Method, apparatus, storage medium and program product for determining exposure parameters

Abstract

The embodiment of the application discloses a method, device, storage medium and program product for determining exposure parameters in X-ray imaging. The method comprises: obtaining a three-dimensional image of a subject; extracting a body shape feature of the subject based on the three-dimensional image; inputting the body shape feature into an exposure parameter prediction model corresponding to an X-ray imaging protocol, wherein the exposure parameter prediction model is trained based on first training data, and the first training data contains historical body shape features in historical exposure operations of the X-ray imaging protocol and historical values of exposure parameters in the historical exposure operations; and receiving an exposure parameter prediction value determined based on the body shape feature from the exposure parameter prediction model. The embodiment of the application automatically determines exposure parameters according to a three-dimensional image of an imaging target, improves exposure accuracy, and reduces manual difficulty.

Description

Technical Field

[0001] This invention relates to the field of medical imaging technology, and in particular to a method, apparatus, storage medium, and program product for determining exposure parameters in X-ray imaging. Background Technology

[0002] X-rays are electromagnetic radiation with wavelengths between ultraviolet and gamma rays. X-rays are penetrating, penetrating materials of varying densities. In medicine, X-rays are generally used to project images of human organs and bones to create medical images. An X-ray imaging system typically includes an X-ray generator assembly, a Bucky-Wall-Stand (BWS) assembly, an examination table assembly, a film cassette assembly containing a flat panel detector, and a remote control unit. The X-ray generator assembly uses high voltage provided by a high-voltage generator to emit X-rays that pass through and irradiate the target, forming a medical image of the target on the flat panel detector. The flat panel detector transmits the medical image information to the control unit. The target can stand near the Bucky-Wall-Stand assembly or lie on the examination table assembly to receive X-ray imaging of different parts of the body, such as the head, chest, abdomen, and joints.

[0003] In X-ray imaging, X-ray exposure parameters (such as tube voltage, tube current, and exposure time) have a significant impact on X-ray image quality. Currently, X-ray exposure parameters are mainly set based on the technician's personal experience, which presents difficulties in implementation. Summary of the Invention

[0004] The present invention provides a method, apparatus, storage medium, and program product for determining exposure parameters in X-ray imaging.

[0005] A method for determining exposure parameters in X-ray imaging, the method comprising:

[0006] Acquire a 3D image of the subject;

[0007] Based on the three-dimensional image, the body shape features of the subject are extracted;

[0008] The body shape features are input into the exposure parameter prediction model corresponding to the X-ray imaging protocol, wherein the exposure parameter prediction model is trained based on the first training data, which includes historical body shape features and historical exposure parameter values ​​from historical exposure operations of the X-ray imaging protocol.

[0009] Receive the exposure parameter prediction values ​​determined based on the body shape characteristics from the exposure parameter prediction model.

[0010] Therefore, the embodiments of the present invention automatically determine exposure parameters based on the three-dimensional image of the subject, improving exposure accuracy and reducing manual difficulty. Furthermore, the correlation between the subject's body shape characteristics and the exposure parameters is considered during the automatic determination process, further improving parameter accuracy.

[0011] In the exemplary implementation, it also includes:

[0012] Establish the first artificial neural network model;

[0013] The first artificial neural network model is trained into the exposure parameter prediction model using the first training data, wherein the training includes:

[0014] The first training data is input into the first artificial neural network model, so that the first artificial neural network model outputs the exposure parameter prediction value corresponding to the first training data; based on the difference between the historical exposure parameter value in the first training data and the exposure parameter prediction value corresponding to the first training data, the loss function value of the first artificial neural network model is determined; the model parameters of the first artificial neural network model are configured so that the loss function value is lower than a first preset threshold; the configured first artificial neural network model is determined as the exposure parameter prediction model.

[0015] Therefore, by introducing artificial neural networks into the exposure parameter prediction process, the implementation difficulty is reduced.

[0016] In an exemplary implementation, extracting the body shape features of the subject based on the three-dimensional image includes:

[0017] The three-dimensional image is input into the body shape feature determination model, wherein the body shape feature determination model is trained based on second training data, the second training data including historical three-dimensional images and body shape features labeled based on the historical three-dimensional images; and body shape features determined based on the three-dimensional image are received from the body shape feature determination model.

[0018] Therefore, it enables the automatic extraction of body shape features from 3D images, thus improving feature extraction efficiency.

[0019] In the exemplary implementation, it also includes:

[0020] Establish a second artificial neural network model;

[0021] The second artificial neural network model is trained into the body shape feature determination model using the second training data, the training including:

[0022] The second training data is input into the second artificial neural network model, so that the second artificial neural network model outputs the predicted body shape features corresponding to the second training data; based on the difference between the labeled body shape features in the second training data and the predicted body shape features, the loss function value of the second artificial neural network model is determined; the model parameters of the second artificial neural network model are configured so that the loss function value is lower than a second preset threshold; the configured second artificial neural network model is determined as the body shape feature determination model.

[0023] Therefore, by introducing artificial neural networks into the body shape feature extraction process, the implementation difficulty has been reduced.

[0024] In an exemplary embodiment, the method further includes: determining whether an adjustment instruction for the predicted exposure parameter value is received; wherein when an adjustment instruction for the predicted exposure parameter value is received, the predicted exposure parameter value is adjusted based on the adjustment instruction, and a first exposure operation is performed on the imaging target based on the adjusted predicted exposure parameter value; and when no adjustment instruction for the predicted exposure parameter value is received, a second exposure operation is performed on the imaging target based on the predicted exposure parameter value.

