Neurology radiation dose risk analysis prompting method and related equipment

Abstract

The invention provides a prompting method for neurology radiation dose risk analysis and related equipment, and is applied to the technical field of data processing. The method comprises the following steps: acquiring a single-energy-level CT image in a preset time period and an initial radiation dose detection model matched with the single-energy-level CT image; obtaining a training sample set; processing the training sample set based on a preset processing rule to generate a training set and a test set; training the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model; performing slicing processing on the single-energy-level CT image based on the target radiation dose detection model to generate a plurality of material parameter values corresponding to single-energy-level CT slice images; processing the material parameter values based on a target radiation dose detection model, and generating material component information corresponding to a plurality of single-energy-level CT slice images; and processing the material component information based on the target radiation dose detection model, and generating a CT radiation dose corresponding to the single-energy-level CT slice image.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a prompting method for analyzing the radiation dose risk in neurology and related devices. Background Art

[0002] CT imaging technology is a method of reconstructing the internal structure of an object by measuring the attenuation signals of the incident X-rays by the scanned object at multiple angles. Due to its operational convenience and detection effectiveness, computed tomography (CT) technology is becoming more and more common in clinical practice. At the same time, the X-ray radiation brought by CT scans has gradually attracted much attention. Excessive radiation doses can induce cancer. However, for many clinical conditions, multiple CT scans are required for regular observation of lesions. Therefore, reducing the radiation dose received by patients during each CT scan has important clinical significance.

[0003] In the prior art, one type of method is to reduce the X-ray intensity to reduce the radiation dose, but the problem is that the noise signal of the CT image is relatively enhanced and the image quality deteriorates. Considering that many lesions are small in size and rich in details, severe noise signals will greatly increase the difficulty of lesion detection and increase the risks of missed detection and misdiagnosis; another type of method is to reduce the radiation dose by reducing the X-ray projection beam, such as reducing the number of projection angles. How to accurately give a risk warning of the radiation dose so as to reduce the further aggravation of the patient's condition is a technical problem that needs to be urgently solved by those skilled in the art.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present application is to provide a prompting method for analyzing the radiation dose risk in neurology and related devices, which can at least overcome the problems existing in the prior art to a certain extent. By obtaining the change degree of the material component information corresponding to the monoenergetic CT image to judge whether an organ has lesions or other conditions, and then calculating the corresponding radiation dose situation, accurately giving a risk warning of the radiation dose so as to reduce the further aggravation of the patient's condition.

[0006] Other features and advantages of the present application will become apparent through the following detailed description, or will be partially learned through the practice of the present invention.

[0007] According to one aspect of the present application, a method for prompting risk analysis of radiation dose in neurology is provided, including: obtaining at least one single-energy CT image within a preset time period and an initial radiation dose detection model matching the single-energy CT image, wherein the generation times of different single-energy CT images are different; obtaining a training sample set for training the initial radiation dose detection model; processing the training sample set based on a preset processing rule to generate a training set and a test set; training the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model; performing slicing processing on the single-energy CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a plurality of single-energy CT slice images; processing the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to a plurality of single-energy CT slice images; processing the substance component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image.

[0008] In an embodiment of the present application, the processing the training sample set based on a preset processing rule to generate a training set and a test set includes: extracting features from the training sample set to determine an original feature library; dividing each feature data set according to the original feature library to generate a training set and a test set; using a classifier to predict each test set divided from the original feature library to determine a prediction result; using a preset algorithm to train each training set divided from the original feature library to obtain a test set class prediction result.

[0009] In an embodiment of the present application, the extracting features from the training sample set to determine an original feature library includes: processing standard substance information based on a preset processing rule to generate standardized features, wherein the standardized features are features of samples or phantoms with clear and known substance compositions; processing the standardized features based on preset feature screening and dimensionality reduction rules to generate original features; generating an original feature library from a plurality of original features.

[0010] In an embodiment of the present application, the processing the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to a plurality of single-energy CT slice images includes: processing the substance parameter values based on the target radiation dose detection model to generate a plurality of CT image pixel values; generating a plurality of image pixel regions based on different CT image pixel values, wherein the difference between the pixel values of the same image pixel region is within a preset threshold; processing the plurality of image pixel regions respectively based on the target radiation dose detection model to generate substance component information corresponding to the plurality of image pixel regions.

[0011] In one embodiment of the present application, the single-energy CT image is sliced based on the target radiation dose detection model to generate substance parameter values corresponding to a plurality of single-energy CT slice images, including: the target radiation dose detection model includes a calculation formula for obtaining substance parameter values, and the calculation formula is: CTHU(keV) = a·e b·keV + c; where a, b, and c represent characteristic parameters, e represents an exponential function, and keV represents the energy level corresponding to the single-energy CT image.

