Analysis Method, System, Device and Medium for Nuclear Medicine Imaging Radiation Report

Abstract

The present invention relates to the field of medical imaging technology, and specifically provides an analysis method, system, device, and medium for nuclear medicine imaging radiation reports. By correlating the extracted characteristic values of the patient's medical imaging data with the characteristic values of the patient's own historical medical imaging data, the comprehensive correlation coefficient between the patient's disease condition and their own historical medical imaging is evaluated, which can fully consider the individual trajectory of the patient's disease development. By correlating the extracted characteristic values of the patient's clinical data with the characteristic values of the clinical data of the same type of patient group, the population deviation degree of the patient's clinical data is calculated, making it more likely to detect new changes in the patient's condition at an early stage. Based on the comprehensive correlation coefficient between the patient's disease condition and their own historical medical imaging and the population deviation degree of the patient's clinical data, a model is constructed to predict the category of the patient's disease progression, enabling a more comprehensive analysis of the patient's condition and better management of the patient's disease process.

Description

Technical Field

[0001] The present invention relates to the field of nuclear medicine imaging technology, and in particular to an analysis method, system, device and medium for nuclear medicine imaging radiological reports. Background Art

[0002] As an important branch of the medical field, nuclear medicine, with the help of radionuclide tracer technology, can provide in-depth insights into the physiological and pathological processes of the human body at the molecular and cellular levels. Nuclear medicine imaging radiology reports are the key summary and interpretation of these examination results, providing an indispensable basis for clinical diagnosis, treatment plan formulation and disease monitoring.

[0003] With the rapid development of medical technology, nuclear medicine imaging is increasingly used in clinical practice, bringing revolutionary breakthroughs in the diagnosis and treatment of many diseases. However, the existing nuclear medicine imaging analysis technology still has many shortcomings that need to be addressed.

[0004] For example, the existing Chinese patent with application number 202410933655.8 discloses an analysis method, system, equipment and medium for nuclear medicine image radiology reports. The solution ensures that the data in nuclear medicine and radiology reports can be referenced to each other by establishing a corresponding relationship, performs two-way search and matching of data between nuclear medicine and radiology reports, and combines radiology structured report information with nuclear medicine reference navigation maps to enhance the accuracy and efficiency of diagnosis. After the nuclear medicine report is completed, it is compared with the radiology report, solving the problem of low efficiency and easy omissions in the current manual analysis of nuclear medicine image radiology reports.

[0005] However, the above patent has the following problems: First, the solution analyzes and evaluates the imaging data of a single examination, and does not involve a comprehensive analysis of the patient's own historical nuclear medicine imaging data corresponding to the data. It may be difficult to detect subtle progression of the disease at the individual level, and cannot fully consider the individual trajectory of the patient's disease development, and has limited ability to detect new changes in the patient's condition at an early stage.

[0006] Second, this solution does not involve placing the individual patient medical imaging data corresponding to the report in the context of a similar patient group, making it difficult to determine whether the symptoms reflected in the report generated based on the individual patient's medical imaging data are within the normal fluctuation range of the group or are abnormal. During the analysis process, this solution does not involve the differences in medical imaging data between individual patients, and uses a single standard to analyze the medical imaging data of all individual patients. As a result, important clues to the development of the disease may be missed, and the unique progression of the patient's disease cannot be accurately identified.

[0007] As the existing Chinese patent with the application number 202410966597.9 discloses a method and system for interpreting medical imaging reports based on a multi-layer deep learning model. This solution preprocesses the collected historical reports, judges the impact of text data, then constructs a model with a specific structure, divides the data into training, validation, and test sets, and obtains the optimal model through training, evaluation, parameter optimization, and fine-tuning. Finally, the preprocessed data of the report to be explained is input into the model to obtain the interpretation result, solving the problem of poor readability of medical imaging reports in the prior art.

[0008] However, the following problems exist in the above patent: This solution focuses on the medical imaging report to be explained itself and does not combine the patient's real-time data or other relevant clinical information, making it difficult to comprehensively reflect the patient's condition. The medical imaging report is only part of the basis for diagnosis. Only by combining the patient's real-time data and other relevant clinical information can we more accurately grasp the overall condition of the patient and thus make more appropriate medical decisions. Summary of the Invention

[0009] To overcome the shortcomings in the background technology, the embodiments of the present invention provide an analysis method, system, device, and medium for nuclear medicine imaging radiation reports, which can effectively solve the problems involved in the above background technology.

[0010] The object of the present invention can be achieved through the following technical solutions: In the first aspect of the present invention, an analysis method for nuclear medicine imaging radiation reports is provided. The method includes the following steps: S1. Obtain the nuclear medicine imaging data and clinical data of the patient.

[0011] S2. Extract the eigenvalue of the patient's nuclear medicine imaging data and clinical data respectively to obtain the eigenvalue of the patient's nuclear medicine imaging data and the eigenvalue of the patient's clinical data.

[0012] S3. Associate the extracted eigenvalue of the patient's nuclear medicine imaging data with the eigenvalue of the patient's own historical nuclear medicine imaging data, and evaluate the comprehensive correlation coefficient between the patient's disease and the patient's own historical nuclear medicine imaging by assigning a correlation factor.

[0013] S4. Associate the extracted eigenvalue of the patient's clinical data with the eigenvalue of the clinical data of the same type of patient group, and calculate the group deviation degree of the patient's clinical data.

[0014] S5. Construct a model based on the comprehensive correlation coefficient between the patient's disease and the patient's own historical nuclear medicine imaging and the group deviation degree of the patient's clinical data to predict the disease progression category of the patient.

[0015] The patient's clinical data includes the patient's gene sequencing data, physical data, and medical history text data.