[0025] Therefore, the predicted values ​​of exposure parameters can be adjusted based on the adjustment operation, thus ensuring the accuracy of the predicted values.

[0026] In an exemplary embodiment, the method further includes: after performing the first exposure operation, storing third training data in a log, the third training data including the adjusted exposure parameter prediction values ​​for performing the first exposure operation and the body shape features of the imaging target; extracting the third training data from the log; and performing an update training process on the exposure parameter prediction model using the third training data, the update training process including: inputting the third training data into the exposure parameter prediction model to output exposure parameter prediction values ​​corresponding to the third training data; determining a loss function value for the exposure parameter prediction model based on the difference between the adjusted exposure parameter prediction values ​​in the third training data and the exposure parameter prediction values ​​corresponding to the third training data; configuring the model parameters of the exposure parameter prediction model so that the loss function value is lower than a third preset threshold; and determining the configured exposure parameter prediction model as the updated exposure parameter prediction model.

[0027] It is evident that updating and training the exposure parameter prediction model using logs can introduce adjustment preferences into the model, making it more in line with user habits.

[0028] In an exemplary embodiment, the body shape feature includes a first body shape feature with a first weight and a second body shape feature with a second weight, wherein the first body shape feature characterizes the overall body shape of the subject, and the second body shape feature characterizes the local body shape of the subject that corresponds to the imaging target of the X-ray imaging protocol; wherein the second weight is greater than the first weight.

[0029] Therefore, the prediction result is obtained by combining the body shape characteristics of the subject and the imaging target. The prediction result comprehensively weighs the overall body shape and local body shape, ensuring accuracy. Moreover, the weight of local body shape is greater than that of overall body shape, which improves the correlation between the prediction result and the imaging target.

[0030] An apparatus for determining exposure parameters in Z-ray imaging, the apparatus comprising:

[0031] The acquisition module is configured to acquire a three-dimensional image of the imaging target;

[0032] The extraction module is configured to extract the body shape features of the imaging target based on the three-dimensional image;

[0033] The input module is configured to input the body shape features into an exposure parameter prediction model corresponding to an X-ray imaging protocol, wherein the exposure parameter prediction model is trained based on first training data, the first training data including historical body shape features and historical values ​​of exposure parameters in historical exposure operations of the X-ray imaging protocol.

[0034] The receiving module is configured to receive exposure parameter prediction values ​​determined based on the body shape characteristics from the exposure parameter prediction model.

[0035] Therefore, the embodiments of the present invention automatically determine exposure parameters based on the three-dimensional image of the subject, improving exposure accuracy and reducing manual difficulty. Furthermore, the correlation between the subject's body shape characteristics and the exposure parameters is considered during the automatic determination process, further improving parameter accuracy.

[0036] In an exemplary embodiment, the extraction module is configured to input the three-dimensional image into a body shape feature determination model, wherein the body shape feature determination model is trained based on second training data, the second training data including historical three-dimensional images and body shape features annotated based on the historical three-dimensional images; and to receive body shape features determined based on the three-dimensional image from the body shape feature determination model.

[0037] In the exemplary implementation, it also includes:

[0038] An exposure module is configured to determine whether an adjustment instruction for the predicted exposure parameter value is received; wherein when an adjustment instruction for the predicted exposure parameter value is received, the predicted exposure parameter value is adjusted based on the adjustment instruction, and a first exposure operation is performed on the imaging target based on the adjusted predicted exposure parameter value; when no adjustment instruction for the predicted exposure parameter value is received, a second exposure operation is performed on the imaging target based on the predicted exposure parameter value.

[0039] Therefore, the predicted values ​​of exposure parameters can be adjusted based on the adjustment operation, thus ensuring the accuracy of the predicted values.

[0040] In the exemplary implementation, it also includes:

[0041] An update module is configured to, after performing the first exposure operation, store third training data in a log, the third training data including the adjusted exposure parameter prediction values ​​and the body shape features of the imaging target after performing the first exposure operation; extract the third training data from the log; and perform an update training process on the exposure parameter prediction model using the third training data, the update training process including: inputting the third training data into the exposure parameter prediction model so that the exposure parameter prediction model outputs exposure parameter prediction values ​​corresponding to the third training data; determining the loss function value of the exposure parameter prediction model based on the difference between the adjusted exposure parameter prediction value in the third training data and the exposure parameter prediction value corresponding to the third training data; configuring the model parameters of the exposure parameter prediction model so that the loss function value is lower than a third preset threshold; and determining the configured exposure parameter prediction model as the updated exposure parameter prediction model.

[0042] It is evident that updating and training the exposure parameter prediction model using logs can introduce adjustment preferences into the model, making it more in line with user habits.

[0043] An apparatus for determining exposure parameters in X-ray imaging, comprising a processor and a memory;

[0044] The memory stores an application program that can be executed by the processor, which causes the processor to perform a method for determining exposure parameters in X-ray imaging as described above.

[0045] A computer-readable storage medium storing computer-readable instructions that, when executed by a processor, implement the method for determining exposure parameters in Z-ray imaging as described above.