[0012] The substance parameter values are respectively processed based on the target radiation dose detection model to generate substance composition information corresponding to a plurality of single-energy CT slice images, including: the target radiation dose detection model includes a calculation formula for obtaining substance composition information, and the calculation formula is:

[0013] where p x,y,z represents the substance corresponding to the pixel at the coordinate position x, y, z, w i represents the content of the i-th reference substance, Mi represents the i-th reference substance, and n represents the number of reference substances constituting the substance composition of the pixel.

[0014] In one embodiment of the present application, after generating the CT radiation dose corresponding to the single-energy CT slice image by processing the substance composition information based on the target radiation dose detection model, it further includes: obtaining the lesion area corresponding to the target single-energy CT slice image based on the CT radiation dose; processing the lesion area based on a preset radiation dose scanning method to generate a CT scanning plan; obtaining the estimated radiation dose value generated by the CT scanning plan; obtaining the historical radiation dose value of the lesion area; generating a target radiation dose value based on the historical radiation dose value and the estimated radiation dose value; if the target radiation dose value is greater than a preset threshold, generating a warning message, where the warning message is used to indicate that the current radiation dose exceeds the safety threshold.

[0015] Another aspect of the present application is a prompting device for analyzing the radiation dose risk in neurology, which is characterized by including: an acquisition module, configured to acquire at least one single-energy CT image within a preset time period and an initial radiation dose detection model matching the single-energy CT image, wherein the generation times of different single-energy CT images are different; and acquire a training sample set for training the initial radiation dose detection model; a processing module, configured to process the training sample set based on a preset processing rule to generate a training set and a test set; train the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model; perform slicing processing on the single-energy CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a plurality of single-energy CT slice images; process the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to the plurality of single-energy CT slice images; and process the substance component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image.

[0016] According to still another aspect of the present application, an electronic device is provided, which is characterized by including: a first processor; and a memory for storing executable instructions of the first processor; wherein the first processor is configured to execute the above-mentioned prompting method for analyzing the radiation dose risk in neurology by executing the executable instructions.

[0017] According to yet another aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a second processor, the above-mentioned prompting method for analyzing the radiation dose risk in neurology is implemented.

[0018] According to yet another aspect of the present application, a computer program product is provided, including a computer program, which is characterized in that when the computer program is executed by a third processor, the above-mentioned prompting method for analyzing the radiation dose risk in neurology is implemented.

[0019] A prompting method and related equipment for radiation dose risk analysis in neurology provided by this application. The server obtains at least one single-energy CT image within a preset time period and an initial radiation dose detection model matching the single-energy CT image, where the generation times of different single-energy CT images are different; obtains a training sample set for training the initial radiation dose detection model; processes the training sample set based on a preset processing rule to generate a training set and a test set; trains the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model; performs slicing processing on the single-energy CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a plurality of single-energy CT slice images; processes the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to the plurality of single-energy CT slice images; processes the substance component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image. By obtaining the change degree of the substance component information corresponding to the single-energy CT image, it is possible to judge whether an organ has lesions or other conditions, and then calculate the corresponding radiation dose situation, accurately give a risk warning for the radiation dose, so as to reduce the further aggravation of the patient's condition.

[0020] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. Brief Description of the Drawings

[0021] Figure 1 The flowchart showing a prompting method for radiation dose risk analysis in neurology provided by an embodiment of this application;

[0022] Figure 2 The structural schematic diagram showing a prompting device for radiation dose risk analysis in neurology provided by an embodiment of this application;

[0023] Figure 3 The structural schematic diagram showing an electronic device provided by an embodiment of this application;

[0024] Figure 4 The schematic diagram showing a storage medium provided by an embodiment of this application. Detailed Description of the Embodiment

[0025] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0026] The following combination of Figure 1To describe a prompting method for radiation dose risk analysis in neurology according to an exemplary embodiment of the present application. It should be noted that the following application scenarios are only shown for the convenience of understanding the spirit and principle of the present application, and the embodiments of the present application are not limited in this regard. On the contrary, the embodiments of the present application can be applied to any applicable scenario.

[0027] In one embodiment, the present application also proposes a prompting method for radiation dose risk analysis in neurology and related devices. Figure 1 Schematically shows a flowchart of a prompting method for radiation dose risk analysis in neurology according to an embodiment of the present application. As Figure 1 shown, this method is applied to a server and includes:

[0028] S101, obtaining at least one monoenergetic CT image within a preset time period and an initial radiation dose detection model matching the monoenergetic CT image.