[0016] According to an implementable manner of the first aspect of the present invention, the specific analysis method for step S2 is as follows: S21. Obtain the historical nuclear medicine image data of each part from the management database, input it into the deep learning model for training, use the trained model to analyze the patient's nuclear medicine image, identify each part in the patient's nuclear medicine image, perform model segmentation, and at the same time extract the feature representations corresponding to each part to form the feature vectors of each part, which are regarded as the eigenvalue of the nuclear medicine image data of the patient.

[0017] S22. Obtain the gene sequencing data of the patient, compare it with the reference genome to determine the position of each sequencing fragment in the reference genome. If the position of a certain sequencing fragment in the reference genome coincides with the known disease-related gene region and there are differences between this sequencing fragment and the reference genome, it is determined as a gene mutation site. Screen out the gene mutation sites related to specific diseases and calculate the occurrence frequency of the specific mutation sites of the patient.

[0018] S23. Collect the physical data of the patient and calculate the derived features to obtain the patient's BMI and body fat percentage.

[0019] S24. Read the medical history text data of the patient, perform structured processing on the unstructured medical history text data, count the number of attacks of specific diseases from the structured medical history data, and calculate the disease attack frequency in combination with time information.

[0020] S25. Extract the degree of symptom improvement after each treatment of the patient, quantify it into specific indicators and classify them. The degree of symptom improvement is divided into significant improvement, partial improvement, and no improvement.

[0021] S26. Regard the occurrence frequency of the patient's specific mutation sites, BMI, body fat percentage, disease attack frequency, and degree of symptom improvement as the eigenvalue of the clinical data of the patient.

[0022] According to an implementable manner of the first aspect of the present invention, the specific operation method for allocating the association factor is as follows: A1. Obtain each historical nuclear medicine image of the patient himself, label the corresponding shooting time point for each of them, sort the historical nuclear medicine images of the patient himself according to the clinical required time period according to the time point, allocate the largest time association factor to the historical nuclear medicine image with the closest time point, and calculate the time association factors of the remaining historical nuclear medicine images of the patient himself using the linear attenuation function.

[0023] A2. Determine the target part of the current disease diagnosis, set the basic part association factor, and adjust the part association factor according to the influence degree of the remaining parts that have an impact on the target part.

[0024] According to an implementable manner of the first aspect of the present invention, the specific operation method of the comprehensive correlation coefficient between the patient's disease condition and his own historical nuclear medicine images is as follows: Using the feature vectors of each part, match the target parts in each historical nuclear medicine image of the patient himself to obtain the historical nuclear medicine images of the patient himself corresponding to the target parts, extract the feature changes of the target parts therein, record them as the feature change amounts of each target part, and calculate the comprehensive correlation coefficient between the patient's disease condition and his own historical nuclear medicine images. : F = ∑ p = 1 l { C p ∗ [ F max ∗ e − k ∗ ( T now − T p Δ T ) ] + ∑ i = 1 n ( F target ∗ I i ) } , where represents the feature change amount of the -th target part, is the number of the -th target part, , represents the time point when the nuclear medicine image of the current patient is taken, represents the time point when the historical nuclear medicine image of the patient himself corresponding to the -th target part is taken, is the set maximum time correlation factor, represents the preset time decay factor, represents the preset reference shooting interval duration, is the natural constant, represents the basic part correlation factor, represents the influence factor of the part corresponding to the -th historical nuclear medicine image of the patient himself on the target part, represents the number of the -th historical nuclear medicine image of the patient himself, , when the part corresponding to the -th historical nuclear medicine image is the target part, .

[0025] According to an implementable manner of the first aspect of the present invention, the specific analysis method of step S4 is as follows: S41. Based on the clinical data of the patient, determine the inclusion criteria for patients of the same type, obtain the clinical data of the group of patients of the same type that meet the inclusion criteria from the management database, and extract the features of the clinical data of the group of patients of the same type.

[0026] S42. Quantitatively encode the degree of symptom improvement in the clinical data of the patient and the clinical data of the group of patients of the same type, and then calculate the means and standard deviations of the clinical data of the group of patients of the same type.

[0027] S43. Calculate the difference between each eigenvalue of the patient's clinical data and the mean of the clinical data corresponding to the same type of patient group, and then divide it by the standard deviation of the occurrence frequency of the mutation site, BMI, body fat percentage, disease attack frequency, and symptom improvement degree of the same type of patient group to confirm the relative position of each eigenvalue of the patient's clinical data among the eigenvalues of the clinical data of the same type of patient group.

[0028] S44. Calculate the population deviation degree of the patient's clinical data according to the relative position of each eigenvalue of the patient's clinical data among the eigenvalues of the clinical data of the same type of patient group. : , where respectively represent the relative positions of the occurrence frequency of the specific mutation site, BMI, body fat percentage, disease attack frequency, and symptom improvement degree of the patient among the corresponding eigenvalues of the clinical data of the same type of patient group. respectively represent the weight factors corresponding to the occurrence frequency of the specific mutation site, BMI, body fat percentage, disease attack frequency, and symptom improvement degree.

[0029] According to an achievable manner of the first aspect of the present invention, the specific operation method of step S5 is as follows: S51. Analyze each patient of the same type based on the comprehensive correlation coefficient between the patient's disease condition and his own historical nuclear medicine images and the analysis method of the population deviation degree of the patient's clinical data, and divide the patients into "low-risk", "medium-risk", and "high-risk" disease progression categories according to the set classification criteria, thereby constructing sample sets for each disease progression category.

[0030] S52. Using each feature and its eigenvalue of the clinical data in each disease progression category sample set as the splitting point, calculate the Gini index value of each feature in each disease progression category sample set, select the feature and its eigenvalue with the smallest Gini index, record this feature as the optimal splitting feature, record this eigenvalue as the optimal splitting point, divide each disease progression category sample set according to the optimal splitting point, and recursively repeat calculating the Gini index, selecting the optimal splitting feature and splitting point for the obtained subsets until the number of samples in the subset is less than the set threshold, thereby constructing each decision tree to form a random forest model.

