A method and system for predicting thermal capacity of medical imaging equipment

Abstract

An embodiment of the present specification provides a method for predicting the thermal capacity of a medical imaging device, the method comprising obtaining a target first prediction model, which is a machine learning model; obtaining attribute information of a tube in the medical imaging device and operating information of the medical imaging device, wherein the operating information includes the current thermal capacity and scanning parameters of the tube; and inputting the attribute information and the operating information into the target first prediction model to determine a target predicted thermal capacity of the tube at a preset time point.

Description

Technical Field

[0001] The present specification relates to the field of medical imaging equipment, and in particular to a method and system for predicting thermal capacity of medical imaging equipment. Background Art

[0002] Medical imaging equipment has become an indispensable device in the existing medical field. When medical imaging equipment is imaging patients, the components in the medical imaging equipment will affect the quality of the imaging. The tube is the core part of the CT (Computed Tomography) scanning equipment. Reducing the loss of the tube is very important to ensure CT scanning imaging. The heat capacity of the tube limits the scanning capability of CT scanning. Too high heat capacity will cause the loss of the tube.

[0003] Therefore, it is necessary to propose a heat capacity prediction method for medical imaging equipment to predict the heat capacity that may be caused by the operation of the tube, and take measures in advance based on the predicted heat capacity to reduce the loss of the tube. Summary of the invention

[0004] One of the embodiments of the present specification provides a method for predicting the thermal capacity of a medical imaging device, the method comprising: obtaining a target first prediction model, the target first prediction model being a machine learning model; obtaining attribute information of a tube in the medical imaging device and working information of the medical imaging device, the working information comprising the current thermal capacity and scanning parameters of the tube; and inputting the attribute information and the working information into the target first prediction model to determine the target predicted thermal capacity of the tube at a preset time point.

[0005] One of the embodiments of the present specification provides a thermal capacity prediction system for medical imaging equipment, the system comprising: a model acquisition module, used to acquire a target first prediction model, the target first prediction model being a machine learning model; an information acquisition module, used to acquire attribute information of a tube in the medical imaging equipment and working information of the medical imaging equipment, the working information including the current thermal capacity and scanning parameters of the tube; and a prediction module, used to input the target first prediction model based on the attribute information and the working information, to determine the target predicted thermal capacity of the tube at a preset time point.

[0006] One of the embodiments of the present specification provides a device for predicting thermal capacity of a medical imaging device, including a processor, wherein the processor is used to execute operations corresponding to the method for predicting thermal capacity of a medical imaging device as described in the previous item.

[0007] One of the embodiments of this specification provides a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes operations corresponding to the thermal capacity prediction method for medical imaging equipment as described in the previous item.

[0008] Some embodiments of the present specification have the following beneficial effects: (1) The accuracy of heat capacity prediction can be improved by predicting heat capacity through a machine learning model. (2) The attribute information of the tube is taken into account during prediction, so that the prediction can be made according to the actual situation of each tube itself, which is more practical. (3) In actual application, based on the prediction results and the actually detected heat capacity values, enhanced training is performed to update the parameters of the first prediction model, which can improve the prediction accuracy of the first prediction model and enable it to make predictions based on the attribute information of the tube of each device. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0010] Figure 1 is a schematic diagram of an application scenario of a heat capacity prediction system according to some embodiments of this specification;

[0011] Figure 2 is an exemplary module diagram of a heat capacity prediction system according to some embodiments of this specification;

[0012] Figure 3 is an exemplary flow chart of heat capacity prediction according to some embodiments of this specification;

[0013] Figure 4 is an exemplary flow chart of a method for obtaining a target first prediction model according to some embodiments of this specification;

[0014] Figure 5 is another exemplary flow chart of a method for obtaining a target first prediction model according to some embodiments of this specification;

[0015] Figure 6 is an exemplary schematic diagram of a target first prediction model structure according to some embodiments of this specification;

[0016] Figure 7 It is an exemplary schematic diagram of the target first prediction model training process shown in some embodiments of this specification. DETAILED DESCRIPTION

[0017] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0018] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0019] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0020] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0021] Figure 1 It is a schematic diagram of an application scenario of a heat capacity prediction system according to some embodiments of this specification.

[0022] like Figure 1 As shown, the thermal capacity prediction system 100 may include a processing device 110 , a network 120 , a user terminal 130 , a storage device 140 , and a medical imaging device 150 .

[0023] In some embodiments, the thermal capacity prediction system 100 can predict the change in thermal capacity of the tube of a medical imaging device during use by implementing the method and / or process disclosed in this specification.

[0024] The processing device 110 can process data and / or information obtained from the terminal 130, the storage device 140 and / or the medical imaging device 150. The processing device 110 can access information and / or data from the terminal 130, the storage device 140 and / or the medical imaging device 150 through the network 120. The processing device 110 can directly connect to the terminal 130, the storage device 140 and / or the medical imaging device 150 to access information and / or data. For example, the processing device 110 can obtain working information of the medical imaging device 150 and attribute information of the tube in the medical imaging device 150 from the medical imaging device 150 and / or the terminal 130. The processing device 110 can process the acquired data and / or information. For example, the processing device 110 can predict the target predicted thermal capacity of the tube at a preset time point based on the acquired working information and the attribute information of the tube. For another example, the processing device 110 can obtain a target first prediction model through training, and predict the target predicted thermal capacity of the tube at a preset time point through the target first prediction model. In some embodiments, the processing device 110 may be a single server or a server group. The processing device 110 may be disposed in the medical imaging device 150. The processing device 110 may be local or remote. The processing device 110 may be implemented on a cloud platform.