[0029] In one embodiment, by collecting twice successively, using Siemens dual-source, GE high-low energy rapid alternating switching, and Philips double-layer detectors, a series of high- and low-energy CT data can be collected to form a series of monoenergetic CT images. Among them, the monoenergetic CT images include, but are not limited to, high monoenergetic CT images and low monoenergetic CT images, respectively simulating CT images at different energies, and the energy span usually ranges from 50 keV to 140 keV. The generation times of different monoenergetic CT images are different. By obtaining monoenergetic CT images of the same patient at the same location but with a certain time interval, the radiation dose value of the patient within the preset time can be obtained.

[0030] S102, obtaining a training sample set for training the initial radiation dose detection model.

[0031] In one implementation, the training sample set includes a number of reference substance information; the training sample set is preprocessed to generate a training sample set with identification information. The data of the training sample set includes patient data from different hospitals. Using the random number table method, it is divided into a training group (281 cases) and an internal test group (122 cases) according to a ratio of 7:3. Among them, the identification information is used to characterize the attributes of various characterization substance information. Based on the training sample set with identification information, the initial radiation dose detection model is trained to generate a target radiation dose detection model. Specifically, obtain the data classification result corresponding to the training sample set. Since the classification result of each training sample set can be determined in advance and can be directly obtained from the outside. Compare the prediction result with the data classification result to determine the first comparison result; compare the prediction result of the test set class with the data classification result to determine the second comparison result; when judging whether the second comparison result and the first comparison result meet the preset requirements, when they meet the preset requirements, it indicates that the detection result of the current radiation dose detection model is relatively accurate. At this time, the current radiation dose detection model can be used as the target radiation dose detection model. The radiation dose detection model includes a reference parameter space for characterizing the distribution position of the characteristic parameters corresponding to the substance components.

[0032] S103. Process the training sample set based on a preset processing rule to generate a training set and a test set.

[0033] In one implementation, feature extraction is performed on the training sample set to determine the original feature library; each feature data set is divided according to the original feature library to generate a training set and a test set; a classifier is used to predict each test set divided by the original feature library to determine the prediction result; a preset algorithm is used to train each training set divided by the original feature library to obtain the prediction result of the test set class.

[0034] In another implementation, the standard substance information is processed based on a preset processing rule to generate standardized features, where the standardized features are the features of a sample or phantom with a clear and known substance composition; the standardized features are processed based on a preset feature screening and dimensionality reduction rule to generate original features; a number of original features are used to generate an original feature library. Feature extraction is performed on the training data set to determine the original feature library, where the feature extraction includes four types of features: original features, statistical features, frequency domain features, and time domain features. The specific introduction of the commonly extracted features is as follows: Original features: The original features are the data obtained after collection and preprocessing. All the information in each sample data matrix is used for model training. To avoid losing sample information, each curve of the sample Xi is directly expanded and stretched into a row vector. Statistical features: The statistical features consider the change trend of all data points on the curve. By extracting statistical features, the dimension of the data sample can be reduced, facilitating the analysis of gene data and accelerating the convergence speed during model training. The extracted statistical features include maximum value, minimum value, mean value, variance, and standard deviation. Frequency domain features: The data set is subjected to wavelet transform, and the coefficients obtained from the second-order wavelet transform are used as new features to form frequency domain features. Time domain features: The time domain features mainly start from the perspective of time to discover the change rules of signals and systems. The time domain features can reflect the information in the curve data during time changes. This process mainly extracts the first-order forward difference of the data set to form the first-order difference time domain features, and an exponential moving average feature processing method is adopted.

[0035] S104. Train the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model.

[0036] In one implementation, multiple data groups are extracted from the training set, where each data group contains a preset number of data samples, and at least one data sample includes identification information, and the data corresponding to the identification information can be CT images with different radiation dose types. The initial radiation dose detection model is trained based on the data samples in the multiple data groups to generate a trained radiation dose detection model.

[0037] In addition, since the data for training the radiation dose detection model may be scarce and the proportion of each stage in the training data is uneven, the Stable Diffusion model is used to generate single-energy CT images. First, based on the real images as the target distribution, that is, the stable distribution, according to the principle of the stable diffusion model, a stable diffusion model suitable for generating single-energy CT images is constructed. This model consists of multiple deconvolution layers and other neural network layers, and reversely generates single-energy CT images through the reverse diffusion process. During the training process, starting from the stable distribution (i.e., the real image distribution), the stable diffusion model is used for the reverse diffusion process. The reverse diffusion process will gradually blur the image until the virtual single-energy CT image corresponding to the stable distribution is obtained. The generated virtual single-energy CT image has corresponding characteristic information, which is thus used for subsequent lesion judgment and substance component identification.