[0046] A computer program product includes a computer program that, when executed by a processor, implements a method for determining exposure parameters in X-ray imaging as described above. Attached Figure Description

[0047] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which will make the above and other features and advantages of the present invention more apparent to those skilled in the art. In the drawings:

[0048] Figure 1 This is an exemplary flowchart of a method for determining exposure parameters in X-ray imaging according to an embodiment of the present invention.

[0049] Figure 2 An exemplary flowchart for determining the body shape characteristics of a subject according to an embodiment of the present invention.

[0050] Figure 3 This is a schematic diagram illustrating the storage of exposure operation logs according to an embodiment of the present invention.

[0051] Figure 4 This is a schematic diagram illustrating an exemplary process for determining exposure parameters in X-ray imaging according to an embodiment of the present invention.

[0052] Figure 5 This is an exemplary structural diagram of an apparatus for determining exposure parameters in X-ray imaging according to an embodiment of the present invention.

[0053] Figure 6 This is an exemplary structural diagram of an apparatus for determining exposure parameters in Z-ray imaging according to an embodiment of the present invention.

[0054]

[0055]

[0056] Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the present invention clearer, the following embodiments are provided to further illustrate the present invention in detail.

[0058] For the sake of brevity and intuitiveness, the following description uses several representative embodiments to illustrate the solution of the present invention. Numerous details in the embodiments are only used to aid in understanding the solution of the present invention. However, it is obvious that the technical solution of the present invention can be implemented without being limited to these details. To avoid unnecessarily obscuring the solution of the present invention, some embodiments are not described in detail, but only a framework is given. In the following text, "comprising" means "including but not limited to," and "according to..." means "at least according to..., but not limited to only according to...". Due to Chinese language habits, unless the quantity of a component is specifically indicated below, it means that the component can be one or more, or can be understood as at least one.

[0059] In view of the shortcomings of existing technologies that rely on technicians' personal experience to set X-ray exposure parameters, the present invention takes into account the correlation between the body shape characteristics of the subject and the exposure parameters (generally, the larger the body shape of the subject, the larger the exposure parameters). The body shape characteristics are automatically determined based on the three-dimensional image of the subject, and then the exposure parameters are automatically determined based on the body shape characteristics. This not only improves the exposure accuracy but also reduces the manual difficulty.

[0060] Figure 1 This is a flowchart illustrating a method for determining exposure parameters in X-ray imaging according to an embodiment of the present invention. Preferably, it can be executed by a controller. Figure 1 The method is shown. The controller can be implemented as a control host integrated into the X-ray imaging system, or as a control unit independent of the control host.

[0061] like Figure 1 As shown, the method 100 includes:

[0062] Step 101: Obtain the three-dimensional image of the subject.

[0063] In one embodiment, in step 101, a camera component may be used to capture a three-dimensional image of the subject. In another embodiment, in step 101, a three-dimensional image of the subject may be retrieved from a storage medium (e.g., a cloud or local database), wherein the three-dimensional image was obtained by capturing the subject using the camera component.

[0064] Here, the light source of the imaging component can be either aligned with or not aligned with the Z-ray source in the X-ray imaging system. When the light source of the imaging component is aligned with the Z-ray source in the X-ray imaging system, the imaging component is usually fixed to the X-ray tube housing or the beam beam housing of the X-ray generating component. For example, a groove for accommodating the imaging component is arranged on the X-ray tube housing or the beam beam housing, and the imaging component is fixed to the groove by means of bolt connection, snap connection, wire rope loop, etc. When the light source of the imaging component is not aligned with the X-ray source in the X-ray imaging system, the imaging component can be placed in any location in the examination room where the subject is located, suitable for imaging the subject, such as on the ceiling, the floor, or on various components of the medical imaging system, etc.

[0065] In one embodiment, the imaging component includes at least one 3D camera. The 3D camera uses 3D imaging technology to capture images of the subject to generate a 3D image of the subject.

[0066] In one embodiment, the imaging assembly includes at least two two-dimensional cameras, each positioned at a predetermined location. In practice, those skilled in the art can select suitable locations as the predetermined locations to position the two-dimensional cameras as needed. The imaging assembly may further include an image processor. The image processor synthesizes the two-dimensional images captured by the individual two-dimensional cameras into a three-dimensional image of the subject, wherein the depth of field used by the image processor in the synthesis can be any depth of field of the two-dimensional images. Optionally, each two-dimensional camera can send its captured two-dimensional images to an image processor outside the imaging assembly, so that the image processor outside the imaging assembly can synthesize the two-dimensional images captured by the individual two-dimensional cameras into a three-dimensional image of the subject, wherein the depth of field used by the image processor outside the imaging assembly in the synthesis process can also be any depth of field of the two-dimensional images. Specifically, the image processor outside the imaging assembly can be implemented as a control host in an X-ray imaging system, or as an independent control unit separate from the X-ray imaging system. Each two-dimensional camera can be positioned in the examination room where the subject is located, at any location suitable for photographing the subject, such as on the ceiling, floor, or on various components of the X-ray imaging system, etc.