[0025] The network 120 may include any suitable network that provides information and / or data exchange that can facilitate the thermal capacity prediction system 100. In some embodiments, one or more components of the thermal capacity prediction system 100 (e.g., the processing device 110, the user terminal 130, the storage device 140, and the medical imaging device 150) may exchange information and / or data through the network 120. The network 120 may include a local area network (LAN), a wide area network (WAN), a wired network, a wireless network, etc., or any combination thereof.

[0026] The terminal 130 refers to one or more terminal devices or software used by the user. In some embodiments, the terminal 130 may be a mobile device, a tablet computer, etc. or any combination thereof. In some embodiments, the terminal 130 may interact with other components in the thermal capacity prediction system 100 through the network 120. For example, the terminal 130 may send one or more control instructions to the medical imaging device 150 to control the processing device 110 to process the working information of the medical imaging device 150 and the attribute information of the tube in the medical imaging device 150, thereby obtaining the target predicted thermal capacity of the tube of the medical imaging device. In some embodiments, the terminal 130 may be a part of the processing device 110. In some embodiments, the terminal 130 may be integrated with the processing device 110 as an operating console for the medical imaging device 150.

[0027] The storage device 140 may be used to store data, instructions and / or any other information. In some embodiments, the storage device 140 may store data and / or information obtained from, for example, the server 110, the terminal 130, the medical imaging device 150, etc. For example, the storage device 140 may store a pre-trained target first prediction model, a target second prediction model, etc. For another example, the storage device 140 may store working parameters of the medical imaging device, attribute information of the tube, the real heat capacity of the tube, etc. The storage device 140 may be set in the medical imaging device 150. In some embodiments, the storage device 140 may include a large-capacity memory, a removable memory, etc., or any combination thereof.

[0028] The medical imaging device 150 can be used to obtain image data of the scanned object. The scanned object may include a biological object (e.g., a human body, an animal, etc.), a non-biological object (e.g., a phantom), etc. In some embodiments, the medical imaging device 150 may be a CT imaging device, a PET-CT imaging device, etc. In some embodiments, the medical imaging device 150 may include an X-ray tube (referred to as "tube"), which may emit a radiation beam (e.g., X-rays) to the scanned object to generate image data. The tube may include a tube core and a tube shell, the tube core including a cathode, a focus, an anode target, a rotating shaft bearing supporting the anode target, etc., and the tube shell includes a coolant, a radiator, etc. As described above, the processing device 110 and the storage device 140 may become part of the medical imaging device 150. Accordingly, thermal capacity prediction may be performed by the medical imaging device.

[0029] Figure 2 is an exemplary module diagram of a thermal capacity prediction system according to some embodiments of the present specification.

[0030] In some embodiments, the heat capacity prediction system 200 may include a model acquisition module 210 , an information acquisition module 220 , and a prediction module 230 .

[0031] The model acquisition module 210 can be used to acquire a target first prediction model, which is a machine learning model. In some embodiments, the model acquisition module 210 can also be used to construct an initial second prediction model, acquire first training data, and update the parameters of the initial second prediction model based on the first training data to determine the target second prediction model, and determine the target first prediction model based on the target second prediction model. In some embodiments, the model acquisition module 210 can acquire second training data during actual application, and update the intermediate target first prediction model based on the second training data to obtain the target first prediction model. For more details on acquiring the target first prediction model, see Figure 4 , Figure 5 , Figure 7 and its related description.

[0032] The information acquisition module 220 can be used to obtain the property information of the tube in the medical imaging device and the working information of the medical imaging device. In some embodiments, the information acquisition module 220 can obtain the real heat capacity of the tube at a preset time point based on the sensor. For more details about the property information of the tube, the working information of the medical imaging device, and obtaining the real heat capacity, see Figure 3 and its related description.

[0033] The prediction module 230 can process the attribute information and the working information based on the target first prediction model to determine the target predicted thermal capacity of the tube at a preset time point. The target first prediction model can be a machine learning model. For more details on determining the target predicted thermal capacity, see Figure 3 For more details about the first preset model, see Figure 6 and its related description.

[0034] It should be understood that Figure 2 The modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above methods and systems can be implemented using computer executable instructions and / or included in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of this specification can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software such as executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0035] It should be noted that the above description of the candidate item display and determination system and its modules is only for convenience of description and does not limit the present specification to the scope of the embodiments. It is understandable that, after understanding the principle of the system, those skilled in the art may arbitrarily combine the modules or form a subsystem to connect with other modules without deviating from the principle. In some embodiments, Figure 2The model acquisition module 210, information acquisition module 220 and prediction module 230 disclosed in the specification can be different modules in a system, or a module can realize the functions of two or more modules. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the protection scope of this specification.

[0036] Figure 3 is an exemplary flow chart of heat capacity prediction according to some embodiments of the present specification. In some embodiments, process 300 may be executed by a processing device (e.g., processing device 110). For example, process 300 may be stored in a storage device in the form of a program or instruction. Figure 2 When the modules shown execute programs or instructions, process 300 can be implemented. In some embodiments, process 300 can be completed using one or more additional operations not described below, and / or not by one or more operations discussed below. Figure 3 As shown, process 300 includes the following steps:

[0037] Step 310 , obtaining a target first prediction model. In some embodiments, step 310 may be performed by the model acquisition module 210 .