[0038] During the reverse diffusion process, noise is introduced to control the diversity and randomness of the generated single-energy CT images. By adjusting the intensity of the noise, the clarity and style of the generated images can be affected, so as to obtain more diverse virtual single-energy CT images. In addition, the initial noise is set. Specifically, this initial noise can be random Gaussian noise or uniformly distributed noise, aiming to introduce some randomness in the early stage of image generation to encourage the model to generate diverse initial images. Then, the noise intensity is adjusted and gradually decreased during the generation process, which can be achieved by controlling the noise parameters in the generation network or gradually reducing the standard deviation of the added noise during the training process. Gradually reducing the noise intensity helps the model to gradually improve the clarity and quality of the image during the generation process. By introducing the initial noise and gradually reducing the noise intensity, the model can generate more random and diverse images, which helps to generate a series of single-energy CT images with different characteristics or cases, and helps the stable diffusion model to more comprehensively learn the distribution of the data. The trained model is more likely to handle the input of noise or incomplete information, thus improving the robustness of the model and making it more applicable to various uncertain factor situations in the real world.

[0039] Based on the test set, the trained radiation dose detection model is processed to generate test results; if the data samples containing identification information in the test results indicate that the radiation dose in the current area is in an abnormal state, then the trained radiation dose detection model is used as the target radiation dose detection model.

[0040] S105, based on the target radiation dose detection model, slice the single-energy CT image to generate the substance parameter values corresponding to a number of single-energy CT slice images.

[0041] In one implementation, the target radiation dose detection model includes a calculation formula for obtaining the substance parameter values. Specifically, the specific calculation formula is as follows:

[0042] CTHU(keV) = a·e b·keV + c; where a, b, and c represent characteristic parameters, e represents the exponential function, and keV represents the energy level corresponding to the single - energy CT image.

[0043] Specifically, CTHU(keV) represents the tissue parameter value at a specific energy level (keV), such as the Hounsfield unit (HU) in a CT image, which is a unit for measuring the absorption degree of radiation in different tissues.

[0044] a, b, and c are the characteristic parameters of the model, which are determined during the model fitting process and are used to describe the characteristics of different tissues' absorption of X - rays.

[0045] e is the base of the natural logarithm, approximately equal to 2.71828.

[0046] (keV) is the energy level corresponding to the single - energy CT image, representing the photon energy of the X - ray.

[0047] The CT image is segmented into different regions or slices, and each region corresponds to a specific tissue or structure. For each segmented image region, the above formula is used to estimate the parameters a, b, and c. This uses curve - fitting techniques, such as non - linear least squares method, to find the best - fitting parameters.

[0048] When the parameters a, b, and c are determined, the CTHU value of each pixel or each image region at a specific energy level can be calculated using the above formula. Based on the calculated CTHU values, a CT image reflecting the characteristics of different tissues or structures is reconstructed.

[0049] Thus, the diagnostic value of the CT image is improved, especially when it is necessary to distinguish different types of tissues or detect micro - lesions. By analyzing the CT images at different energy levels, doctors can obtain more information about tissue characteristics and thus make more accurate diagnoses.

[0050] S106. Process the substance parameter values respectively based on the target radiation dose detection model to generate the substance component information corresponding to a number of single - energy CT slice images.

[0051] In one implementation, the target radiation dose detection model includes a calculation formula for obtaining substance component information, and the calculation formula is:

[0052]

[0053] where p x,y,z represents the substance component information corresponding to the pixel at the coordinate position x, y, z, and w i represents the content of the i - th reference substance, and all wi The sum is 1, satisfying the normalization condition. Mi represents the specific property or parameter value of the i-th reference substance, which can be density, Hounsfield unit (HU), or other relevant physical or chemical properties. n represents the number of reference substances that make up the material components of the pixel.

[0054] Specifically, a set of reference substances Mi is determined, which can represent different types of tissues or materials that may be encountered in the CT image. The content w of each reference substance is determined for each pixel point. i The sum of these contents must be 1.

[0055] Using the above formula, multiply the content of the reference substance of each pixel point by its corresponding reference substance property, and then sum to obtain the material component information of the pixel. According to the calculated material component information of each pixel, the CT image is reconstructed to reflect the distribution of different material components. Using the reconstructed image, doctors or researchers can perform further analysis to identify and diagnose different tissue types or lesions.