[0067] In one embodiment, the imaging assembly may include at least one two-dimensional camera and at least one depth sensor. The at least one two-dimensional camera and at least one depth sensor are mounted in the same location. The imaging assembly may further include an image processor. The image processor uses the depth of field provided by the depth sensor and the two-dimensional photograph provided by the two-dimensional camera to jointly generate a three-dimensional image of the subject. Optionally, the two-dimensional camera sends the captured two-dimensional image of the subject to an image processor outside the imaging assembly, and the depth sensor sends the acquired depth of field to the image processor outside the imaging assembly, so that the image processor outside the imaging assembly uses the depth of field and the two-dimensional photograph to jointly generate a three-dimensional image of the subject. Preferably, the image processor outside the imaging assembly can be implemented as a control host in an X-ray imaging system, or as an independent control unit separate from the Z-ray imaging system. The two-dimensional camera can be arranged in the examination room where the subject is located, at any location suitable for photographing the subject, such as on the ceiling, on the floor, or on various components of the Z-ray imaging system, etc.

[0068] After the camera component acquires a 3D image of the subject, it can send the 3D image to the execution unit via a wired or wireless interface. Figure 1 The process controller. Preferably, the wired interface includes at least one of the following: a universal serial bus interface, a controller area network interface, a serial port, etc.; the wireless interface includes at least one of the following: an infrared interface, a near-field communication interface, a Bluetooth interface, a VPN interface, a wireless broadband interface, etc.

[0069] The above exemplary descriptions illustrate a typical example of a camera assembly capturing images of a subject to generate a three-dimensional image. Those skilled in the art will recognize that such descriptions are merely exemplary and are not intended to limit the scope of protection of the embodiments of the present invention.

[0070] Step 102: Extract the body shape features of the subject based on the 3D image.

[0071] Body shape is a general description and assessment of the human form, mainly including the length, width, and proportions of each part. Body shape is related to the subject's motor abilities and other functions. Generally, the larger the subject's body shape, the higher the exposure parameters should be.

[0072] Body shape characteristics are used to characterize the body shape of the test subject. In one implementation, body shape characteristics include a first body shape characteristic that characterizes the overall body shape of the test subject. For example, the first body shape characteristic may include: height, weight, upper limb length, lower limb length, upper limb to lower limb length ratio, chest circumference, waist circumference, hip circumference, waist-to-hip ratio, height-waist circumference index, and body mass index (BMI), etc.

[0073] In one implementation, body shape features include a second body shape feature characterizing the body shape of the subject, corresponding to the imaging target of the X-ray imaging protocol. The imaging target is the target within the subject that requires X-ray imaging based on the X-ray imaging protocol. For example, the imaging target could be the subject's palm, waist, abdomen, or spine, etc. The X-ray imaging protocol is the specific protocol (e.g., the Organ-Tissue Protocol (OGP)) used during the X-ray imaging of the imaging target. The X-ray imaging protocol can be determined based on user selections in a human-computer interface. For example, when the imaging target is the subject's abdomen, the second body shape feature could include: abdominal fat thickness, the superior ventral angle, the upper boundary of the abdomen, etc. When the imaging target is the subject's spine, the second body shape feature could include: spinal length, vertebral body length, intervertebral disc length, and the ratio of spinal length to height, etc.

[0074] The above exemplary descriptions illustrate typical examples of the first and second body features. Those skilled in the art will recognize that such descriptions are merely exemplary and are not intended to limit the scope of protection of the embodiments of the present invention.

[0075] In one implementation, extracting body shape features of a subject based on a 3D image includes: inputting the 3D image into a body shape feature determination model, wherein the body shape feature determination model is trained based on second training data, the second training data including historical 3D images and body shape features annotated based on the historical 3D images; and receiving body shape features determined based on the 3D image from the body shape feature determination model. Therefore, automatic extraction of body shape features from 3D images is achieved, improving feature extraction efficiency. The historical 3D images may include: 3D images of the subject during exposure operations that have been performed historically using X-ray imaging protocols; the body shape features annotated based on the historical 3D images may be: body shape features annotated after manually reviewing the historical 3D images, or body shape features automatically annotated based on machine algorithms, etc. In one implementation, it further includes: establishing a second artificial neural network model; and training the second artificial neural network model into a body shape feature determination model using the second training data. The training process includes: inputting second training data into a second artificial neural network model to output predicted body shape features corresponding to the second training data; determining the loss function value of the second artificial neural network model based on the difference between the labeled body shape features and the predicted body shape features in the second training data; configuring the model parameters of the second artificial neural network model so that the loss function value is lower than a second preset threshold; and defining the configured second artificial neural network model as the body shape feature determination model. Therefore, by introducing an artificial neural network into the body shape feature extraction process, the implementation difficulty is reduced.

[0076] In another implementation, a 3D image of the subject is used as modeling data, and a 3D model of the subject is obtained through simulation modeling. Then, the subject's body shape features are read from the 3D model. For example, simulation modeling methods include wireframe modeling, solid modeling, surface modeling, etc. For instance, in surface modeling, matching marker points or image contours are extracted from the 3D image. Based on the marker points, image contours, and volume-invariant constraints, the 3D deformation and motion parameters of the human body are estimated, and then a human body model is drawn using spheres and conic surfaces of revolution.

[0077] Step 103: Input the body shape features into the exposure parameter prediction model corresponding to the X-ray imaging protocol. The exposure parameter prediction model is trained based on the first training data, which includes historical body shape features and historical exposure parameter values ​​from historical exposure operations of the X-ray imaging protocol.