[0038] The target first prediction model can be used to predict the heat capacity of the tube, and the target first prediction model can be a machine learning model. The parameters of the target first prediction model are obtained through training. The training data used for training includes first training data and second training data. The training can include initial training and can also include enhanced training. Among them, the training based on the first training data can be called "initial training", and the training based on the second training data can be called "enhanced training". For more information about "initial training" and "enhanced training", please refer to Figure 4 , 5 , 7 and related descriptions.

[0039] In some embodiments, the target first prediction model may include one or more layers. For more information about the target first prediction model structure, see Figure 6 and its related description.

[0040] In some embodiments, the type of the target first prediction model may include a convolutional neural network (CNN), a deep neural network (DNN), etc.

[0041] In some embodiments, the model acquisition module 210 may acquire the target first prediction model through training. The trained target first prediction model may be stored in a storage device (eg, storage device 140), and the model acquisition module 210 may acquire the target first prediction model from the storage device.

[0042] Step 320 , obtaining the property information of the tube in the medical imaging device and the working information of the medical imaging device. In some embodiments, step 320 may be performed by the information acquisition module 220 .

[0043] The medical imaging device can be used to obtain image data of the scanned object, and the medical imaging device may include a tube. For more details about the medical imaging device and its tube, see Figure 1 and its related description.

[0044] The attribute information of the tube may be data information related to the tube itself. In some embodiments, the attribute information of the tube may include but is not limited to at least one of the target material, target surface diameter, target angle, tube core material, filter capacity of the tube wall, heat dissipation method, anode heat dissipation coefficient, anode cooling rate, and support method between the rotor and the bearing. For example, the target material of the tube may be tungsten, molybdenum, rhodium, rhenium, etc.; the target surface diameter may be 50 mm, etc.; the target angle may be 10° or 5°, etc.; the tube core material may be ceramic, glass, etc.; the filter capacity of the tube wall may be 5 mA L / 140 KV, etc.; the heat dissipation method may be air cooling, water cooling, oil cooling, etc.; the anode heat dissipation coefficient may be 4 KW / min, etc.; the anode cooling rate may be 10 KW / min, etc.; the support method between the rotor and the bearing may be a mechanical ball bearing, a liquid metal bearing, etc. It should be noted that the attribute information of the tube may also be other information related to the tube.

[0045] In some embodiments, the attribute information of the tube may be stored in a storage device, and the information acquisition module 220 may read the attribute information of the tube from the storage device. The information acquisition module 220 may directly acquire the attribute information of the tube from the medical imaging device. In some embodiments, the attribute information of the tube of different types of medical imaging devices may be different. The attribute information of the tube of the same type of medical imaging device may also be different. For example, the attribute information of the tube in CT devices produced by different manufacturers may be different.

[0046] The working information of the medical imaging device may be data related to the imaging of the scanned object by the medical imaging device. In some embodiments, the working information may include the current thermal capacity and scanning parameters of the tube in the medical imaging device.

[0047] The current thermal capacity may be the thermal capacity of the tube at the current moment. The current moment may vary according to circumstances. For example, at the current moment, the patient is being scanned, and the current thermal capacity is the thermal capacity before the medical imaging device starts scanning. For another example, if the scan has already started, the current thermal capacity is the thermal capacity of the medical imaging device during the scanning process.

[0048] In some embodiments, the information acquisition module 220 can obtain the current heat capacity of the tube through a sensor. The sensor can be built into the medical imaging device and is used to determine the heat capacity of the tube in the medical imaging device. The sensor can be a heat capacity sensor, and the heat capacity can be directly obtained through the heat capacity sensor. The sensor can also be other sensors. For example, the sensor can be a temperature sensor and a heat sensor. The temperature detected by the temperature sensor and the heat detected by the heat sensor can be converted into the current heat capacity of the tube through a calculation formula. In some embodiments, the sensor can obtain the heat capacity of the tube at regular intervals. For example, the sensor obtains the heat capacity of the tube at intervals of 0.005s and sends the heat capacity to a storage device, etc. In some embodiments, the current heat capacity of the tube can also be obtained in other ways.

[0049] Scanning parameters are parameter information related to the scanning process of the medical imaging device. In some embodiments, the scanning parameters may include but are not limited to the tube voltage, tube current, focal spot size, anode rotation speed, scanning time of the medical imaging device, etc.

[0050] In some embodiments, the information acquisition module 220 can obtain scanning parameters from a medical imaging device, and the medical imaging device can automatically generate scanning parameters according to characteristic parameters of the scanned object (such as age, weight, physical characteristics, symptoms, etc.). In some embodiments, the information acquisition module 220 can obtain scanning parameters from a terminal (e.g., terminal 130), and the scanning parameters can be set and generated by medical personnel. The scanning parameters can be stored in a storage device (e.g., storage device 140), and the information acquisition module 220 can obtain the scanning parameters from the storage device.

[0051] Step 330 , input the attribute information of the tube and the working information of the medical imaging device into the target first prediction model to determine the target predicted thermal capacity of the tube at a preset time point. In some embodiments, step 330 may be performed by the prediction module 230 .

[0052] The preset time point is a time point after the current time and after the medical imaging device starts scanning. In some embodiments, the preset time point can be the time point when the scan ends, or it can be a time point between the start and end of the scan. The preset time point can be determined by the scan time of the medical imaging device. For example, the scan time is 1 minute, and the start scan time is 16:10, then the preset time point can be determined to be 16:11 based on the scan time. For another example, the preset time point can be set to 40 seconds after the start of the scan, 2 minutes after the start of the scan, etc. In some embodiments, the preset time point can be set by medical personnel.