[0056] In another implementation, the material parameter values are processed based on the target radiation dose detection model to generate several CT image pixel values; several image pixel regions are generated based on different CT image pixel values, where the difference in pixel values of the same image pixel region is within a preset threshold; the target radiation dose detection model is used to process several image pixel regions respectively to generate the material component information corresponding to several image pixel regions. The difference in pixel values of the same image pixel region is within a preset threshold, such as the CT scan part (head, neck, chest, abdomen, limbs), organ type (brain, lung, intestine, bone, muscle, etc.), whether there is a contrast agent or an implant in the body, to improve the accuracy of material component discrimination. This solution does not limit the CT image pixel values of each region, and can be set according to the actual needs of the applicant. For example, if the tissue region is the lung, the candidate tissues within the region are lung, muscle, blood, and fat; if the tissue region is bone, the candidate tissues within the region are bone, fat, and blood.

[0057] S107, process the material component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image.

[0058] In one implementation, the classified tissues and organs are corresponded with the publicly available database data to obtain the types and densities of elements contained in each organ, and are input into the target radiation dose detection model together with the CT scan system parameters and scan process parameters.

[0059] Obtain the current material component information of the single-energy CT slice image, determine whether the corresponding cells, tissues, and body fluids are the same as the organs in this part conventionally, and obtain the cumulative CT radiation dose value of this part based on the change of the material component information, so as to judge whether it is suitable to continue using the same examination method or use other methods to replace it in the future. In addition, in order to avoid misjudgment, the single-energy CT slice image of the same part before the preset time period will also be obtained for the current single-energy CT slice image, so as to judge whether the change of the organ in this part has a positive correlation with the radiation dose. This application does not limit the radiation dose generated by each CT detection of each organ, which can be limited according to the actual needs of the applicant.

[0060] In this application, the server obtains at least one single-energy CT image within a preset time period and an initial radiation dose detection model matching the single-energy CT image, where the generation times of different single-energy CT images are different; obtains a training sample set for training the initial radiation dose detection model; processes the training sample set based on a preset processing rule to generate a training set and a test set; trains the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model; performs slicing processing on the single-energy CT image based on the target radiation dose detection model to generate material parameter values corresponding to a plurality of single-energy CT slice images; processes the material parameter values respectively based on the target radiation dose detection model to generate material component information corresponding to a plurality of single-energy CT slice images; processes the material component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image. By obtaining the degree of change of the material component information corresponding to the single-energy CT image, it is judged whether the organ has lesions or other conditions, and then the corresponding radiation dose situation is calculated, and the risk of the radiation dose is accurately warned to reduce the further aggravation of the patient's condition.

[0061] Optionally, in another embodiment based on the above method of this application, after generating the CT radiation dose corresponding to the single-energy CT slice image by processing the material component information based on the target radiation dose detection model, it further includes:

[0062] Obtain the lesion area corresponding to the target single-energy CT slice image based on the CT radiation dose;

[0063] Process the lesion area based on a preset radiation dose scanning method to generate a CT scanning plan;

[0064] Obtain the estimated radiation dose value generated by the CT scanning plan;

[0065] Obtain the historical radiation dose value of the lesion area;

[0066] Generate a target radiation dose value based on the historical radiation dose value and the estimated radiation dose value;

[0067] If the target radiation dose value is greater than a preset threshold, generate a warning message, where the warning message is used to indicate that the current radiation dose exceeds the safety threshold.

[0068] In one implementation, for a target single-energy CT slice image, first delineate the target region related to the lesion, mark the center point of the region, select the detector channels corresponding to the target region, and adjust the CT scanning bed so that the marked center point of the region is at the rotation center position of the CT scan, and scan the region related to the lesion area. In addition, adjust the angular interval of continuous X-ray exposure according to specific clinical requirements. For regular CT scans and observations of the lesion, since the target region is often much smaller than the global scanning region, reconstruct the small-range lesion region and collect the projection data related to the lesion area. On the premise of meeting the actual usage requirements, reduce the radiation dose received by the human body during a single CT scan to a large extent.

[0069] If the warning value is exceeded, it will strongly prompt the doctor to control the radiation dose for this CT scan (for example, show that the cumulative radiation dose received by the patient's lungs is too large), and give recommended CT scan parameters with reference to the CT imaging system model of this CT imaging. The doctor can adjust the parameters of this CT scan for CT imaging appropriately according to the system's suggestions and combined with experience.