[0078] In one embodiment, the method further includes: establishing a first artificial neural network model; training the first artificial neural network model into an exposure parameter prediction model using first training data, wherein the training includes: inputting the first training data into the first artificial neural network model to output predicted exposure parameter values ​​corresponding to the first training data; determining a loss function value for the first artificial neural network model based on the difference between historical exposure parameter values ​​in the first training data and the predicted exposure parameter values ​​corresponding to the first training data; configuring model parameters of the first artificial neural network model to make the loss function value lower than a first preset threshold; and determining the configured first artificial neural network model as the exposure parameter prediction model. During the training process, the first artificial neural network model automatically learns the correlation between historical body shape features and historical exposure parameter values, and has the ability to predict exposure parameters using body shape features. The historical exposure operations of the X-ray imaging protocol refer to exposure operations that have been historically executed using the X-ray imaging protocol. The body shape features of the subject in the historical exposure operations (referred to as historical body shape features) and the exposure parameters in the historical exposure operations (referred to as historical exposure parameter values) are obtained. First training data is constructed based on the historical body shape features and historical exposure parameter values.

[0079] Specifically, the exposure parameters may include at least one of the following: X-ray tube voltage, X-ray tube current, exposure time, exposure dose, the product of X-ray tube current and exposure time, and exposure density.

[0080] Therefore, the exposure parameter prediction model trained using historical exposure parameter values ​​and historical body shape features from historical exposure operations incorporates the correlation between body shape features and exposure parameters. Consequently, the exposure parameter prediction model can improve the accuracy of exposure parameters.

[0081] Step 104: Receive the predicted exposure parameters based on body shape characteristics from the exposure parameter prediction model.

[0082] Therefore, the embodiments of the present invention automatically determine exposure parameters based on the three-dimensional image of the subject, improving exposure accuracy and reducing manual difficulty. Furthermore, the correlation between the subject's body shape characteristics and the exposure parameters is considered during the automatic determination process, further improving parameter accuracy.

[0083] In one implementation, the body shape features include a first body shape feature with a first weight and a second body shape feature with a second weight, wherein the second weight is greater than the first weight. Correspondingly, the historical body shape features in the first training data, the second training data, and the third training data also include a first historical body shape feature with a first weight and a second historical body shape feature with a second weight. Thus, the prediction result is obtained by utilizing both the overall body shape features of the test subject and the local body shape features of the imaging target. The prediction result comprehensively weighs the overall body shape of the test subject and the local body shape of the imaging target, ensuring accuracy. Moreover, the weight of the local body shape is greater than the weight of the overall body shape, improving the correlation between the prediction result and the imaging target.

[0084] In one embodiment, the method further includes: displaying the predicted exposure parameter value (e.g., displaying the predicted exposure parameter value via a user interface); determining whether an adjustment instruction for the predicted exposure parameter value has been received (e.g., received via the user interface); wherein when an adjustment instruction for the predicted exposure parameter value is received, the predicted exposure parameter value is adjusted based on the adjustment instruction, and a first exposure operation is performed on the imaging target based on the adjusted predicted exposure parameter value; when no adjustment instruction for the predicted exposure parameter value is received, a second exposure operation is performed on the imaging target based on the predicted exposure parameter value. Therefore, the predicted exposure parameter value can be adjusted based on the adjustment operation, ensuring the accuracy of the predicted value.

[0085] In one embodiment, the method further includes: storing third training data in the log, the third training data including exposure parameter values ​​for performing the exposure operation and the body shape features of the imaging target; extracting the third training data from the log; and performing an update training process on the exposure parameter prediction model using the third training data. The update training process includes: inputting the third training data into the exposure parameter prediction model so that the exposure parameter prediction model outputs exposure parameter prediction values ​​corresponding to the third training data; determining the loss function value of the exposure parameter prediction model based on the difference between the exposure parameter values ​​in the third training data and the exposure parameter prediction values ​​corresponding to the third training data; configuring the model parameters of the exposure parameter prediction model so that the loss function value is lower than a third preset threshold; and determining the configured exposure parameter prediction model as the updated exposure parameter prediction model. It can be seen that updating the exposure parameter prediction model using logs allows for the introduction of adjustment preferences into the exposure parameter prediction model, making it more in line with user habits.

[0086] Figure 2 An exemplary flowchart for determining the body shape characteristics of a subject according to an embodiment of the present invention.

[0087] First, the imaging process 20 is performed. In the imaging process 20, a 3D camera is used to capture the subject to obtain a 3D image of the subject. Then, the 3D image of the subject is used as modeling data to obtain a 3D model 21 of the subject through simulation modeling. Next, the body shape features 22 of the subject are extracted from the 3D model 21. The body shape features 22 include: (1) the overall body shape features 60 of the subject; (2) the body shape features 70 of the subject that correspond to the imaging target of the X-ray imaging protocol, which can be called local body shape features. The overall body shape features 60 can be one or more, such as the subject's upper and lower limb length ratio 221, chest circumference 222, and waist circumference 223. The local body shape features 70 can be one or more. For example, when the imaging target is the abdomen, the local body shape features 70 can include abdominal fat thickness 321, upper abdominal angle 322, and upper abdominal boundary position 323.

[0088] Figure 3 This is a schematic diagram of log storage according to an embodiment of the present invention. When the subject enters the preparation state 30, a three-dimensional image 31 is obtained by photographing the subject. Body shape features 32 are extracted using the three-dimensional image 31. Exposure parameters 34 are set on the control host's settings interface. When an exposure button press event 35 is detected, the exposure parameters 34 and body shape features 32 are associated and stored in the log 33. Subsequently, the exposure parameters 34 and body shape features 32 can be extracted from the log 33 as the first training data for the exposure parameter prediction model.