[0053] In some embodiments, the preset time point can be determined according to actual conditions (e.g., whether the tube is aged, characteristic parameters of the object to be scanned, scanning mode of the medical imaging device, working environment, etc.). For example, tube aging may reduce the heat capacity of the tube (i.e., the heat carrying capacity index of the tube). When the tube is aged, the preset time point can be set to an earlier time point after the start of scanning (e.g., 30s after the start of scanning) to avoid the situation in which the heat capacity exceeds the heat capacity during the scanning process, resulting in the interruption of the scanning. For another example, when the object to be scanned is an obese object, the tube is working at a high load. Long-term high-load operation will affect the change of the heat capacity and the heat dissipation coefficient of the tube. The preset time point can be set to an earlier time point after the start of scanning (e.g., 40s after the start of scanning) to avoid the situation in which the heat dissipation coefficient of the tube decreases, resulting in the heat capacity exceeding the heat capacity.

[0054] The target predicted thermal capacity refers to the predicted value of the thermal capacity of the tube at a preset time point. In some embodiments, the prediction module 230 can determine the target predicted thermal capacity of the tube using a target first prediction model. The target first prediction model refers to the model that actually performs the thermal capacity prediction application. For more details about the target first prediction model, see Figure 6 and its related description.

[0055] In some embodiments, if the target predicted thermal capacity does not meet the preset conditions, corresponding measures can be taken in advance. The preset condition can be that the target predicted thermal capacity is less than a preset thermal capacity threshold. For example, if the scan has already started, that is, the current thermal capacity is the thermal capacity at a certain point in time during the scan, if the target predicted thermal capacity does not meet the preset conditions, the processing device can issue an instruction to stop the scan, or send a prompt message to the user of the terminal (such as terminal 130) (for example, the prompt message can be "the predicted thermal capacity at a certain point in time will exceed the thermal capacity threshold"). For another example, if the scan has not yet started, that is, the current thermal capacity is the thermal capacity before the scan starts, if the target predicted thermal capacity does not meet the preset conditions, the processing device can issue an instruction that the scan cannot be performed or send a prompt message to the user of the terminal (such as terminal 130) to avoid accidents such as excessive thermal capacity during the scan, resulting in damage to the tube or interruption of the scan.

[0056] In order to gradually improve the prediction accuracy of the target first prediction model, the target first prediction model can also be trained during the actual application process.

[0057] In some embodiments, the information acquisition module 220 can obtain the actual heat capacity of the tube at a preset time point through a sensor, and use the actual heat capacity at the preset time point and the target predicted heat capacity at the preset time point as the second training data for updating the model parameters. For example, after the actual prediction of the target first prediction model, the model acquisition module 210 can directly update the parameters of the target first prediction model based on the actual heat capacity at the preset time point and the target predicted heat capacity at the preset time point, and use the updated target first prediction model for the next heat capacity prediction application. The object of the next prediction can be the tube of the same or different medical imaging device as this time. For more details on updating the parameters of the target first prediction model based on the actual heat capacity at the preset time point and the target predicted heat capacity at the preset time point, see Figure 5 and its related description.

[0058] In some embodiments, the thermal tolerance difference between the target predicted thermal capacity and the actual thermal capacity can be used to determine whether the target first prediction model needs to be updated, or whether the target predicted thermal capacity and the actual thermal capacity can be used as the second training data. For example, when the thermal tolerance difference between the target predicted thermal capacity and the actual thermal capacity is greater than the threshold thermal tolerance difference, the model acquisition module 210 can update the parameters of the target first prediction model based on the target predicted thermal capacity and the actual thermal capacity; if it is less than the threshold thermal tolerance difference, it can be chosen not to update. In some embodiments of the present specification, the accuracy of thermal capacity prediction can be improved by predicting thermal capacity through a machine learning model. The attribute information of the tube is considered during prediction, so that the prediction can be made according to the actual situation of each tube itself, which is more practical.

[0059] Figure 4 is an exemplary flow chart of obtaining a target first prediction model according to some embodiments of the present specification. In some embodiments, process 400 may be executed by a processing device (e.g., processing device 110). For example, process 400 may be stored in a storage device in the form of a program or instruction. Figure 2 When the modules shown execute programs or instructions, process 400 can be implemented. In some embodiments, process 400 can be completed using one or more additional operations not described below, and / or not by one or more operations discussed below. Figure 4 As shown, process 400 includes the following steps:

[0060] Step 410 , constructing an initial second prediction model. In some embodiments, step 410 may be performed by the model acquisition module 210 .

[0061] The initial second prediction model may be a machine learning model. The initial second prediction model is a model that has not been trained or has not been trained well. In some embodiments, the parameters of the initial second prediction model are obtained by initialization.

[0062] In some embodiments, similar to the structure of the target first prediction model, the initial second prediction model may include multiple layers, i.e., an input layer, an incremental prediction layer, a fusion layer, and an output layer. In some embodiments, the initial second prediction model may differ from the structure of the target first prediction model. For example, the input layer of the initial second prediction model has two nodes to receive two inputs, the current heat capacity and the scanning parameter. In some embodiments, the initial second prediction model may receive two inputs, the current heat capacity and the scanning parameter, and output the target predicted heat capacity.

[0063] In some embodiments, the initial second prediction model may have the same structure as the target first prediction model. For example, the initial second prediction model includes three input nodes to receive three inputs of the property information of the tube, the current heat capacity, and the scanning parameters. In some embodiments, the initial second prediction model may receive three inputs of the property information of the tube, the current heat capacity, and the scanning parameters, and output the target predicted heat capacity.

[0064] For the structure of the target first prediction model, see Figure 6 and its related description.