[0031] S53. Input the nuclear medicine image data and clinical data of the selected patient into each decision tree in the random forest in turn, and output the prediction of the disease progression category of the selected patient.

[0032] The second aspect of the present invention proposes an analysis system for nuclear medicine image radiology reports, which specifically includes the following modules: a medical data acquisition module for acquiring the nuclear medicine image data and clinical data of the patient.

[0033] A feature extraction module for extracting eigenvalue according to the acquired nuclear medicine image data and clinical data of the patient respectively.

[0034] A historical data association module is used to associate the extracted characteristic values of the patient's nuclear medicine image data with the characteristic values of the patient's own historical nuclear medicine image data, and evaluate the comprehensive correlation coefficient between the patient's disease condition and the patient's own historical nuclear medicine images by assigning correlation factors.

[0035] A patient data association module is used to associate the extracted characteristic values of the patient's clinical data with the characteristic values of the clinical data of the same type of patient group, and calculate the group deviation degree of the patient's clinical data.

[0036] A model construction module is used to construct a model based on the comprehensive correlation coefficient between the patient's disease condition and the patient's own historical nuclear medicine images and the group deviation degree of the patient's clinical data, and predict the category of the patient's disease progression.

[0037] In a third aspect of the present invention, an analysis device for nuclear medicine image radiology reports is proposed. The device includes a processor, a memory and a communication bus, and a computer-readable program executable by the processor is stored on the memory.

[0038] The communication bus realizes the connection and communication between the processor and the memory.

[0039] When the processor executes the computer-readable program, the steps in an analysis method for nuclear medicine image radiology reports as described in the present invention are realized.

[0040] In a fourth aspect of the present invention, an analysis medium for nuclear medicine image radiology reports is provided. The medium is burned with a computer program, and when the computer program runs in the memory of the server, the analysis method for nuclear medicine image radiology reports as described in the present invention is realized.

[0041] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: First, the present invention associates the extracted characteristic values of the patient's nuclear medicine image data with the characteristic values of the patient's own historical nuclear medicine image data, and evaluates the comprehensive correlation coefficient between the patient's disease condition and the patient's own historical nuclear medicine images, which can fully consider the individual trajectory of the patient's disease development. Even if the changes in the disease condition are relatively subtle, compared with the methods that only use single images or do not consider the individual historical situation, it is more likely to detect new changes in the patient's condition at an early stage.

[0042] Second, the present invention associates the extracted characteristic values of the patient's clinical data with the characteristic values of the clinical data of the same type of patient group, and calculates the group deviation degree of the patient's clinical data, which can measure the abnormality degree of the patient relative to the same type of patient group, and helps to accurately identify those individuals at high risk in the group.

[0043] III. The present invention constructs a model based on the comprehensive correlation coefficient between a patient's disease condition and their own historical nuclear medicine images, as well as the population deviation of the patient's clinical data, to predict the category of the patient's disease progression. The multi-dimensional analysis method can consider the factors affecting disease progression more comprehensively compared to a single-dimensional prediction model, thereby improving the accuracy of predicting the category of disease progression. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described with reference to the accompanying drawings. However, the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the following drawings.

[0045] Figure 1 It is a flowchart of the method implementation steps of the present invention.

[0046] Figure 2 It is a schematic diagram of the connection of system modules of the present invention.

[0047] Figure 3 It is a schematic diagram of an embodiment of a device in an example of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0049] Embodiment 1

[0050] See Figure 1 As shown, the present invention provides an analysis method for nuclear medicine image radiation reports, including the following steps: S1. Obtain the nuclear medicine image data and clinical data of the patient.

[0051] The patient's clinical data includes the patient's gene sequencing data, physical data, and medical history text data.

[0052] S2. Extract the eigenvalue of the patient's nuclear medicine image data and clinical data respectively to obtain the eigenvalue of the patient's nuclear medicine image data and the eigenvalue of the patient's clinical data.

[0053] The specific analysis method for step S2 is as follows: S21. Obtain the historical nuclear medicine image data of each part from the management database, input it into the deep learning model for training, use the trained model to analyze the patient's nuclear medicine image, identify each part in the patient's nuclear medicine image, perform model segmentation, and at the same time extract the feature representations corresponding to each part to form the feature vectors of each part, regarded as the eigenvalue of the patient's nuclear medicine image data; it helps in the early detection of diseases and can be recognized by the model when there are slight changes in the features of certain parts, thus enabling early intervention.

[0054] It should be noted that the specific training steps of the deep learning model are as follows: S211. Divide the historical nuclear medicine image data of each part into a training set, a validation set, and a test set according to a set ratio, and perform the operation of normalizing the gray value to map the gray value of the historical nuclear medicine image data of each part to a specific range.

[0055] S212. Select a loss function and an optimization algorithm, input the preprocessed historical nuclear medicine image data of each part into the selected deep learning model, calculate the prediction result of the model through forward propagation, use the loss function to calculate the loss value between the prediction result and the true label, calculate the gradient of the loss value with respect to the model parameters through the backpropagation algorithm, and use the optimization algorithm to update the model parameters according to the calculated gradient.

[0056] S213. After each training round ends, calculate the performance metrics of the model on the validation set (a part of the data reserved during data division and not participating in training), record the loss value and the performance metrics of the validation set during the training process, draw the loss curve and the performance metric curve. When the performance metrics of the validation set no longer improve, it is considered that the model training converges and the training process ends.