[0089] Figure 4This is a schematic diagram illustrating an exemplary process for determining exposure parameters in X-ray imaging according to an embodiment of the present invention.

[0090] An exposure parameter prediction model 42 is pre-trained. Specifically, this includes: using body shape features 40 stored in the logs and exposure parameters 41 associated with those features as first training data. Here, the association between body shape features 40 and exposure parameters 41 means that exposure has already been performed on historical test subjects with body shape features 40 using exposure parameters 41. The first training data is input into a first artificial neural network model, which outputs predicted exposure parameter values ​​corresponding to the first training data. Based on the difference between historical exposure parameter values ​​in the first training data and the predicted exposure parameter values ​​corresponding to the first training data, the loss function value of the first artificial neural network model is determined. The model parameters of the first artificial neural network model are configured so that the loss function value is lower than a first preset threshold. The configured first artificial neural network model is then determined as the exposure parameter prediction model 42.

[0091] After training the exposure parameter prediction model 42, the exposure parameters can be automatically determined using the exposure parameter prediction model 42. The specific process of determining the exposure parameters includes: using a camera component to photograph the subject to obtain a three-dimensional image 43 of the subject. The three-dimensional image 43 is input into the body shape feature determination model 44, which is trained based on second training data, including historical three-dimensional images and body shape features labeled based on the historical three-dimensional images. The body shape feature determination model 44 determines the body shape features of the subject based on the three-dimensional image of the subject. The body shape feature determination model 44 sends the body shape features of the subject to the exposure parameter prediction model 42. The exposure parameter prediction model 42 predicts the exposure parameter prediction value 45 based on the body shape features of the subject. The exposure parameter prediction value 45 is displayed on the user interface. Then, in the judgment process 46, it is determined whether a confirmation instruction for the exposure parameter prediction value 45 displayed on the user interface has been received. If yes (corresponding to the "Y" branch), exposure processing 48 is executed; if not (corresponding to the "N" branch), manual adjustment processing of the exposure parameters 47 is executed. In the manual exposure parameter adjustment process 48, the predicted exposure parameter value 45 is manually adjusted, and then the exposure process 48 is performed using the manually adjusted predicted exposure parameter value 45.

[0092] Figure 5 This is a structural diagram of an apparatus for determining exposure parameters in X-ray imaging according to an embodiment of the present invention. Figure 5As shown, the apparatus 500 for determining exposure parameters in X-ray imaging includes: an acquisition module 501 configured to acquire a three-dimensional image of the imaging target; an extraction module 502 configured to extract body shape features of the imaging target based on the three-dimensional image; an input module 503 configured to input the body shape features into an exposure parameter prediction model corresponding to an X-ray imaging protocol, wherein the exposure parameter prediction model is trained based on first training data, the first training data including historical body shape features and historical exposure parameter values ​​from historical exposure operations of the X-ray imaging protocol; and a receiving module 504 configured to receive the predicted exposure parameter values ​​determined based on the body shape features from the exposure parameter prediction model.

[0093] In an exemplary embodiment, the apparatus 500 further includes: a training module 506 configured to establish a first artificial neural network model; and to train the first artificial neural network model into an exposure parameter prediction model using first training data, wherein the training includes: inputting the first training data into the first artificial neural network model to output exposure parameter prediction values ​​corresponding to the first training data; determining a loss function value of the first artificial neural network model based on the difference between historical exposure parameter values ​​in the first training data and the exposure parameter prediction values ​​corresponding to the first training data; configuring model parameters of the first artificial neural network model to make the loss function value lower than a first preset threshold; and determining the configured first artificial neural network model as an exposure parameter prediction model.

[0094] In an exemplary embodiment, the extraction module 502 is configured to input a three-dimensional image into a body shape feature determination model, wherein the body shape feature determination model is trained based on second training data, the second training data including historical three-dimensional images and body shape features annotated based on the historical three-dimensional images; and to receive body shape features determined based on the three-dimensional image from the body shape feature determination model.

[0095] In an exemplary embodiment, the training module 506 is configured to establish a second artificial neural network model; and to train the second artificial neural network model into a body shape feature determination model using second training data. The training includes: inputting the second training data into the second artificial neural network model so that the second artificial neural network model outputs predicted body shape features corresponding to the second training data; determining the loss function value of the second artificial neural network model based on the difference between the labeled body shape features and the predicted body shape features in the second training data; configuring the model parameters of the second artificial neural network model so that the loss function value is lower than a second preset threshold; and determining the configured second artificial neural network model as a body shape feature determination model.

[0096] In an exemplary embodiment, an exposure module 507 is also included, configured to determine whether an adjustment instruction for the predicted exposure parameter value is received; wherein when an adjustment instruction for the predicted exposure parameter value is received, the predicted exposure parameter value is adjusted based on the adjustment instruction, and an exposure operation is performed on the imaging target based on the adjusted predicted exposure parameter value; when no adjustment instruction for the predicted exposure parameter value is received, an exposure operation is performed on the imaging target based on the predicted exposure parameter value.