[0065] It should be noted that the "three inputs" or "two inputs" in this specification do not limit the number of input values, but are used to describe the data types of the inputs. Similarly, the "three input nodes" and "two input nodes" in this specification do not limit the number of nodes, but describe the corresponding relationship with the input data types, indicating that each data type has a corresponding input node, but the specific number is not limited in this specification.

[0066] Step 420 , obtaining first training data, where the first training data includes a first training sample and a first label thereof. In some embodiments, step 420 may be performed by the model acquisition module 210 .

[0067] The first training data may be one or more groups. In some embodiments, the first training sample may include the current thermal capacity of the sample of the first sample tube and the sample scanning parameters. That is, the initial second prediction model may be trained based on the current thermal capacity of the sample of the first sample tube and the sample scanning parameters.

[0068] In some embodiments, the first training sample may include the current thermal capacity of the sample of the first sample tube, the sample scanning parameters and the sample attribute information of the first sample tube. That is, the initial second prediction model may be trained based on the current thermal capacity of the sample of the first sample tube, the sample attribute information and the sample scanning parameters.

[0069] In some embodiments, when the first training sample includes the current thermal capacity of the sample of the first sample tube, the sample scanning parameters, and the sample attribute information of the first sample tube, the sample attribute information of the first sample tube can be set to a preset value. For example, the preset value is 0. Accordingly, the initial second prediction model is actually only trained based on the current thermal capacity of the sample and the sample scanning parameters.

[0070] As described above, the input node of the initial second prediction model can receive two inputs or three inputs. In some embodiments, when the initial second prediction model has a different structure from the target first prediction model, that is, when the input node of the initial second prediction model receives two inputs, the first training sample includes the current heat capacity and sample scanning parameters of the first sample tube. When the initial second prediction model has the same structure as the target first prediction model, that is, when the input node of the initial second prediction model receives three inputs, the first training sample includes the current heat capacity, sample scanning parameters and sample attribute information of the first sample tube. The sample attribute information of the first sample tube can be a preset value (e.g., 0) or the actual attribute information of the first sample tube.

[0071] The first label corresponding to the first training sample is the real heat capacity of the first sample tube at the preset time point of the first sample. The current heat capacity of the first sample tube refers to the heat capacity at a certain time point before the preset time point of the first sample. Similar to the current heat capacity, the current heat capacity of the sample can be the heat capacity of the first sample tube under different circumstances. The first sample tubes of different first training samples can be the same or different.

[0072] In some embodiments, the first training data may be stored in a storage device, and the model acquisition module 210 may acquire the first training data from the storage device. The first training data may be acquired based on historical scanning, wherein the first label of the first training sample may be acquired by a sensor during the historical scanning process.

[0073] Step 430 : Update the initial second prediction model based on the first training data to determine the target second prediction model. In some embodiments, step 430 may be performed by the model acquisition module 210 .

[0074] The parameters of the target second prediction model are determined through training. The target second prediction model has the same structure as the initial second prediction model. It can be understood that, similar to the initial second prediction model, the structure of the target second prediction model can be the same as or different from the structure of the target first prediction model, and the structure of the target second prediction model is determined by the initial second prediction model.

[0075] In some embodiments, the model acquisition module 210 can input the first training sample with the first label into the initial second prediction model, construct a loss function based on the output of the initial second prediction model (i.e., the target predicted heat capacity) and the first label, and iteratively update the parameters of the initial second prediction model based on the loss function. When the trained model meets the preset conditions, the training ends and the target second prediction model is determined. The preset conditions are that the loss function converges, the number of iterations reaches a threshold, etc.

[0076] Step 440 , determining a target first prediction model based on the target second prediction model. In some embodiments, step 440 may be performed by the model acquisition module 210 .

[0077] When the target first prediction model and the target second prediction model have the same structure, in some embodiments, the model acquisition module 210 can directly use the target second prediction model as the target first prediction model. When the target first prediction model and the target second prediction model have different structures, in some embodiments, the model acquisition module 210 can migrate the parameters of the target second prediction model to the initial first prediction model, and use the first prediction model after parameter migration as the target first prediction model.

[0078] The initial first prediction model refers to a model that has not been trained or has not been trained yet. The structure of the initial first prediction model is the same as that of the target first prediction model. The parameters of the initial first prediction model are obtained by initialization.

[0079] In some embodiments, the model acquisition module 210 may determine the intermediate first prediction model based on the target second prediction model, and further determine the target first prediction model based on the intermediate first prediction model.

[0080] The intermediate first prediction model refers to a model determined by updating the parameters of the initial first prediction model based on the first training data. The intermediate first prediction model has not been subjected to enhanced training.

[0081] In some embodiments, the model acquisition module 210 can directly use the intermediate first prediction model as the target first prediction model for heat capacity prediction application. In other words, the target first prediction model applied at this time has only undergone initial training. In some embodiments, the model acquisition module 210 can update the intermediate first prediction model based on the second training data (i.e., enhanced training), and use the updated intermediate first prediction model as the target first prediction model for heat capacity prediction application. In other words, the target first prediction model applied at this time has undergone initial training and enhanced training. For more information about enhanced training, see Figure 5 and its related description.

[0082] As described above, the target first prediction model may have the same structure as the target second prediction model. In some embodiments, the model acquisition module 210 may directly use the target second prediction model obtained by training based on the first training data as the intermediate first prediction model. As described above, the target first prediction model may have a different structure from the target second prediction model. In some embodiments, the model acquisition module 210 may migrate the parameters of the target second prediction model to the initial first prediction model, and use the initial first prediction model after parameter migration as the intermediate first prediction model. In some embodiments, the model acquisition module 210 may migrate all or part of the parameters of the target second prediction model to the initial first prediction model. For example, the model acquisition module 210 may update the parameters of the incremental prediction layer, the fusion layer, and the output layer in the initial first prediction model only based on the migration. After the migration, some or all of the initialization parameters in the initial first prediction model are updated to the trained parameters.