[0057] It should be noted that in a specific embodiment, for the patient's pulmonary nuclear medicine image, input it into the trained deep learning model. The model identifies the left upper lobe of the lung by analyzing the patient's pulmonary nuclear medicine image. Using the segmentation function of the model, outline the boundary range of the left upper lobe of the lung and obtain its feature representation, such as the average gray value and texture complexity of the boundary region of the left upper lobe of the lung. The average gray value after normalization is 120, and the texture complexity obtained according to the texture calculation method is 0.8. The shape feature is represented by a three-dimensional vector [ 0 . 2 , 0 . 1 , 0 . 3 ] indicating the quantization values representing different shape features. Then the feature vector of the left upper lobe of the lung is [ 120 , 0 . 8 , 0 . 2 , 0 . 1 , 0 . 3 ] .

[0058] S22. Obtain the patient's gene sequencing data, align it with the reference genome to determine the position of each sequencing fragment in the reference genome. If the position of a certain sequencing fragment coincides with the known disease-related gene region in the reference genome and there are differences between this sequencing fragment and the reference genome, it is determined as a gene mutation site. Screen out the gene mutation sites related to specific diseases, calculate the occurrence frequency of specific mutation sites in the patient, and the pathogenic factors or disease risk factors at the gene level of the patient can be clarified.

[0059] It should be noted that the specific operation method for calculating the occurrence frequency of specific mutation sites in the patient is as follows: For each determined gene mutation site related to a specific disease, count the number of times it appears in the patient's gene sequencing data, divide the number of occurrences of each mutation site by the total number of sequencing fragments, and obtain the occurrence frequency of this mutation site in the patient's gene.

[0060] S23. Collect the patient's physical data, perform derivative feature calculations to obtain the patient's BMI and body fat percentage; the patient's body composition can be quantified, the patient's physical condition can be reflected from a macroscopic perspective, and the patient's overall health level and the risk of obesity-related diseases can be evaluated. For example, too high a BMI may be related to an increased risk of various chronic diseases such as cardiovascular diseases and diabetes.

[0061] It should be noted that the physical data includes basic physiological indicators such as height, weight, and waist circumference, which are respectively denoted as , , , 。

[0062] S24. Read the patient's medical history text data, perform structured processing on the unstructured medical history text data, count the number of occurrences of specific diseases from the structured medical history data, and calculate the disease attack frequency in combination with time information; it helps to better judge the development trend and treatment effect of the disease.

[0063] It should be noted that the specific operation method for performing structured processing on the unstructured medical history text data is as follows: S241. Remove the irrelevant symbols, special characters, and garbled codes in the patient's medical history text data, unify the formats of dates, numbers, units, etc. in the text, and use natural language processing tools to split the continuous text into independent phrases.

[0064] S242. Through named entity recognition technology, identify the key entities in the text, such as patient basic information, disease names, symptom manifestations, examination items, treatment methods, and drug names, determine the relationships between each entity, and store the structured information into the corresponding database according to the designed data structure.

[0065] It should be noted that in a specific embodiment, there is a relationship of suffering from a disease between the "patient" and the "disease", a relationship of manifestation between the "disease" and the "symptom", a relationship of treatment between the "treatment means" and the "disease", etc.

[0066] It should be noted that the specific operation method for calculating the disease attack frequency is as follows: Extract the time information related to the disease attack from the structured medical history data, including the visit time, the time when the symptom first appears, and the disease diagnosis time. Unify the extracted time information into a standard time format. According to the disease name and symptom description records in the structured data, determine the independent events of each disease attack, count the number of disease attacks within a set period, and divide the number of disease attacks by the corresponding period duration to obtain the disease attack frequency.

[0067] S25. Extract the degree of symptom improvement after each treatment of the patient, quantify it into specific indicators and classify them. The degree of symptom improvement is divided into significant improvement, partial improvement, and no improvement, which can intuitively evaluate the treatment effect.

[0068] It should be noted that the specific analysis method for the degree of symptom improvement is as follows: Statistically calculate the disease attack frequency of the patient within a set duration before and after each treatment. By dividing the disease attack frequency of the patient within the set duration before each treatment by the disease attack frequency of the patient within the set duration after each treatment, obtain the disease attack reduction rate of each treatment of the patient. Set the disease attack reduction rate ranges for significant improvement, partial improvement, and no improvement respectively, and obtain the degree of symptom improvement after each treatment of the patient by matching.

[0069] S26. Regard the occurrence frequency of specific mutation sites, BMI, body fat percentage, disease attack frequency, and degree of symptom improvement of the patient as the clinical data characteristic values of the patient; Integrating data from multiple aspects into clinical data characteristic values can comprehensively describe the clinical condition of the patient.

[0070] The specific operation method for allocating the correlation factor is as follows: A1. Obtain each historical nuclear medicine image of the patient himself, mark the corresponding shooting time point for each of them, sort the patient's own historical nuclear medicine images according to the clinical required time period based on the time point, allocate the largest time correlation factor to the historical nuclear medicine image with the closest time point, and calculate the time correlation factors of the remaining historical nuclear medicine images of the patient himself using a linear attenuation function; During the diagnosis process, recent images can often reflect the latest state of the patient's current condition and should be given more attention. Earlier images may also contain useful information, but their importance is relatively lower. In this way, the image information at different times can be systematically integrated to improve the accuracy of diagnosis.

[0071] It should be noted that the specific analysis method for calculating the time correlation factors of the remaining historical nuclear medicine images of the patient himself is: Set the slope , Through the formula Calculate the time correlation factor of the remaining historical nuclear medicine images of the patient himself / herself. It represents the time interval between the specified nuclear medicine image shooting time and the most recent nuclear medicine image shooting time. is the intercept; as the time interval increases, the time correlation factor will gradually decrease according to a linear rule, indicating that the images farther from the current time have a lower degree of correlation.