[0097] In an exemplary embodiment, the system further includes: an update module 508 configured to store third training data in a log, the third training data including exposure parameter values ​​for performing the exposure operation and body shape features of the imaging target; extracting the third training data from the log; and performing an update training process on the exposure parameter prediction model using the third training data, the update training process including: inputting the third training data into the exposure parameter prediction model so that the exposure parameter prediction model outputs exposure parameter prediction values ​​corresponding to the third training data; determining the loss function value of the exposure parameter prediction model based on the difference between the exposure parameter values ​​in the third training data and the exposure parameter prediction values ​​corresponding to the third training data; configuring the model parameters of the exposure parameter prediction model so that the loss function value is lower than a third preset threshold; and determining the configured exposure parameter prediction model as the updated exposure parameter prediction model.

[0098] In an exemplary embodiment, the body shape feature includes a first body shape feature with a first weight and a second body shape feature with a second weight, wherein the first body shape feature characterizes the overall body shape of the subject and the second body shape feature characterizes the local body shape of the subject that corresponds to the imaging target of the X-ray imaging protocol; wherein the second weight is greater than the first weight.

[0099] Figure 6 This is a structural diagram of an apparatus for determining exposure parameters in X-ray imaging, having a memory-processor architecture according to an embodiment of the present invention.

[0100] like Figure 6As shown, the apparatus 600 for determining exposure parameters in X-ray imaging includes a processor 601, a memory 602, and a computer program stored in the memory 602 and executable on the processor 601. When executed by the processor 601, the computer program implements the method for determining exposure parameters in X-ray imaging as described above. Specifically, the memory 602 can be implemented as various storage media such as an electrically erasable programmable read-only memory (EEPROM), flash memory, or a programmable programmable read-only memory (PROM). The processor 601 can be implemented as including one or more central processing units (CPUs) or one or more field-programmable gate arrays (FPGAs), wherein the FPGA integrates one or more CPU cores. Specifically, the CPU or CPU core can be implemented as a CPU, an MCU, or a DSP, etc.

[0101] It should be noted that not all steps and modules in the above processes and structural diagrams are mandatory; some steps or modules can be omitted as needed. The execution order of the steps is not fixed and can be adjusted as required. The division of modules is merely for the convenience of description and functional division. In actual implementation, a module can be implemented by multiple modules, and the functions of multiple modules can also be implemented by the same module. These modules can be located in the same device or in different devices.

[0102] The hardware modules in each embodiment can be implemented mechanically or electronically. For example, a hardware module may include specially designed permanent circuitry or logic devices (such as dedicated processors, such as FPGAs or ASICs) to perform specific operations. A hardware module may also include programmable logic devices or circuitry (such as general-purpose processors or other programmable processors) temporarily configured by software to perform specific operations. The choice between mechanical implementation, dedicated permanent circuitry, or temporarily configured circuitry (such as software-configured circuitry) can be made based on cost and time considerations.

[0103] The present invention also provides a machine-readable storage medium storing instructions for causing a machine to execute the methods described herein. Specifically, a system or apparatus equipped with a storage medium storing software program code implementing the functions of any of the embodiments described above can be provided, and the computer (or CPU or MPU) of the system or apparatus can read and execute the program code stored in the storage medium. Furthermore, an operating system or similar device operating on a computer can perform some or all of the actual operations through instructions based on the program code. The program code read from the storage medium can also be written to a memory located in an expansion board inserted into a computer or to a memory located in an expansion unit connected to the computer, and then, based on the instructions in the program code, a CPU or similar device installed on the expansion board or expansion unit can execute some or all of the actual operations, thereby implementing the functions of any of the embodiments described above. Storage medium embodiments for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, the program code can be downloaded from a server computer or the cloud via a communication network.

[0104] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method (100) for determining exposure parameters in X-ray imaging, characterized in that, The method (100) includes: Acquire a 3D image of the subject (101); Based on the three-dimensional image, the body shape features of the subject are extracted (102). The body shape features are input into the exposure parameter prediction model corresponding to the X-ray imaging protocol, wherein the exposure parameter prediction model is trained based on the first training data, which includes historical body shape features and historical exposure parameter values ​​(103) of the historical exposure operations of the X-ray imaging protocol. Receive the exposure parameter prediction values ​​(104) determined based on the body shape characteristics from the exposure parameter prediction model. The body shape features include a first body shape feature with a first weight and a second body shape feature with a second weight, wherein the first body shape feature represents the overall body shape of the subject, and the second body shape feature represents the local body shape of the subject that corresponds to the imaging target of the X-ray imaging protocol; wherein the second weight is greater than the first weight.

2. The method (100) according to claim 1, characterized in that, Also includes: Establish the first artificial neural network model; The first artificial neural network model is trained into the exposure parameter prediction model using the first training data, wherein the training includes: The first training data is input into the first artificial neural network model so that the first artificial neural network model outputs the exposure parameter prediction value corresponding to the first training data; Based on the difference between the historical exposure parameter values ​​in the first training data and the predicted exposure parameter values ​​corresponding to the first training data, the loss function value of the first artificial neural network model is determined; the model parameters of the first artificial neural network model are configured so that the loss function value is lower than a first preset threshold; the configured first artificial neural network model is determined as the exposure parameter prediction model.

3. The method (100) according to claim 1, characterized in that, The step of extracting the body shape features of the subject based on the three-dimensional image (102) includes: The three-dimensional image is input into the body shape feature determination model, wherein the body shape feature determination model is trained based on second training data, the second training data including historical three-dimensional images and body shape features labeled based on the historical three-dimensional images; The body shape feature determination model receives body shape features determined based on the three-dimensional image.