[0083] The structures of the target first prediction model and the target second prediction model are different. Through the parameter migration of the target second prediction model, some parameters in the initial first prediction model are updated, and other parameters are still initialized values. For example, in the target first prediction model, the parameters of the input nodes corresponding to the attribute information of the tube are still initialized values, and other parameters are trained parameters. On this basis, the model acquisition module 210 can update the parameters of the input nodes corresponding to the attribute information of the tube through enhanced training based on the target first prediction model in actual applications.

[0084] The attribute information of the tube is closely related to the medical imaging device itself. In actual application, the attribute information of the tubes of different medical imaging devices is different. In some embodiments of the present specification, the initial training can be based only on the current thermal capacity and sample scanning parameters of the sample tube, without considering the sample attribute information of the sample tube, so that the target first prediction model obtained after the initial training can basically achieve thermal capacity prediction. In actual application, enhanced training is performed based on the prediction results and the actual detected thermal capacity values, and the parameters of the target first prediction model are updated to improve the prediction accuracy of the target first prediction model, so that it can be predicted according to the attribute information of the tube of each device. It can be understood that when acquiring the first training data, it is not necessary to consider the attribute information of the tube in the medical imaging device, which can reduce the cost of acquiring training data while ensuring the accuracy of model prediction.

[0085] In some embodiments, the model acquisition module 210 can directly train the initial first prediction model based on the first training data to determine the target first prediction model. The training samples include the current thermal capacity of the sample tube, the sample scanning parameters and the sample attribute information of the sample tube. In some embodiments, the sample attribute information of the sample tube is the actual attribute information. The label of the training sample is the same as that of the training initial second prediction model. The model acquisition module 210 constructs a loss function based on the label and the output of the initial first prediction model, and updates the parameters of the initial first prediction model. In some embodiments, after the model acquisition module 210 obtains the intermediate first prediction model, the intermediate first prediction model can be directly used as the target first prediction model, or the parameters of the intermediate first prediction model can be updated based on the second training data (i.e., enhanced training) as the target first prediction model.

[0086] It should be noted that the above description of process 400 is only for example and explanation, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 400 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0087] Figure 5 is another exemplary flow chart of obtaining a target first prediction model according to some embodiments of the present specification. In some embodiments, process 500 may be executed by a processor (e.g., processing device 110). For example, process 500 may be stored in a storage device in the form of a program or instruction. Figure 2 When the modules shown execute programs or instructions, process 500 can be implemented. In some embodiments, process 500 can be completed using one or more additional operations not described below, and / or not by one or more operations discussed below. Figure 5 As shown, process 500 includes the following steps:

[0088] Step 510 , obtaining second training data, where the second training data includes a second training sample and a second label thereof. In some embodiments, step 510 may be performed by the model acquisition module 210 .

[0089] In some embodiments, the second training sample may include a sample target predicted heat capacity of the second sample tube at a second sample preset time point. The second training sample may be obtained during an actual application. For example, in an actual application, the target predicted heat capacity of the tube determined using the target first prediction model may be used as the second training sample.

[0090] The second label corresponding to the second training sample is the real heat capacity of the second sample tube at the preset time point of the second sample. In some embodiments, the information acquisition module 220 can obtain the real heat capacity of the second sample tube at the preset time point of the second sample through a sensor. For example, in actual applications, the real heat capacity of the tube at the preset time point obtained by the sensor can be used as the second label.

[0091] The second sample tubes of different second training samples may be the same or different. The first sample tube and the second sample tube may be the same or different.

[0092] Step 520 : Determine an intermediate first prediction model based on the target second prediction model. In some embodiments, step 520 may be performed by the model acquisition module 210 .

[0093] For the intermediate first prediction model and determining the intermediate first prediction model based on the target second prediction model, please refer to step 440 and its related description.

[0094] Step 530 , based on the predicted second training data, the intermediate first prediction model is updated to obtain the target first prediction model. In some embodiments, step 530 may be performed by the model acquisition module 210 .

[0095] In some embodiments, the model acquisition module 210 may compare the target predicted heat capacity of the second sample tube at the second sample preset time point with the actual heat capacity of the tube at the preset time point, update the parameters of the intermediate first prediction model, and use the intermediate first prediction model with the updated parameters as the target first prediction model. For example, the model acquisition module 210 constructs a loss function based on the target predicted heat capacity of the second sample tube at the second sample preset time point and the second label, and updates the parameters of the intermediate first prediction model based on the loss function.

[0096] Figure 6 It is an exemplary schematic diagram of the target first prediction model structure shown in some embodiments of this specification.

[0097] like Figure 6 As shown, in some embodiments, the target first prediction model includes an input layer 610 , an incremental prediction layer 620 , a fusion layer 630 , and an output layer 640 .

[0098] The input layer 610 can be used to input the property information of the tube, the current heat capacity and the scanning parameters. The input layer includes input nodes for receiving the property information of the tube, the current heat capacity and the scanning parameters.

[0099] The increment prediction layer 620 can determine the increment of the thermal capacity of at least one position in the tube at a preset time point based on the attribute information and the working information (ie, the scanning parameters and the current thermal capacity). The thermal capacity increment refers to the amount of thermal capacity increase between the current time and the preset time point.

