lung function feature set and the associated physiological parameters. The patient's measured values ​​of various related physiological parameters are compared with the standard evaluation thresholds for each related physiological parameter. If the measured value of a related physiological parameter exceeds the standard evaluation threshold, the related physiological parameter is judged to have recovered abnormally; otherwise, the related physiological parameter is judged to have recovered normally. If there are associated physiological parameters for abnormal recovery, then it is determined that there are abnormal function parameters, and the associated physiological parameters for abnormal recovery and their values ​​are output; otherwise, it is determined that there are no associated physiological parameters for abnormal recovery.

5. The method for assessing lung function in patients recovering from pneumonia according to claim 4, characterized in that, The process of matching abnormal parameter types with a potential complication database, analyzing the correlation between abnormal parameters and a preset disease feature model, and outputting a secondary disease risk warning specifically includes: Extract the associated physiological parameters that show abnormal recovery, and construct a feature set of abnormal associated physiological parameters; Several potential complications during pneumonia recovery were identified, and based on medical experience data, a sample-associated physiological parameter feature set for each potential complication was constructed. The similarity between the abnormal associated physiological parameter feature set and the sample associated physiological parameter feature set of potential complications is calculated. If the similarity is higher than the warning threshold, the corresponding potential complication is output and a secondary disease risk warning is output. Otherwise, no response is made.

6. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform a method for assessing lung function in patients recovering from pneumonia as described in any one of claims 1-5.

7. A computer-readable storage medium storing computer-readable instructions, characterized in that, When the computer-readable instructions are executed by a processor, they implement a method for assessing lung function in patients recovering from pneumonia, as described in any one of claims 1-5.

Images