[0072] A2. Determine the target site of the current disease diagnosis, set the basic site correlation factor, and adjust the site correlation factor according to the influence degree of each other site that affects the target site; this can improve the comprehensiveness of the diagnosis. The occurrence and development of many diseases are not limited to the target site itself, but also related to the surrounding or associated sites. For example, heart diseases may be related to lung function, blood vessel conditions, etc. Considering these associated sites and adjusting the correlation factor can more comprehensively evaluate the condition of the target site.

[0073] It should be noted that the specific operation method for adjusting the site correlation factor is: set the basic site correlation factor , and based on the normal physiological functions of the human body, analyze the internal relationship between the corresponding site and the target site, and thus set the influence factor of each site on the target site.

[0074] S3. Associate the extracted characteristic values of the patient's nuclear medicine image data with the characteristic values of the patient's own historical nuclear medicine image data, and evaluate the comprehensive correlation coefficient between the patient's disease and his / her own historical nuclear medicine image by assigning correlation factors.

[0075] The specific operation method for the comprehensive correlation coefficient between the patient's disease and his / her own historical nuclear medicine image is: use the characteristic vectors of each site to match the target site in each of the patient's own historical nuclear medicine images, obtain the patient's own historical nuclear medicine images corresponding to the target site, and extract the characteristic changes of the target site therein, denoted as the characteristic change amounts of each target site; this can accurately locate the situation of the target site in different historical images, which helps to more accurately judge the development process of the target site disease.

[0076] It should be noted that in a specific embodiment, 5 nuclear medicine images of the lungs taken by a specified patient every year in the past 5 years are obtained, and corresponding timestamps are marked for these 5 nuclear medicine images. For example, the earliest one was taken on January 1, 2020, denoted as t1, the one taken on January 5, 2021 is denoted as t2, and so on to obtain the nuclear medicine images of the patient's lungs at each time point.

[0077] Use the method in step S21 to obtain the feature vectors of each part. For the right middle lobe of the patient's lung, match the same right middle lobe in the patient's pulmonary nuclear medicine images at each time point. Then, for the well-matched right middle lobe, extract the feature changes of the patient's right middle lobe over time points. For specific details, refer to Table 1, in which some representative data are listed.

[0078] Table 1. Feature vectors of some collected right middle lobes

[0079]

[0080] Calculate the comprehensive correlation coefficient between the patient's disease condition and their own historical nuclear medicine images : F = ∑ p = 1 l { C p ∗ [ F max ∗ e − k ∗ ( T now − T p Δ T ) ] + ∑ i = 1 n ( F target ∗ I i ) } , where represents the feature change amount of the th target part, is the number of the th target part, , represents the time point when the current patient's nuclear medicine image is taken, represents the time point when the th target part corresponding to the patient's own historical nuclear medicine image is taken, is the set maximum time correlation factor, represents the preset time decay factor, represents the preset reference shooting interval duration, is the natural constant, represents the basic part correlation factor, represents the influence factor of the th corresponding part of the patient's own th historical nuclear medicine image on the target part, represents the number of the th patient's own historical nuclear medicine image, , when the th historical nuclear medicine image corresponding part is the target part,

[0081] It should be noted that in the comprehensive correlation coefficient between the patient's disease condition and their own historical nuclear medicine images, the time correlation degree and part correlation degree of the target part are two important influencing factors. In the first part, reflects the degree of feature change of the target part from history to the present, reflects the influence of the time factor on the correlation degree, represents the time point when the current patient's nuclear medicine image is taken, represents the The time point of the patient's own historical nuclear medicine image corresponding to a target site, as the time interval increases, the value will decrease, meaning that the correlation between the historical image and the current image will decrease due to the passage of time. In the second part, refers to traversing the patient's own historical nuclear medicine images, is a value set according to the actual situation and is used to measure the influence degree of the historical image part on the target site.

[0082] Suppose , , , the simulation results can be obtained. For specific reference, see Table 2, in which some representative data are listed.

[0083] Table 2. Part of the data related to the target site collected and the calculation results

[0084]

[0085] Calculated from the above simulation data: , that is, the comprehensive correlation coefficient between the patient's disease condition and his own historical nuclear medicine image is 3.06.

[0086] S4. Correlate the extracted characteristic values of the patient's clinical data with the characteristic values of the clinical data of the same type of patient group, and calculate the group deviation degree of the patient's clinical data.

[0087] The specific analysis method of step S4 is as follows: S41. According to the patient's clinical data, determine the inclusion criteria for the same type of patients, obtain the clinical data of the same type of patient group that meet the inclusion criteria from the management database, and extract the characteristics of the clinical data of the same type of patient group; it can ensure that the studied same type of patient group has a high correlation and improve the accuracy of subsequent analysis.

[0088] S42. Quantitatively code the degree of symptom improvement in the patient's clinical data and the clinical data of the same type of patient group, and then calculate the means and standard deviations of the clinical data of the same type of patient group; quantitative coding provides a basis for scientific analysis and can more objectively compare the differences in symptom improvement between different patients.

[0089] It should be noted that the specific operation method for the means and standard deviations of the clinical data of the same type of patient group is as follows: quantitative intervals are set for "significant improvement", "partial improvement", and "no improvement" respectively. For the target patient, according to the above quantitative correspondence relationship, the degree of symptom improvement after each treatment is coded. For the clinical data of the same type of patient group, the degree of symptom improvement is coded according to the unified quantitative correspondence relationship. For the data of the degree of symptom improvement of the same type of patient group after quantitative coding, its mean is calculated, and on the basis of calculating the mean, the standard deviation is further calculated. The means and standard deviations of the clinical data of the same type of patient group are calculated according to this method.