[0100] like Figure 1 The bulb includes a plurality of components. In some embodiments, at least one position in the bulb may be a position of a different component in the bulb or a different position point on a different component. In some embodiments, the heat capacity increments at different positions in the bulb may be different. For example, the heat capacity increment of the filament is greater than the heat capacity increment of the buffer tube. For example, at least one position includes n positions, and the heat capacity increments at the n positions are heat capacity increment 1, heat capacity increment 2, ... heat capacity increment n.

[0101] The fusion layer 630 can determine the initial predicted heat capacity of at least one position at a preset time point based on the current heat capacity and the heat capacity increment of at least one position in the tube at the preset time point. In some embodiments, for one of the at least one position, the fusion layer 630 can fuse the heat capacity increment of the position, the heat capacity increment of other positions in the at least one position, and the current heat capacity to determine the initial predicted heat capacity of the position at the preset time point. The other positions in the at least one position refer to one or more positions other than the at least one position in the at least one position. It can be understood that the heat capacity increments at different positions in the tube may be different, and the heat capacity changes between different positions may be affected by each other.

[0102] In some embodiments, for one of the at least one positions, other positions to be fused can be determined in a variety of ways, and the fusion layer 630 can fuse other positions and the heat capacity increment of the position and the current heat capacity of the tube to determine the initial predicted heat capacity of the position at a preset time point. For example, other fused positions corresponding to the position can be determined based on the position relationship. For example, other positions that are closer to the position (for example, the distance is less than a threshold) can be used as other fused positions. For another example, other fused positions corresponding to the position can be preset.

[0103] In some embodiments, for any one of the at least one positions, the fusion layer 630 may directly fuse the heat capacity increment and the current heat capacity of the position to determine the initial predicted heat capacity of the position at a preset time point. For example, for position a, the fusion layer 630 may fuse the heat capacity increment and the current heat capacity of position a to determine the initial predicted heat capacity of position a at a preset time.

[0104] In some embodiments, there may be multiple fusion modes, such as averaging, weighted averaging, taking the median, taking the maximum value, etc. For example, when weighted averaging is performed, the value of the weight may be determined based on the influence factors of other positions on the position.

[0105] Exemplarily, the calculation formula (1) for the initial predicted heat capacity may be:

[0106] y i =m1 *x 1 +m 2 *x 2 +…+m j *x j +m i *x i +k*C (1)

[0107] Among them, y i is the initial predicted heat capacity at location i, x 1 、x 2 , ..., x j are the heat capacity increments of other locations where fusion is performed at location i, and x i is the heat capacity increment at position i, C is the current heat capacity, m 1 、m 2 ,...,m j 、m i x 1 、x 2 , ..., x j 、x i Impact factor, k is the coefficient. m 1 、m 2 ,...,m j 、m i and k can be preset in advance or obtained through model training.

[0108] As another example, at least one position includes position 1, position 2 and position 3. The initial predicted heat capacities of position 1, position 2 and position 3 are y1, y2 and y3 respectively, wherein position 1 does not have other fused positions, position 2 has other fused positions as position 1, and position 3 has other fused positions as position 1 and position 2. The initial predicted heat capacities of position 1, position 2 and position 3 can be determined by the following formula:

[0109] y1=t 1 *x 1 +t 2 *C (1)

[0110] y2=t 3 *y 1 +t 4 *x 2 +t 5 *C (2)

[0111] y3=t 6 *y1+t 7 *y2+t 8 *x 3 +t 9 *C...... (3)

[0112] Among them, y1, y2, and y3 are the initial predicted heat capacities of positions 1, 2, and 3 respectively, and x 1 、x 2 、x 2 are the heat capacity increments at positions 1, 2, and 3, respectively, C is the current heat capacity, t 1 ,t 2 ,t 3 ,t 4 ,t 5 ,t 6 ,t 7 ,t 8 ,t 9 The parameters obtained by model training.

[0113] It is understandable that the calculation method for obtaining the initial predicted heat capacity after fusion may be other methods, and the above formula is only used as an example.

[0114] The output layer 640 may determine the target predicted heat capacity based on the initial predicted heat capacity of at least one position at a preset time point. For example, the output layer 640 may fuse the initial predicted heat capacity of at least one position and use the fused value as the target predicted heat capacity. The fusion method includes averaging, weighted averaging, etc. For another example, the output layer 640 may select the maximum value of multiple initial predicted heat capacities as the target predicted heat capacity of the tube at the preset time point.

[0115] In some embodiments, the model type of one or more layers in the target first prediction model may be a neural network model, such as CNN, DNN, etc. For example, the incremental prediction layer 620 may be a CNN network.

[0116] Figure 7 2 is an exemplary schematic diagram of a target first prediction model training process according to some embodiments of the present specification. In some embodiments, training the target first prediction model may be performed by the processing device 110 (eg, the model acquisition module 210).

[0117] The training of the target first prediction model may include initial training and enhanced training. In actual application, the target first prediction model may be obtained through initial training, or through initial training and enhanced training.

[0118] like Figure 7 As shown, the initial training may include two methods: first initial training 710 and second initial training 720. The model acquisition module 210 may perform initial training based on any method to obtain the target first prediction model 1. The target first prediction model 1 has not been enhanced. The target first prediction model 1 is an intermediate prediction model.

[0119] The first initial training 710 trains the initial second prediction model based on the first training data to determine the target second prediction model, and obtains the target first prediction model 1 based on the target second prediction model. For more details on determining the target first prediction model based on the second prediction model, see Figure 4 And the related description will not be repeated here.

[0120] The first initial training 720 trains the initial first prediction model based on the first training data to obtain the target first prediction model 1.