4. The method (100) according to claim 3, characterized in that, Also includes: Establish a second artificial neural network model; The second artificial neural network model is trained into the body shape feature determination model using the second training data, the training including: The second training data is input into the second artificial neural network model so that the second artificial neural network model outputs predicted body shape features corresponding to the second training data; Based on the difference between the labeled body shape features in the second training data and the predicted body shape features, the loss function value of the second artificial neural network model is determined; the model parameters of the second artificial neural network model are configured so that the loss function value is lower than a second preset threshold; the configured second artificial neural network model is determined as the body shape feature determination model.

5. The method (100) according to claim 1, characterized in that, Also includes: Determine whether an adjustment instruction for the predicted value of the exposure parameter has been received; When an adjustment instruction for the predicted value of the exposure parameter is received, the predicted value of the exposure parameter is adjusted based on the adjustment instruction, and a first exposure operation is performed on the imaging target based on the adjusted predicted value of the exposure parameter. When no adjustment instruction for the predicted exposure parameter value is received, a second exposure operation is performed on the imaging target based on the predicted exposure parameter value.

6. The method (100) according to claim 5, characterized in that, Also includes: After performing the first exposure operation, third training data is stored in the log. The third training data includes the adjusted exposure parameter prediction values ​​after performing the first exposure operation and the body shape characteristics of the subject. Extract the third training data from the log; The exposure parameter prediction model is updated using the third training data. The update training process includes: The third training data is input into the exposure parameter prediction model, so that the exposure parameter prediction model outputs the exposure parameter prediction value corresponding to the third training data; based on the difference between the adjusted exposure parameter prediction value in the third training data and the exposure parameter prediction value corresponding to the third training data, the loss function value of the exposure parameter prediction model is determined; the model parameters of the exposure parameter prediction model are configured so that the loss function value is lower than a third preset threshold; the configured exposure parameter prediction model is determined as the updated exposure parameter prediction model.

7. An apparatus (500) for determining exposure parameters in X-ray imaging, characterized in that, The device (500) includes: The acquisition module (501) is configured to acquire a three-dimensional image of the subject; The extraction module (502) is configured to extract the body shape features of the subject based on the three-dimensional image; The input module (503) is configured to input the body shape features into an exposure parameter prediction model corresponding to an X-ray imaging protocol, wherein the exposure parameter prediction model is trained based on first training data, the first training data including historical body shape features and historical values ​​of exposure parameters in historical exposure operations of the X-ray imaging protocol. The receiving module (504) is configured to receive the exposure parameter prediction value determined based on the body shape characteristics from the exposure parameter prediction model; The body shape features include a first body shape feature with a first weight and a second body shape feature with a second weight, wherein the first body shape feature represents the overall body shape of the subject, and the second body shape feature represents the local body shape of the subject that corresponds to the imaging target of the X-ray imaging protocol; wherein the second weight is greater than the first weight.

8. The apparatus (500) according to claim 7, characterized in that, The extraction module (502) is configured to input the three-dimensional image into a body shape feature determination model, wherein the body shape feature determination model is trained based on second training data, the second training data including historical three-dimensional images and body shape features labeled based on the historical three-dimensional images; and to receive body shape features determined based on the three-dimensional image from the body shape feature determination model.

9. The apparatus (500) according to claim 7, characterized in that, Also includes: An exposure module (505) is configured to determine whether an adjustment instruction for the predicted value of the exposure parameter is received; wherein when an adjustment instruction for the predicted value of the exposure parameter is received, the predicted value of the exposure parameter is adjusted based on the adjustment instruction, and a first exposure operation is performed on the imaging target based on the adjusted predicted value of the exposure parameter. When no adjustment instruction for the predicted exposure parameter value is received, a second exposure operation is performed on the imaging target based on the predicted exposure parameter value.

10. The apparatus (500) according to claim 9, characterized in that, Also includes: The update module (506) is configured to store third training data in the log after the first exposure operation is performed. The third training data includes the adjusted exposure parameter prediction values ​​and the body shape characteristics of the subject after the first exposure operation is performed. Extract the third training data from the log; The exposure parameter prediction model is updated using the third training data. The update training process includes: inputting the third training data into the exposure parameter prediction model to output an exposure parameter prediction value corresponding to the third training data; determining a loss function value for the exposure parameter prediction model based on the difference between the adjusted exposure parameter prediction value in the third training data and the exposure parameter prediction value corresponding to the third training data; configuring the model parameters of the exposure parameter prediction model so that the loss function value is lower than a third preset threshold; and determining the configured exposure parameter prediction model as the updated exposure parameter prediction model.

11. An apparatus (600) for determining exposure parameters in X-ray imaging, characterized in that, Includes a processor (601) and a memory (602); The memory (602) stores an application program that can be executed by the processor (601) to cause the processor (601) to perform a method (100) for determining exposure parameters in X-ray imaging as claimed in any one of claims 1-6.

12. A computer-readable storage medium, characterized in that, It stores computer-readable instructions that, when executed by a processor, implement the method (100) for determining exposure parameters in X-ray imaging as described in any one of claims 1-6.

13. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the method (100) for determining exposure parameters in X-ray imaging as described in any one of claims 1-6.

Images