[0070] By applying the above technical solutions, the server obtains at least one single-energy CT image within a preset time period and an initial radiation dose detection model matching the single-energy CT image, where the generation times of different single-energy CT images are different; obtains a training sample set for training the initial radiation dose detection model; processes the standard substance information based on a preset processing rule to generate standardized features, where the standardized features are the features of a sample or phantom with a clear and known material composition; processes the standardized features based on a preset feature screening and dimensionality reduction rule to generate original features; generates an original feature library from several original features; divides each feature data set according to the original feature library to generate a training set and a test set; uses a classifier to predict each test set divided by the original feature library to determine the prediction result; uses a preset algorithm to train each training set divided by the original feature library to obtain the prediction result of the test set class; extracts multiple data groups from the training set, where each data group contains a preset number of data samples, and at least one data sample includes identification information; trains the initial radiation dose detection model based on the data samples in the multiple data groups to generate a trained radiation dose detection model.

[0071] Process the trained radiation dose detection model based on the test set to generate test results; if the data samples containing identification information in the test results indicate that the radiation dose in the current area is in an abnormal state, then use the trained radiation dose detection model as the target radiation dose detection model; perform slicing processing on the single-energy CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a number of single-energy CT slice images; process the substance parameter values based on the target radiation dose detection model to generate a number of CT image pixel values; generate a number of image pixel regions based on different CT image pixel values, where the difference in pixel values of the same image pixel region is within a preset threshold; process each of the number of image pixel regions based on the target radiation dose detection model to generate substance composition information corresponding to each image pixel region; process the substance composition information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image.

[0072] Obtain the lesion area corresponding to the target single-energy CT slice image based on the CT radiation dose; process the lesion area based on a preset radiation dose scanning method to generate a CT scanning plan; obtain the estimated radiation dose value generated by the CT scanning plan; obtain the historical radiation dose value of the lesion area; generate a target radiation dose value based on the historical radiation dose value and the estimated radiation dose value; if the target radiation dose value is greater than a preset threshold, generate a warning message, where the warning message is used to indicate that the current radiation dose exceeds the safety threshold. By obtaining the change degree of the substance composition information corresponding to the single-energy CT image, the situation of whether the organ has lesions can be judged, and then the corresponding radiation dose situation can be calculated, and the risk of the radiation dose can be accurately warned to reduce the further aggravation of the patient's condition.

[0073] In one implementation, as Figure 2 shown, the present application further provides a prompt device for risk analysis of radiation dose in neurology, including:

[0074] An acquisition module 201, configured to acquire at least one single-energy CT image within a preset time period and an initial radiation dose detection model matching the single-energy CT image, where the generation times of different single-energy CT images are different; acquire a training sample set for training the initial radiation dose detection model.

[0075] The processing module 202 is configured to process the training sample set based on a preset processing rule to generate a training set and a test set; train the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model; perform slicing processing on the single-energy CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a plurality of single-energy CT slice images; process the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to a plurality of single-energy CT slice images; and process the substance component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image.

[0076] In another embodiment of the present application, the processing module 202 is configured to process the training sample set based on a preset processing rule to generate a training set and a test set, including:

[0077] Extract features from the training sample set to determine an original feature library;

[0078] Divide each feature data set according to the original feature library to generate a training set and a test set;

[0079] Use a classifier to predict each test set divided from the original feature library to determine a prediction result;

[0080] Use a preset algorithm to train each training set divided from the original feature library to obtain a test set class prediction result.

[0081] In another embodiment of the present application, the processing module 202 is configured to extract features from the training sample set to determine an original feature library, including:

[0082] Process the standard substance information based on a preset processing rule to generate standardized features, where the standardized features are features of a sample or phantom with a clear and known substance composition;

[0083] Process the standardized features based on a preset feature screening and dimensionality reduction rule to generate original features;

[0084] Generate an original feature library from a plurality of original features.

[0085] In another embodiment of the present application, the processing module 202 is configured to process the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to a plurality of single-energy CT slice images, including:

[0086] Process the substance parameter values based on the target radiation dose detection model to generate a plurality of CT image pixel values;

[0087] Generate a number of image pixel regions based on different CT image pixel values, wherein the difference in pixel values of the same image pixel region is within a preset threshold;

[0088] Process the number of image pixel regions respectively based on the target radiation dose detection model to generate substance component information corresponding to the number of image pixel regions.