[0121] When the target first prediction model 1 is actually applied, the parameters can be updated through enhanced training 730 to improve the prediction ability of the model. When the processing device uses the target first prediction model 1 for the first prediction, the obtained target predicted heat capacity and the corresponding real heat capacity are used as a set of second training data, and the parameters of the target first prediction model 1 are updated based on the set of training data to obtain the target first prediction model 2. The target first prediction model 2 can be used as the model for the next prediction. When the processing device uses the target first prediction model 2 for the second prediction, the target predicted heat capacity obtained by the second prediction and the corresponding real heat capacity are used as a set of second training data, and the parameters of the target first prediction model 2 are updated based on the set of training data to obtain the target first prediction model 3. By analogy, the target first prediction model can be subjected to multiple enhanced training.

[0122] In some embodiments, when the thermal tolerance difference between the target predicted thermal capacity and the actual thermal capacity obtained after multiple enhanced training of the target first prediction model meets certain conditions (eg, the thermal tolerance difference is less than a thermal tolerance threshold), the enhanced training may be stopped.

[0123] While the target first prediction model is used to determine the target predicted heat capacity of the tube, the parameters of the target first prediction model can be continuously updated according to the actual heat capacity corresponding to the tube, so that the updated target first prediction model can achieve more accurate prediction.

[0124] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0125] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0126] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0127] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0128] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0129] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0130] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A method for predicting the thermal capacity of medical imaging equipment, characterized in that: The method comprises: Obtaining a target first prediction model, wherein the target first prediction model is a machine learning model; Acquiring attribute information of a tube in a medical imaging device and operating information of the medical imaging device, wherein the operating information includes current thermal capacity and scanning parameters of the tube; and The attribute information and the working information are input into the target first prediction model to determine the target predicted thermal capacity of the tube at a preset time point; wherein the target first prediction model includes an input layer, an incremental prediction layer, a fusion layer and an output layer, The input layer is used to input the attribute information and the work information; The increment prediction layer determines the heat capacity increment of at least one position in the tube at the preset time point based on the attribute information and the working information; The fusion layer determines, based on the current heat capacity and the heat capacity increment of the at least one position in the bulb at the preset time point, an initial predicted heat capacity of the at least one position at the preset time point; and The output layer determines the target predicted heat capacity based on the initial predicted heat capacity of the at least one location at the preset time point.

2. The method according to claim 1, characterized in that The obtaining of the first prediction model of the target comprises: constructing an initial second prediction model; Acquire first training data, the first training data including a first training sample and a first label thereof, the first training sample including a current sample heat capacity and a sample scanning parameter of a first sample tube, the first label being a real heat capacity of the first sample tube at a preset time point of the first sample; Update the initial second prediction model based on the first training data to determine a target second prediction model; and Based on the target second prediction model, the target first prediction model is determined.

3. The method according to claim 2, characterized in that The first training sample also includes sample attribute information of the sample tube.

4. The method according to claim 2, characterized in that: The determining the target first prediction model based on the target second prediction model comprises: Acquire second training data, the second training data including a second training sample and a second label thereof, the second training sample including a sample target predicted heat capacity of a second sample tube at a second sample preset time point, and the second label being a real heat capacity of the second sample tube at the second sample preset time point; Determining an intermediate first prediction model based on the target second prediction model; The intermediate first prediction model is updated based on the second training data to obtain the target first prediction model.

5. The method according to claim 1, characterized in that The fusion layer determines, based on the current heat capacity and the heat capacity increment of the at least one position in the tube at the preset time point, the initial predicted heat capacity of the at least one position at the preset time point, including: For one of the at least one position, the fusion layer fuses the heat capacity increment of the position, the heat capacity increments of other positions in the at least one position, and the current heat capacity to determine the initial predicted heat capacity of the position at the preset time point.

6. The method according to claim 1, characterized in that The fusion layer determines, based on the current heat capacity and the heat capacity increment of the at least one position in the tube at the preset time point, the initial predicted heat capacity of the at least one position at the preset time point, including: For any one of the at least one position, the fusion layer fuses the heat capacity increment and the current heat capacity of the position to determine the initial predicted heat capacity of the position at the preset time point.

7. A thermal capacity prediction system for medical imaging equipment, characterized in that: The system comprises: A model acquisition module, used to acquire a target first prediction model, wherein the target first prediction model is a machine learning model; an information acquisition module, used to acquire attribute information of a tube in a medical imaging device and operating information of the medical imaging device, wherein the operating information includes current thermal capacity and scanning parameters of the tube; and A prediction module is used to input the attribute information and the working information into the target first prediction model to determine the target predicted thermal capacity of the tube at a preset time point; wherein the target first prediction model includes an input layer, an incremental prediction layer, a fusion layer and an output layer, The input layer is used to input the attribute information and the work information; The increment prediction layer determines the heat capacity increment of at least one position in the tube at the preset time point based on the attribute information and the working information; The fusion layer determines, based on the current heat capacity and the heat capacity increment of the at least one position in the bulb at the preset time point, an initial predicted heat capacity of the at least one position at the preset time point; and The output layer determines the target predicted heat capacity based on the initial predicted heat capacity of the at least one location at the preset time point.

8. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions, and when the computer instructions are executed by the processor, the thermal capacity prediction method of the medical imaging device according to any one of claims 1 to 6 is implemented.

9. A medical imaging device, characterized in that: The device comprises at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is used to execute at least part of the computer instructions to implement the thermal capacity prediction method for medical imaging equipment as described in any one of claims 1 to 6.

Images