[0090] S43. Calculate the difference between each eigenvalue of the patient's clinical data and the mean of the corresponding clinical data of the same type of patient group, and then divide it by the standard deviations of the mutation site occurrence frequency, BMI, body fat percentage, disease attack frequency, and symptom improvement degree of the same type of patient group to confirm the relative position of each eigenvalue of the patient's clinical data among the eigenvalues of the clinical data of the same type of patient group; it can convert each eigenvalue of the patient to a relative scale with the same type of patient group as the reference, which helps to identify the differences between the patient and the same type of patient group and provides a reference for personalized diagnosis and treatment.

[0091] It should be noted that the specific calculation method for the difference between each eigenvalue of the patient's clinical data and the mean of the corresponding clinical data of the same type of patient group is as follows: obtain each eigenvalue of the patient's clinical data and each mean of the corresponding clinical data of the same type of patient group, correspond the clinical data to which each eigenvalue of the patient belongs one by one to obtain the corresponding clinical data of the same type of patient group, and obtain the difference between each eigenvalue of the patient's clinical data and the mean of the corresponding clinical data of the same type of patient group by taking the difference.

[0092] S44. Calculate the population deviation degree of the patient's clinical data according to the relative position of each eigenvalue of the patient's clinical data among the eigenvalues of the clinical data of the same type of patient group. : , where respectively represent the relative positions of the occurrence frequency of the patient's specific mutation site, BMI, body fat percentage, disease attack frequency, and symptom improvement degree among the corresponding eigenvalues of the clinical data of the same type of patient group. respectively represent the weight factors corresponding to the occurrence frequency of the specific mutation site, BMI, body fat percentage, disease attack frequency, and symptom improvement degree; it can more accurately reflect the special situation of the patient, helps to better locate the disease characteristics of the patient in the group background, and provides an important basis for subsequent prediction of disease progression, etc.

[0093] It should be noted that can be set to 0.3. It can be set to 0.2, It can be set to 0.2, It can be set to 0.15, It can be set to 0.15. Specific mutation sites are often closely related to the pathogenesis of diseases. For example, in some genetic diseases, specific gene mutations are the root causes of the diseases. The change in their occurrence frequency may directly reflect the genetic tendency, disease risk, and potential pathophysiological processes of the diseases; BMI and body fat percentage are important indicators for measuring the overall health status and metabolic level of patients. They are closely related to various chronic diseases, such as cardiovascular diseases, diabetes, etc. In patients with the same type of diseases, the abnormalities of BMI and body fat percentage may affect the progression and treatment response of the diseases. Although the disease attack frequency and symptom improvement degree are also important aspects for evaluating the patient's condition, they are relatively more susceptible to various factors, including treatment regimens, patient compliance, lifestyle changes, etc. Compared with specific mutation sites, BMI, and body fat percentage, their intrinsic connection with the disease is not as close. The symptom improvement degree may be somewhat subjective, and different patients may have differences in the perception and description of symptoms. At the same time, the disease attack frequency and symptom improvement degree are often the phased manifestations during the disease development process, with a certain lag, and cannot directly reflect the potential characteristics of the disease like specific mutation sites. Therefore, the weight corresponding to the occurrence frequency of the patient's specific mutation site is the largest, the weights corresponding to BMI and body fat percentage are the second, and the weights corresponding to the disease attack frequency and symptom improvement degree are the smallest.

[0094] It should be noted that there are five influencing factors in the population deviation degree of the patient's clinical data, namely the occurrence frequency of specific mutation sites, BMI, body fat percentage, disease attack frequency, and symptom improvement degree. Considering the roles of different clinical characteristics in evaluating the degree of deviation of patients from the population, the deviation information in multiple dimensions is integrated through weighted summation. Finally, taking the absolute value is because what is concerned is the degree of deviation, rather than the direction of deviation (whether it is higher or lower than the population mean). In this way, the obtained value can comprehensively reflect the degree of deviation of the patient from the same type of patient population under the combined action of multiple clinical characteristics. For specific reference, see Table 3, in which some representative data are listed.

[0095] Table 3. Characteristic values corresponding to some collected clinical data

[0096]

[0097] Through the above simulation calculations, the population deviation of Patient 4 is 1.5, which is the highest among all patients, indicating that this patient has the greatest difference from the population in terms of comprehensive clinical characteristics. The population deviation of Patient 1 is the lowest at 0.241, with a relatively small difference from the population, suggesting that the condition or physical status of Patient 4 is more special in this group of patients, and more personalized considerations may be needed when formulating treatment plans or judging the development of the condition. Comparing Patient 2 and Patient 3, although their population deviations are both relatively high, the values of the occurrence frequency of specific mutation sites, BMI, and body fat percentage of Patient 2 contribute more to the deviation, while for Patient 3, it is mainly the values of the occurrence frequency of specific mutation sites and the disease attack frequency that have a significant impact on the deviation, indicating that the main driving characteristics for different patients to deviate from the population are different.

[0098] S5. Construct a model based on the comprehensive correlation coefficient between the patient's disease condition and their own historical nuclear medicine images and the population deviation of the patient's clinical data to predict the disease progression category of the patient.

[0099] The specific operation method of step S5 is as follows: S51. Analyze each group of similar patients based on the analysis method of the comprehensive correlation coefficient between the patient's disease condition and their own historical nuclear medicine images and the population deviation of the patient's clinical data, and divide the patients into "low-risk", "medium-risk", and "high-risk" disease progression categories according to the set classification criteria, thereby constructing a sample set for each disease progression category; this helps to make a preliminary and intuitive judgment on the severity and development trend of the patient's condition.

[0100] S52. Using each feature and its feature value in the clinical data of each disease progression category sample set as split points, calculate the Gini index value of each feature in each disease progression category sample set, select the feature with the smallest Gini index and its feature value, record this feature as the optimal split feature and this feature value as the optimal split point, divide each disease progression category sample set according to the optimal split point, and then recursively repeat the calculation of the Gini index, selection of the optimal split feature and split point for the obtained subsets until the sample size of the subset is less than the set threshold, thereby constructing a random forest model composed of each decision tree; the smallest Gini index means that when dividing the sample set according to this feature and its feature value, it can maximize the reduction of the impurity of the sample set, making the divided subsets more "pure" in terms of category, which helps to conduct in-depth research on diseases at different progression stages, such as exploring the differences in pathological mechanisms, treatment responses, etc. among diseases in different risk categories.