[0089] In another implementation manner of the present application, the processing module 202 is configured to perform slicing processing on the monoenergetic CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a number of monoenergetic CT slice images, including:

[0090] The target radiation dose detection model includes a calculation formula for obtaining substance parameter values, and the calculation formula is:

[0091] CTHU(keV) = a·e b·keV + c; wherein, a, b, and c represent characteristic parameters, e represents an exponential function, and keV represents the energy level corresponding to the monoenergetic CT image.

[0092] In another implementation manner of the present application, the processing module 202 is configured to process the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to a number of monoenergetic CT slice images, including:

[0093] The target radiation dose detection model includes a calculation formula for obtaining substance component information, and the calculation formula is:

[0094]

[0095] wherein, p x,y,z represents the substance component information corresponding to the pixel at the coordinate position x, y, z, w i represents the content of the i-th reference substance, Mi represents the i-th reference substance, and n represents the number of reference substances constituting the substance component of the pixel.

[0096] In another implementation manner of the present application, after the processing module 202 is configured to process the substance component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the monoenergetic CT slice image, it further includes:

[0097] Obtain the lesion region corresponding to the target monoenergetic CT slice image based on the CT radiation dose;

[0098] Process the lesion region based on a preset radiation dose scanning method to generate a CT scanning scheme;

[0099] Obtain the estimated radiation dose value generated by the CT scan protocol;

[0100] Obtain the historical radiation dose value of the lesion area;

[0101] Generate a target radiation dose value based on the historical radiation dose value and the estimated radiation dose value;

[0102] If the target radiation dose value is greater than a preset threshold, generate a warning message, where the warning message is used to indicate that the current radiation dose exceeds the safety threshold.

[0103] An embodiment of the present application provides an electronic device, as Figure 3 shown, the electronic device 3 includes a first processor 300, a memory 301, a bus 302, and a communication interface 303. The first processor 300, the communication interface 303, and the memory 301 are connected through the bus 302; a computer program that can run on the first processor 300 is stored in the memory 301, and when the first processor 300 runs the computer program, it executes the prompt method for analyzing the risk of radiation dose in neurology provided by any of the foregoing embodiments of the present application.

[0104] Among them, the memory 301 may include a high-speed random access memory (RAM: Random Access Memory), and may also include a non-volatile memory, such as at least one disk memory. Through at least one communication interface 303 (which can be wired or wireless), a communication connection is realized between the system network element and at least one other network element, and the Internet, wide area network, local area network, metropolitan area network, etc. can be used.

[0105] The bus 302 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. Among them, the memory 301 is used to store a program. After receiving an execution instruction, the first processor 300 executes the program. The prompt method for analyzing the risk of radiation dose in neurology disclosed in any of the foregoing embodiments of the present application can be applied to the first processor 300 or implemented by the first processor 300.

[0106] The first processor 300 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in the first processor 300 or the instructions in the form of software. The above first processor 300 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be embodied as being executed by a hardware decoding processor or by a combination of hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory 301, and the first processor 300 reads the information in the memory 301 and combines its hardware to complete the steps of the above method.

[0107] The electronic device provided by the above embodiment of the present application and the prompt method for analyzing the radiation dose risk in the department of neurology provided by the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by the application program stored therein.

[0108] The embodiment of the present application provides a computer-readable storage medium, such as Figure 4 shown, the computer-readable storage medium 401 stores a computer program, and when the computer program is read and run by the second processor 402, it implements the prompt method for analyzing the radiation dose risk in the department of neurology as described above.

[0109] The technical solution of the embodiment of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing an electronic device (which may be an air conditioner, a refrigeration device, a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in the embodiment of the present application. The aforementioned storage medium includes: various media such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disc that can store program codes.

[0110] The computer-readable storage medium provided by the above embodiments of the present application and the prompting method for the risk analysis of radiation dose in neurology provided by the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0111] The embodiments of the present application provide a computer program product, including a computer program, and the computer program is executed by a third processor to implement the method as described above.

[0112] The computer program product provided by the above embodiments of the present application and the prompting method for the risk analysis of radiation dose in neurology provided by the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0113] It should be noted that in the present application, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0114] Each embodiment in the present application is described in a related manner. The same or similar parts among the embodiments can be referred to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the embodiments of the prompting method, electronic device, electronic equipment, and readable storage medium for evaluating the risk analysis of radiation dose in neurology, since they are basically similar to the embodiments of the above-mentioned prompting method for the risk analysis of radiation dose in neurology, the description is relatively simple, and the relevant parts can be referred to the partial description of the embodiments of the above-mentioned prompting method for the risk analysis of radiation dose in neurology.

[0115] Although the present application is disclosed as above, the present application is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the protection scope of the present application should be subject to the scope defined by the claims.