[0101] It should be noted that the specific operation method for calculating the Gini index value of each feature in each disease progression category sample set is as follows: For each disease progression category sample set, calculate its overall Gini index. The calculation formula for the Gini index is , where represents the category, , is the proportion of the -th class of samples in the sample set. The sample set is divided based on the splitting point, and the Gini index of each sub-sample set after division is calculated respectively, and weighted summation is performed according to the size of the sub-sample set to obtain the feature Gini index based on this splitting point.

[0102] S53. Input the nuclear medicine image data and clinical data of the selected patient into each decision tree in the random forest in sequence, and output the prediction of the disease progression category of the selected patient; different decision trees may have different preliminary judgments on the same patient, but by integrating the results of each decision tree, the error prediction caused by data noise or local features of individual decision trees can be reduced.

[0103] Example 2

[0104] Referring to Figure 2 shown, the present invention provides an analysis system for nuclear medicine image radiology reports, including the following modules: a medical data acquisition module for acquiring nuclear medicine image data and clinical data of a patient.

[0105] A feature extraction module for extracting feature values respectively according to the acquired nuclear medicine image data and clinical data of the patient.

[0106] A historical data association module for associating the feature values of the nuclear medicine image data of the patient extracted with the feature values of the patient's own historical nuclear medicine image data, and evaluating the comprehensive correlation coefficient of the patient's disease condition with his own historical nuclear medicine image by assigning an association factor.

[0107] A patient data association module for associating the feature values of the clinical data of the patient extracted with the feature values of the clinical data of the same type of patient group, and calculating the group deviation degree of the patient's clinical data.

[0108] A model construction module for constructing a model based on the comprehensive correlation coefficient of the patient's disease condition with his own historical nuclear medicine image and the group deviation degree of the patient's clinical data, and predicting the disease progression category of the patient.

[0109] A management database for storing historical nuclear medicine image data of each part.

[0110] The management database is connected to the feature extraction module, the historical data association module, the patient data association module, and the model construction module. The model construction module is connected to the historical data association module and the patient data association module. The feature extraction module is connected to the medical data acquisition module, the historical data association module, and the patient data association module.

[0111] Example 3

[0112] Please refer to Figure 3 As shown, the present invention provides an analysis device for nuclear medicine imaging radiation reports. The device includes a processor, a memory, and a communication bus. A computer-readable program executable by the processor is stored on the memory.

[0113] The communication bus realizes the connection and communication between the processor and the memory.

[0114] When the processor executes the computer-readable program, it realizes the steps in an analysis method for nuclear medicine imaging radiation reports as described in the present invention.

[0115] Embodiment 4

[0116] The present invention provides an analysis medium for nuclear medicine imaging radiation reports. The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to realize the steps in an analysis method for nuclear medicine imaging radiation reports as described in the present invention.

[0117] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limitations of the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention, and still be covered by the protection scope of the present invention.

Claims

1. A method for analyzing nuclear medicine imaging radiology reports, characterized in that: The steps include: S1. Obtain the patient's nuclear medicine imaging data and clinical data; S2. Extracting feature values ​​from the patient's nuclear medicine imaging data and clinical data respectively to obtain feature values ​​from the patient's nuclear medicine imaging data and clinical data; The specific analysis method of step S2 includes: S21. Obtain historical nuclear medicine image data of each part from the management database, input it into the deep learning model for training, use the trained model to analyze the patient's nuclear medicine images, identify each part in the patient's nuclear medicine images, perform model segmentation, and extract feature representations corresponding to each part to form feature vectors of each part, which are regarded as feature values ​​of the patient's nuclear medicine image data; S3. Associate the extracted nuclear medicine imaging data feature values ​​of the patient with the patient's own historical nuclear medicine imaging data feature values, and evaluate the comprehensive correlation coefficient between the patient's symptoms and his own historical nuclear medicine imaging by assigning correlation factors ; The specific operation method of allocating the correlation factor is: A1. Obtain the patient's own historical nuclear medicine images, mark the corresponding shooting time points for each of them, sort the patient's own historical nuclear medicine images according to the time points according to the clinical demand time period, assign the maximum time correlation factor to the historical nuclear medicine image with the closest time point, and use the linear attenuation function to calculate the time correlation factors of the patient's other historical nuclear medicine images; A2. Determine the target site for the current disease diagnosis, set the basic site association factor, and adjust the site association factor according to the degree of influence of other sites that affect the target site; Comprehensive correlation coefficient between the patient's symptoms and his / her historical nuclear medicine images The specific evaluation method is: Using the feature vectors of each part, the target parts in the patient's own historical nuclear medicine images are matched to obtain the patient's own historical nuclear medicine images corresponding to the target part, and the characteristic changes of the target part are extracted, recorded as the characteristic changes of each target part, and the comprehensive correlation coefficient between the patient's symptoms and his own historical nuclear medicine images is calculated. , ,in Indicates The characteristic change of the target part, Indicates the number of the target part, , Indicates the time point at which the current patient's nuclear medicine image was taken. Indicates The time point at which the patient's historical nuclear medicine images were taken for each target part. is the maximum time correlation factor set, Indicates the preset time decay factor, Indicates the preset reference shooting interval duration. is a natural constant, represents the basic site association factor, Indicates the preset patient's own The influence factor of the corresponding part of the historical nuclear medicine image on the target part, Indicates that the patient The serial number of your own historical nuclear medicine images, , when When the corresponding part of the historical nuclear medicine image is the target part, ; S4. Associate the extracted clinical data characteristic values ​​of the patient with the clinical data characteristic values ​​of the same type of patient group, and calculate the group deviation of the patient's clinical data ; S5. Based on , Build a model to predict the patient's disease progression category.

2. The method for analyzing nuclear medicine radiology reports according to claim 1, characterized in that: The patient clinical data includes the patient's gene sequencing data, physical data, and medical history text data.

3. The method for analyzing nuclear medicine radiology reports according to claim 2, characterized in that: The specific analysis method of step S2 also includes: S22. Obtain the patient's gene sequencing data, compare it with the reference genome, determine the position of each sequencing fragment in the reference genome, and if the position of a sequencing fragment in the reference genome coincides with a known disease-related gene region, and the sequencing fragment is different from the reference genome, it is determined to be a gene mutation site, and the gene mutation sites associated with a specific disease are screened out, and the frequency of occurrence of the patient's specific mutation site is calculated; S23. Collect the patient's physical data and perform derived feature calculation to obtain the patient's BMI and body fat percentage; S24. Read the patient's medical history text data, structure the unstructured medical history text data, count the number of attacks of a specific disease from the structured medical history data, and calculate the frequency of disease attacks in combination with time information; S25. Extract the degree of symptom improvement after each treatment, quantify it into specific indicators and classify it into significant improvement, partial improvement and no improvement; S26. The frequency of occurrence of the patient’s specific mutation sites, BMI, body fat percentage, disease attack frequency, and degree of symptom improvement are regarded as the patient’s clinical data characteristic values.

4. The method for analyzing nuclear medicine radiology reports according to claim 1, characterized in that: The specific analysis method of step S4 is: S41. Determine the inclusion criteria for patients of the same type based on the patient's clinical data, obtain clinical data of a group of patients of the same type that meet the inclusion criteria from the management database, and perform feature extraction on the clinical data of the group of patients of the same type; S42. Quantify and encode the degree of symptom improvement in the clinical data of the patient and the clinical data of the same type of patient group, and then calculate the means and standard deviations of the clinical data of the same type of patient group; S43. Calculate the difference between each characteristic value of the patient's clinical data and the mean of the clinical data corresponding to the same type of patient group, and then divide it by the standard deviation of the mutation site occurrence frequency, BMI, body fat percentage, disease attack frequency, and symptom improvement degree of the same type of patient group to confirm the relative position of each characteristic value of the patient's clinical data among the characteristic values ​​of the clinical data of the same type of patient group; S44. Calculate the group deviation of the patient's clinical data based on the relative position of each characteristic value of the patient's clinical data in the clinical data of the same type of patient group : ,in They respectively represent the relative positions of the occurrence frequency of a patient's specific mutation site, BMI, body fat percentage, disease attack frequency, and symptom improvement degree in the corresponding characteristic values ​​of the clinical data of the same type of patient group. They respectively represent the weight factors corresponding to the frequency of occurrence of specific mutation sites, BMI, body fat percentage, disease onset frequency, and degree of symptom improvement.

5. The method for analyzing nuclear medicine radiology reports according to claim 1, characterized in that: The specific operation method of step S5 is: S51. Analyze the patients of the same type based on the comprehensive correlation coefficient between the patient's symptoms and their own historical nuclear medicine images and the group deviation analysis method of the patient's clinical data, and divide the patients into "low risk", "medium risk" and "high risk" disease progression categories according to the set classification criteria, thereby constructing a sample set for each disease progression category; S52. Taking each feature and feature value of the clinical data in each disease progression category sample set as the splitting point, calculate the Gini index value of each feature in each disease progression category sample set, select the feature and feature value with the smallest Gini index, record the feature as the optimal splitting feature, and record the feature value as the optimal splitting point, divide each disease progression category sample set according to the optimal splitting point, and recursively repeat the calculation of the Gini index, the selection of the optimal splitting feature and the splitting point for the obtained subsets, until the number of samples in the subset is less than the set threshold, thereby constructing a random forest model composed of each decision tree; S53. Input the nuclear medicine imaging data and clinical data of the selected patient into each decision tree in the random forest in turn, and output the disease progression category prediction of the selected patient.

6. A system for analyzing nuclear medicine imaging radiology reports, used to perform the steps of the method according to any one of claims 1 to 5, characterized in that: The system specifically includes the following modules: Medical data acquisition module, used to obtain nuclear medicine imaging data and clinical data of patients; A feature extraction module is used to extract feature values ​​from the patient's nuclear medicine imaging data and clinical data, respectively, to obtain the patient's nuclear medicine imaging data feature values ​​and clinical data feature values; A historical data association module is used to associate the extracted nuclear medicine imaging data characteristic values ​​of the patient with the characteristic values ​​of the patient's own historical nuclear medicine imaging data, and to evaluate the comprehensive correlation coefficient between the patient's symptoms and his own historical nuclear medicine imaging by assigning correlation factors; A patient data association module is used to associate the extracted clinical data characteristic values ​​of the patient with the clinical data characteristic values ​​of the same type of patient group, and calculate the group deviation of the patient's clinical data; A model building module is used to build a model based on the comprehensive correlation coefficient between the patient's symptoms and his or her own historical nuclear medicine images and the group deviation of the patient's clinical data to predict the patient's disease progression category; Management database, used to store historical nuclear medicine imaging data of various parts.

7. An analysis device for nuclear medicine imaging radiology reports, the device comprising a processor, a memory and a communication bus, the memory storing a computer readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, the processor implements the steps of the method for analyzing nuclear medicine imaging radiology reports as described in any one of claims 1 to 5.

8. An analytical medium for nuclear medicine imaging radiological reports, characterized in that: The medium is burned with a computer program, and the computer program implements the method described in any one of claims 1 to 5 when running in the memory of the server.

Images