Claims

1. A prompting method for radiation dose risk analysis in neurology, characterized in that, Including: Obtaining at least one single - energy CT image within a preset time period and an initial radiation dose detection model matching the single - energy CT image, wherein the generation times of different single - energy CT images are different; Obtaining a training sample set for training the initial radiation dose detection model; Processing the training sample set based on a preset processing rule to generate a training set and a test set; Training the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model; Performing slicing processing on the single - energy CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a plurality of single - energy CT slice images; Processing the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to a plurality of single - energy CT slice images; Processing the substance component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single - energy CT slice image.

2. The method according to claim 1, wherein The processing the training sample set based on a preset processing rule to generate a training set and a test set includes: Performing feature extraction on the training sample set to determine an original feature library; Dividing each feature data set according to the original feature library to generate a training set and a test set; Using a classifier to predict each test set divided from the original feature library to determine a prediction result; Using a preset algorithm to train each training set divided from the original feature library to obtain a test set class prediction result.

3. The method according to claim 2, wherein The performing feature extraction on the training sample set to determine an original feature library includes: Processing standard substance information based on a preset processing rule to generate standardized features, wherein the standardized features are features of samples or phantoms with clear and known substance compositions; Processing the standardized features based on a preset feature screening and dimensionality reduction rule to generate original features; Generating an original feature library from a plurality of original features.

4. The method according to claim 1, characterized in that The processing the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to a plurality of single - energy CT slice images includes: Processing the substance parameter values based on the target radiation dose detection model to generate a plurality of CT image pixel values; Generating a plurality of image pixel regions based on different CT image pixel values, wherein the difference in pixel values of the same image pixel region is within a preset threshold; Processing a plurality of image pixel regions respectively based on the target radiation dose detection model to generate substance component information corresponding to the plurality of image pixel regions.

5. The method according to claim 1, characterized in that, The performing slicing processing on the single - energy CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a plurality of single - energy CT slice images includes: The target radiation dose detection model includes a calculation formula for obtaining substance parameter values, and the calculation formula is: CTHU (keV) = a·e b·keV + c; Wherein, a, b, and c represent characteristic parameters, e represents an exponential function, and keV represents the energy level corresponding to the single - energy CT image.

6. The method according to claim 5, wherein Processing the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to a plurality of single - energy CT slice images includes: The target radiation dose detection model includes a calculation formula for obtaining substance component information, and the calculation formula is: where p x，y，z represents the substance component information corresponding to the pixel at the coordinate positions x, y, and z, w i represents the content of the i-th reference substance, Mi represents the i-th reference substance, and n represents the number of reference substances that make up the substance components of the pixel.

7. The method according to claim 1, characterized in that After processing the substance component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image, it further includes: Obtaining the lesion area corresponding to the target single-energy CT slice image based on the CT radiation dose; Processing the lesion area based on a preset radiation dose scanning method to generate a CT scanning plan; Obtaining the estimated radiation dose value generated by the CT scanning plan; Obtaining the historical radiation dose value of the lesion area; Generating a target radiation dose value based on the historical radiation dose value and the estimated radiation dose value; If the target radiation dose value is greater than a preset threshold, generating a warning message, where the warning message is used to indicate that the current radiation dose exceeds the safety threshold.

8. A prompting device for risk analysis of radiation dose in neurology department, characterized in that, The device includes: An acquisition module, configured to acquire at least one single-energy CT image within a preset time period and an initial radiation dose detection model matching the single-energy CT image, where the generation times of different single-energy CT images are different; and acquire a training sample set for training the initial radiation dose detection model; A processing module, configured to process the training sample set based on a preset processing rule to generate a training set and a test set; train the initial radiation dose detection model based on the training set and the test set to generate a target radiation dose detection model; perform slice processing on the single-energy CT image based on the target radiation dose detection model to generate substance parameter values corresponding to a plurality of single-energy CT slice images; process the substance parameter values respectively based on the target radiation dose detection model to generate substance component information corresponding to the plurality of single-energy CT slice images; and process the substance component information based on the target radiation dose detection model to generate the CT radiation dose corresponding to the single-energy CT slice image.

9. An electronic device, characterized in that, It includes: A first processor; And a memory for storing executable instructions of the first processor; Wherein, the first processor is configured to execute the prompting method for risk analysis of radiation dose in neurology described in any one of claims 1 to 7 by executing the executable instructions.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a second processor, it implements the prompting method for risk analysis of radiation dose in neurology described in any one of claims 1 to 7.

Images